Microsoft Word - ERSC090525-F - edited_corrected.doc 31-40 SQU Journal For Science, 15 (2010) © 2010 Sultan Qaboos University 31 Study of Groundwater Potentiality and Sea Water Intrusion along the Coastal Plain, Wadi Thuwal, KSA- A Case Study Based on DC Resistivity Mansour A. Al-Garni* and Hamdy I.E. Hassanein Geophysics Department, Faculty of Earth Sciences, King Abdulaziz University, Saudi Arabia, *Email: maalgarni@kau.edu.sa. دراسة احتمالية وجود المياه الجوفية وأثر اختراق مياه البحر إلى منطقة وادي ثول خالل السهل الساحلي، المملكة العربية دراسة بالمقاومة الكهربية –السعودية حسانين . القرني وحمدي ا. منصور ع ه صالحة وتلك التي قد تكون تأثرت باختراق أن غرض الدراسة هو تحديد المناطق التي يمكن أن توجد فيها ميا :خالصة على السهل الساحلي بوادي ثول الذي 2 كلم170فقد تم قياس المقاومة الكهربية على مساحة تقرب من . مياه البحر اليها ئيأن القياسات المقاومية التي اجريت على اساس النموج الطبقي الكهربا. يحده البحر األحمر غربا وتالل بركانية شرقا لذلك فان التحليل االحصائي . طبقات جيولوجيةتعبر عن تعطي قيما واسعة النطاق التي اليمكن ان تكون حقيقية المتعدد وقد قمنا . كان ضروريا للتغلب على هذه الظاهرة وللحصول على الصورة الحقيقية الطبقية الجيولوجية لتفسير البيانات التي تشمل على عدد من ) SLM(كورة الى اخرى احصائية للكهربية االرضية المذ اكهربيةبتحوير النماذج الطبقية فالنموذج المحور حدد لنا عمق . الطبقات المساوية للطبقات الجيولوجية تحت كل موقع تم فيه قراءة المقاومية الكهربية .حر في المنطقةطبقة الصخور االساسية، والطبقات الحاملة للماء والمستوى المائي واثر اختراق ماء الب ABSTRACT: The present study mainly aims to outline zones that have groundwater potentiality with good quality and those which are affected by sea water intrusion. The electrical resistivity data were acquired over an area of about 170 km2 of a coastal plain, Wadi Thuwal, which is bounded by the Red Sea in the west and the volcanic hills in the east. In such an area, resistivity measurements, using n-layering model, generally reveal a wide range of resistivity values which do not reflect the reality. Hence, the statistical analysis has to be involved to overcome this problem and to make the final interpretation reliable. In our case, the n-layer models were modified to other statistical geoelectric models (SLM), consisting of a number of layers equivalent to the stratigraphic layering beneath each VES site. The modified models were used to outline the depth to the bed rock, groundwater accumulation zones and water table as well as to define the effect of sea water intrusion through the study area. KEYWORDS: Wadi Thuwal; Coastal plain; DC resistivity; N-layering model; Statistical model. MANSOUR A. AL-GARNI and HAMDY I.E. HASSANEIN 32 1. Introduction he DC- resistivity in general has been successfully used to delineate the fresh water/saline water interface (El-Waheidi, 1992; Choudhury et al 2001) as well as water content (Kessels et al 1985). The studied area is about 170 km2 along the Red Sea coastal plain where it lies at the downstream of Wadi Khulase (Figure 1). It is mainly covered by Holocene Wadi deposits, Quaternary basalts (Harrat Thuwal) and Miocene sediments. It is shown that the structural pattern beneath the study area is affected by more than one phase of faulting. The surface of Miocene deposits is considered as a bed rock (M. Abdulwahid unpublished data). The lithology and configuration of the surface of Miocene deposits play the main significant role in building up the drainage pattern beneath the eolian and alluvian cover of the study area. This work represents an attempt to study the subsurface structural setting and to delineate the ground water distribution in Wadi Thuwal area, as well as the seawater intrusion effect through it. 2. Data acquisition A vertical electrical sounding (VES) survey was conducted at 88 selected sites using Schlumberger array to measure the surface and subsurface variation in resistivity. The well-known ELREC-T system (IRES) was used as an electrical resistivity meter, utilizing 1200 kW. 2.1 Vertical electrical sounding data analysis 2.2.1 N-layering technique A VES survey provides a series of apparent resistivity values as a function of depths. These values are obtained at each VES location, using different electrode spacings. The half current electrode spacing (AB/2) reached up to 140 m where the spacing between the potential electrodes (MN) varied from 2m to 20m according to the rule of thumb of Schlumberger array (AB≥5MN). These measurements were made along 8 E-W profiles, covering the study area. The resistivity curves fall into the following types: QH, KQ, HK, Q and H. Sampling of the continuous smoothed curve at the rate of 6 logarithmically equally spaced points per logarithmic cycle was done to obtain a digitized sounding curve. In general, sampling the apparent resistivity is done from right to left, starting from the largest current electrode spacing, where the effect of near surface inhomogeneity is disregarded (Al-Garni, 1996). We used Zohdy’s method (Zohdy, 1989) to invert the VES field resistivity measurements to a number of horizontal geoelectric layers. The study area has a large aerial extension with significant variable surface and subsurface structural and environmental depositional regimes built up over a long period of time, where these two factors have led to a wide range of resistivity values (0.9-2568.5 ohm.m). Therefore, statistical analysis was suggested to define and classify the obtained resistivity data as normal distribution groups with known statistical parameters related to certain lithological and/or structural environment units and constrained by reliable depth limits. 2.2.2 Statistical analysis The process of statistical analysis involves all the data as one population to check their normality. The standard deviation of the measured resistivity values all over the area is 227. The Kolmogorov-Smirnov test of normality (Clark and Evans, 1954; Montgomery and Runger, 1994; Cressie, N.A.C, 1991; and King, Ronald S. and Bryant Julstrom 1982) shows that this population of resistivity measurements do not have a normal distribution, where the Kolomgrov-Smirnov statistical index >> the critical value (Table 1). This indicates that the population of data assemblage could be classified into different populations; hence, the statistical analysis was used to implement the classification. T STUDY OF GROUND WATER POTENTIALITY AND SEA WATER INTRUSION 33 Table 1. Results of statistical analysis of the resistivity data of the study area. # of data Min Max mean S.D. K-S C.K-S at α=.05 761 0.86 2568.5 127.8 227 0.288 0.049 Legend Sand and Drainage lines Aeolian sand dip direction normal Inferred Alluvial Asphaltic roads Harrat flood Desert roads Miocene VES locations Study area a b Figure 1. Location (a) and Geology and the VES sites distribution of Wadi Thuwal area, KSA (Al-Garni et al 2009) (b). Accordingly, the calculated resistivity values, using n-layering modeling technique, are classified into six different normal distribution classes each of which corresponds to certain lithological ensembles of distinguishable resistivity character. Table 2 shows the statistical parameters and the results of normality distribution analysis. To define the depth limitations of each one of the classes, the data involved in each class was plotted individually and compiled as a depth resistivity graph (Figure 2), which shows that the resistivity values decrease with increasing depth. The upper 10 m, which is related to the surface and near surface lithological variations, has a wide resistivity range (1-2570 ohm.m). The resistivity of the section between 10 and 20 m ranges between 0.8 -200 ohm.m. It is underlain by the third section (20-30 m), with resistivity values, ranging between 1.8-40.0 ohm.m. The last section, which is deeper than 30 m, has a resistivity range between 2.5-20 ohm.m. The geoelectric n-layering model (NLM), which was obtained from the n-layer modeling technique (Zohdy, 1989), can be modified to another equivalent geoelectric statistical layering model (SLM). In this case, the number of geoelectric layers will be reduced and it will lead to more realistic lithological units. For example, at VES 1001, ten layers are interpreted using n-layer technique (NLM) whereas this number is reduced to six geoelectric layers using statistical analysis (SLM). MANSOUR A. AL-GARNI and HAMDY I.E. HASSANEIN 34 Table 2. Results of normality test and the 6 unit statistical parameters. Code number Statistical parameters 1 2 3 4 5 6 Number of values 194 151 134 168 104 10 range 0.9- 8.1-24.7 25.2-78.6 80.8-255.3 260.2-933.3 960.3-2568.5 Mean 4.1 14.2 45.7 149.6 478.9 1322.4 Standard deviation 1.9 4.8 16.1 52.3 186.3 482.4 Kolmogorov-Smirnov stat. index 0.096 0.105 0.107 0.102 0.129 0.253 Critical K-S stat, pha =.05 0.097 0.109 0.116 0.104 0.132 0.409 -5 4 -5 2 -5 0 -4 8 -4 6 -4 4 -4 2 -4 0 -3 8 -3 6 -3 4 -3 2 -3 0 -2 8 -2 6 -2 4 -2 2 -2 0 -1 8 -1 6 -1 4 -1 2 -1 0 -8 -6 -4 -2 0 depth in m 0.1 1 10 100 1000 10000 R es is tiv ity in O hm .m The relation between resistivity variations with depth 260-960 ohm.m 80-260 ohm.m 25-80 ohm.m 8-25 ohm.m 0-8 ohm.m >960 ohm.m Figure 2. Resistivity clusters versus depths. The developed classification, resistivity contours, and the geology are superimposed all together to implement the correlation (Figure 3). This shows certain distinctive zones of low resistivity values, taking an elongated extension, which is mostly attributed to the surface courses of the floods during the successive rainy seasons. The interpreted low resistivity values, using statistical geoelectric layers (SGL) (1, 2 and 3), are present on the surface of such zones. The interpreted higher resistivity values, using SGL (4, 5 and 6), occur mainly in areas that are covered by alluvial terraces. The interpreted depth beneath each model was reduced to sea level in order to reveal the relation between the resistivity distributions and the geomorphologic and environmental variations. The obtained resistivity values and their corresponding depths were used to construct eight W-E cross-sections along profiles, coded as 1000, STUDY OF GROUND WATER POTENTIALITY AND SEA WATER INTRUSION 35 2000, 3000, 4000, 5000, 6000, 7000 and 8000, respectively (Figure 4). The interpreted resistivity classes (Table 2) were considered as contour intervals. Figure 3. Distribution of SGL, surface resistivity contours, and geology of Wadi Thuwal area. 3. Interpretation of the resistivity data The lateral and vertical resistivity variations along each profile show that the basement rocks and the overlain sedimentary Miocene deposits were affected by different structural events, changing their structural setting at the surface of the bedrock. Hence, the depositional and hydrological regimes of the recent Quaternary deposits are controlled by these changes. (Figure 4) shows that the main water table level of the study area is the sea level, ranging between 10 and 20 m from the surface. The bedrock is characterized by low resistivity values (1 to 8 ohm.m), extending beneath the entire study area. This indicates that the lithological composition of the bedrock is mainly clay, which may be attributed to the salinity content near the shoreline (at the north western part of the study area, Figure 6). The bedrock is overlain by two geoelectric layers. The first, coded No.2, ranging between 8 and 25 ohm.m in resistivity, represents sediments of sand and gravel saturated with water regardless of the quality. The second, coded No.3, which ranges in resistivity between 25 and 80 ohm.m, represents sediments of sand and gravel partially saturated with water (Figure 4). These two units are considered to be the most significant units for conducting groundwater exploration, where the thickness of water saturation of these zones varies according to the level of precipitation. The resistivity behavior of unit No.3 is almost the same as that of No.2, yet the thickness of the water saturation zone is greater. These two zones can be observed along profiles 2000 (V2007 and 2014), 3000 (V3006, 3008, 3014 and 3015), 4000 (bewteenV4008 and 4011), 5000 (V5008 and 5009), 6000 (V6005 and 6006), 7000 (V7003) and 8000(V8002) (Figure 4). MANSOUR A. AL-GARNI and HAMDY I.E. HASSANEIN 36 The three geo-electric units No. 4, 5 and 6 are present frequently, occupying the upper part of the geo- electric sections, and they are characterized by high resistivity values. These units have no significant impact on the water exploration. However, their lithological characters may control the flow of the floods and water percolation through them. Figures (5a and b) show the topographic contour maps (depth to sea level) of the upper surface of both geo-electric units No.1 and 2, respectively, which reveal the effect of the structural events. There are five zones (SZ1, SZ2, SZ3, SZ4, and SZ5) of low resistivity values, which can be delineated based on the correlation between the resistivity and bedrock topography (Figure 5a). Three zones (SZ1, SZ2 and SZ3), which are located below sea level, are invaded by the sea water intrusion where high salinity of groundwater is expected. The other two zones (SZ4 and SZ5), which are located above sea level, prevalently contain an abundance of clay accumulations. Figure (5b) shows five significant catchment zones (Z1, Z2, Z3, Z4 and Z5) within the second geo-electric layer. The resistivity values of this layer and its relative topography (above the sea level) indicate the presence of water of good quality. Figure (5c) shows the thickness variations of this layer all over the studied area, where the average thickness of the groundwater bearing layer at these zones is about 10 m. Figure (6) shows the interpreted faults and outlines the sea water intrusion zones (SWI) and fresh water occurrences (FWO-1, FWO-2, FWO-3 and FWO-4). It shows also that the sea water intrusion and fresh water flow are controlled by the subsurface structures. The area at the east of the fault F4 is affected by sea water intrusion, whereas the most expected fresh water occurrences are located at the west of F7. 0 2000 4000 6000 8000 10000 12000 14000 16000 Easting in m from 511911E -1 00 0 0 10 00 E le va tio n (m ) *4 0 V80 01 V80 02 V80 03 V80 04 Geoelectric Section along VES profile 8000-to-sea level Sea level 0 2000 4000 6000 8000 10000 12000 14000 16000 Easting in m from 511911E -1 00 0 0 10 00 E le va ti o n (m ) *4 0 V70 01 V70 02 V70 03 V70 04 V70 05 V70 06V70 07 Geoelectric Section along VES profile 7000-to-sea level Sea level 0 2000 4000 6000 8000 10000 12000 14000 16000 Easting in m from 511911E -1 00 0 0 10 00 E le va tio n (m ) *4 0 V60 01 V60 02 V60 03 V60 04 V60 05 V60 06 V60 07 V60 08 Geoelectric Section along VES profile 6000-to-sea level Sea level Easting in m from 511911E 00 50 0 15 00 E le va tio n (m ) *4 0 V50 01 V50 02 V50 03 V50 04 V50 05 V50 06 V50 07 V50 08 V50 09 Geoelectric Section along VES profile 5000-to-sea level Sea level 400 600 800 960 1000 1400 2000 3000 L5 L6 Figures to be completed on the next page. STUDY OF GROUND WATER POTENTIALITY AND SEA WATER INTRUSION 37 0 2000 4000 6000 8000 10000 12000 14000 16000 Easting in m from 511911E -1 00 0 0 10 00 20 00 E le va tio n (m ) *4 0 V40 01 V4 002 V40 03 V40 04 V40 05 V40 06 V4 007V40 08 V40 09 V40 10 V4 011 V40 12 Sea level Geoelectric Section along VES profile 4000-to-sea level 0 2000 4000 6000 8000 10000 12000 14000 16000 Easting in m from 511911E -1 00 0 0 10 00 20 00 E le va tio n (m ) *4 0 V300 1 V3002 V300 3 V300 4 V300 5 V300 6 V300 7 V300 8 V300 9 V101 0 V3011 V0 12 V301 3 V3014 V 3015 V3 016 Sea level Geoelectric Section along VES profile 3000-to-sea level 0 2000 4000 6000 8000 10000 12000 14000 16000 Easting in m from 511911E -1 00 0 0 10 00 20 00 E le va tio n (m ) *4 0 V200 1 V200 2 V200 3 V200 4 V200 5 V200 6 V200 7 V200 8 V200 9 V201 0 V201 1 V201 2 V201 3 V201 4 V201 5 Sea level Geoelectric Section along VES profile2000-to-sea level 0 2000 4000 6000 8000 10000 12000 14000 16000 Easting in m from 511911E -1 00 0 0 10 00 20 00 E le va tio n (m ) *4 0 V100 1 V100 2 V100 3 V100 4 V100 5 V10 07 V100 8 V100 9 V101 0 V101 1 V101 2 V101 3 V101 4V101 5V101 6 Sea level Geoelectric Section along VES profile1000-to-sea level Figure 4. Geoelectric resistivity sections relative to sea level. 4. Conclusion and recommendation The present study shows that the water-table level generally ranges between depths of about 10 m to 20 m under the ground surface. The bedrock is characterized by low resistivity, ranging between 1 and 8 ohm.m. This range of resistivity values indicates that the lithological composition of the bedrock is mainly of clay. The interpreted structural regime helped to delineate the sea water intrusion as well as the fresh water catchment zones. The bedrock is overlain by two water bearing geo-electric layers with different degrees of water saturation. The first layer’s resistivity ranges between 8 and 25 ohm.m, representing sands and gravels saturated with water. The significant water accumulations are observed as thicker zones along the extension of this layer. The conducted study defined four main fresh water occurrence zones with average thickness up to 10 m (FWO1, FWO2, FWO3, FWO5) for abundant groundwater supply. More detailed study is recommended to be conducted at these zones before drilling. The statistical approach shows that it was an appropriate method for interpreting the resistivity data in this study. Hence, the methodology can be generalized and it is advised that it be applied in such coastal plain environment studies. MANSOUR A. AL-GARNI and HAMDY I.E. HASSANEIN 38 0 2000 4000 6000 8000 10000 12000 14000 16000 Eastern in m from 511911E UTM 0 20 00 40 00 60 00 80 00 10 00 0 N or th in g in m fr om 2 33 99 33 N U T M Depth relative to sea level 2 4 6 8 10 re si st iv ity in o hm .m 20 VES sites SZ4 SZ2 SZ3 SZ5 SZ4 SZ1 relatively high salinity zone SZ1 a 0 2000 4000 6000 8000 10000 12000 14000 Eastern in m from 511911E UTM 0 20 00 40 00 60 00 80 00 10 00 0 N or th in g in m fr om 2 33 99 33 N U T M layer 2 depth contoures from sea level in m 2 4 6 8 10 12 14 16 18 20 30 60 100 200 300 600 900 1000 2000 re si st iv ity in o hm .m Z2 Z1 Z3 Z1 Z5 Z4 20 Subsurface water channels ground water occurencesZ2 Legend b 0 2000 4000 6000 8000 10000 12000 14000 Eastern in m from 511911E UTM 0 20 00 40 00 60 00 80 00 10 00 0 N or th in g in m fr om 2 33 99 33 N U T M Calculated thickness of layer no.2 2 4 6 8 10 12 14 16 18 20 30 60 100 200 300 600 900 1000 2000 re si st iv ity in o hm .m Z2 Z3 Z1 Z5Z4 20 Z2 ground water occurences Legend c Figure 5. The depth contour map of the upper surface of the interpreted SGL no.1 (a), the depth contour map of the interpreted SGL no. 2 (b), and thickness contour map of layer no. 2, showing zones of water of relatively good quality (c). STUDY OF GROUND WATER POTENTIALITY AND SEA WATER INTRUSION 39 0 2000 4000 6000 8000 10000 12000 14000 16000 Easting from ref.point 511951m E 0 20 00 40 00 60 00 80 00 10 00 0 12 00 0 14 00 0 N or th in g fro m re f. po in t 2 44 99 33 m E 1 1 1 1 1 1 111 1 1 1 1 1 1 1 11 1 1 1 1 1 1 1 1 1 1 11 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 11111111 1 11 1 1 1 11111111 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 22 2 2 F1 F2 F3 F4 F5F6 F7 Sea water intrusion(saline water) S W I SCA FWO-4 FWO-2 salty clay acumulations Fresh water occurrences 1 2 layer 1 distribution under the study area resistivity < 8 ohm.m layer 2 distribution under the study area resistivity range=(8-25) SWI FWO SCA FWO-3 FWO-1 FWO-5 Inferred fault FaultF1F1 Figure 6. Interpretation results of Wadi Thuwal area. 5. Acknowledgment The authors are greatly thankful to the King Abdulaziz City for Science and Technology (KACST) for supporting this study AT-26-82. The authors would like to express their sincere thanks to the editors and the anonymous reviewers, for the thorough review that highly improved the original manuscript. 6. References ABDULWAHID, M. (unpublished data), Geology and structural framework of Harrat Thuwal, North of Jeddah, Saudi Arabia. AL-GARNI, M.A. 1996. Direct current resistivity investigation of groundwater in the lower Mesilla Valley, New Mexico and Texas: MS thesis, Colorado School of Mines, 120 p. AL-GARNI, M.A., EL-BEHIRY, M.G, GOBASHY, M.M, HASSANEIN, H.I., and EL-KALIOUBY, H.M. 2009. Geophysical studies to assess groundwater potentiality and quality at Wadi Thuwal area, North of Jedda, KSA. King Abdulaziz City for Science and Technology, project No. At-26-82, p. 394. CHOUDHURY, K., SAHA, D.K., and CHAKRABORTY, P. 2001. Geophysical study for saline water intrusion in a coastal alluvial terrain. J. Apllied Geophy., 46: 189-200. CLARK, P. AND EVANS F. 1954. Distance to nearest neighbor as a measure of spatial relationships in populations. Ecology, 35: 445-453. CRESSIE, N. A. C. 1991. Statistics for Spatial Data. John Wiley & Sons Inc., New York. MANSOUR A. AL-GARNI and HAMDY I.E. HASSANEIN 40 EL-WAHEIDI, M.M., MERLANTI, F., and PAVAN, M. 1992. Geoelectrical resistivity survey of the central part of Azraq plain (Jordan) for indentifying saltwater/freshwater interface. J. Applied Geophy., 29: 125- 133. KESSELS, W., FLENTGEH, I., KOLDITZ, H. 1985. DC geoelectric sounding to determine water content in the salt mine asse (FRG). Geophysical Prospecting, 33: 456-446. KING, RONALD S. and BRYANT JULSTROM. 1982. Applied Statistics Using the Computer. Alfred Publishing Company, Sherman Oaks, California. Montgomery, D.C. and Runger G.C. 1994. Applied Statistics and Probability for Engineers. John Wiley & Sons Inc., New York. ZOHDY. A.A.R. 1989. A new method for the automatic interpretation of Schlumberger and Wenner sounding curves. Geophysics, 54: 245-253. Received 25 May 2009 Accepted 21 April 2010