80 Qamar Zaman and Shahid Nawaz Khan/ GEOSI Vol 5 No 1 (2020) 80-91 The Application of Geographic Information System (GIS) and Remote Sensing in Quantifying Snow Cover and Precipitation in Kabul Basin Qamar Zaman* and Shahid Nawaz Khan Institute of Geographical Information Systems, School of Civil and Environmental Engineering, National University of Sciences and Technology, Islamabad, 44000, Pakistan *Corresponding Author : qamar.civil@suit.edu.pk Received 16 December 2019/ Revised 25 January 2020/ Accepted 31 January 2020/ Published 10 April 2020 Abstract Water Resources availability is very important to social and economic well-being of the people and has huge impacts on the socio-economic scenarios of a country. Precipitation and snow cover area assessment is some of the major inputs in hydrologic modelling and also for assessing and managing water resources in a basin. The change in the water availability in a basin has huge socio-economic impacts because of the water usage for food production, industries, and many others. The main aim of this study was to measure the snow cover area and precipitation from 2001 to 2015 in the Kabul basin. Moderate Resolution Image Spectroradiometer (MODIS) and Tropical Rainfall measuring Mission (TRMM) data were used to study snow cover area and precipitation respectively during 2001-2015. 8-day snow cover product for 15 years (January) was used to analyse the snow cover while monthly data of TRMM (3B43) were used to analyse the rainfall from 2001-2015. Different image processing techniques were applied on the data retrieved using GIS and Remote Sensing softwares. Initially, SCA was seen increasing, but during the last 3-4 years, it kept decreasing gradually. Rainfall was initially recorded as low, while later on, it was recorded high and reached the highest during 2010. Keywords: MODIS; Snow Cover; TRMM; Precipitation; Kabul Basin; Remote Sensing 1. Introduction Snow cover area is an important component of a regional as well as global climate system and also a critical storage system of fresh water on the surface (Larson et al., 2009). Equally, precipitation is also one of the important input parameters for land surface hydrological modelling of a basin and its quantification is necessary for assessing available water resources (Zhang & Srinivasan, 2010). Both snow cover area estimation and precipitation quantification, thus, are important to assess the water resources. Traditionally, snow cover is measured by monitoring the snow physically in a network which gives accurate information about the snow cover. However, the method is not feasible for long range measurement because of its spatial scale, i.e. the sparse network GEOSFERA INDONESIA p-ISSN 2598-9723, e-ISSN 2614-8528 Vol.5 No. 1 (2020), 80-91, April, 2020 https://jurnal.unej.ac.id/index.php/GEOSI DOI : 10.19184/geosi.v5i1.14896 Accredited by the Ministry of Research , Technology , and Higher Education of the Republic of Indonesia, No. 30/E/KPT/2019. https://jurnal.unej.ac.id/index.php/GEOSI https://drive.google.com/file/d/1rSnVAs6cuHOwHl5BJ87CL2L6K5DQz7s6/view 81 Qamar Zaman and Shahid Nawaz Khan/ GEOSI Vol 5 No 1 (2020) 80-91 having very few stations as compared to the area which is unable to deliver the amount of information needed for the accurate snow cover estimation (Atif et al., 2015). Similarly, rainfall is measured using rain gauges which give information about variations in trends o f precipitation monthly. Nevertheless, because the rain gauges cannot be deployed in remote places and some areas are difficult to reach, they account for 25-30% of the surface of the earth (New et al., 2001). Therefore, a mechanism is needed to monitor the snow cover and rainfall at a large spatial and temporal scale remotely (Ershadi et al., 2005). The modern satellite system in space has the capabilities to monitor different natural and human induced dynamics at different spatial, spectral and temporal scale (Daly et al., 2012). These satellite systems operate round the clock and provide the captured data of almost all of the world for, mostly, free or at some nominal price. The data retrieved from these systems can be analysed for making policy decision including water supply and demand. The Moderate Resolution imaging spectroradiometer (MODIS) which was launched in 1999 has the capabilities to study the earth’s geophysical phenomena, which include mapping the areal extent of snow using SNOWMAP algorithm (Hall et al., 1995). The SNOWMAP algorithm was developed by using thematic mapper data to automatically map the area extent of the snow using different MODIS bands on a daily basis. Different snow products of MODIS are available at national Snow and Ice Data center from September, 13, 2000 to date which include daily and 8 days snow cover (Hall et al., 2001). Tropical Rainfall Measuring Mission (TRMM) is a joint mission between NASA of United states of America and National space development agency of Japan, which measures the precipitation and energy exchange of tropical and subtropical regions (Kummerow et al., 1998). TRMM Microwave Imager (TMI) and Visible and Infrared Radiometer system (VIRS) are primary rainfall measurement instrument on board on TRMM. Different products of TRMM are available online, including daily 3B42 and monthly 3B43 products. Previous related studies argue that the relationship between snow, global warming, and land use can influence the changes in landforms (Beato Bergua et al., 2019). Snowfall can be identified by using a two-dimensional model, so that disaster risk reduction can be optimized (Nakamura et al., 2019). Snowmelt is affected by slopes and causes high infiltration (Miao et al., 2019). MODIS-based vegetation maps can be used to calculate the capacity and intensity of snow in a basin (Pyankov et al., 2018). The ASTER GDAM and Landsat 8 OLI images can be used to analyse the snowfall parameters (Kumar et al., 2018). Nevertheless, the previous studies did not consider the aspects of rainfall as water 82 Qamar Zaman and Shahid Nawaz Khan/ GEOSI Vol 5 No 1 (2020) 80-91 resources in the snow area, Therefore, it is necessary to study the use of GIS and remote sensing to improve the quality of water resources. The differences in the rainfall among the regions based on latitude are very important to know in relation to climate change. Hence, the main aim of this study was to measure the snow cover area and precipitation from 2001 to 2015 in the Kabul basin. 2. Methods The Kabul river basin has a catchment area of approximately 94,000 sq. Kilometers. The river drains two countries, i.e. Pakistan and Afghanistan, and is located in the northwestern side of Pakistan and north eastern side of Afghanistan (Aziz, 2014). The river Kabul flows from east to west, ultimately joining the Indus River near Attock, in Pakistan Khyber Pakhtunkhwa province from where it finally merges into Arabian Sea (Lashkaripour et al., 2008). Many barrages have been constructed on various locations on the river to support the system of extensive agricultural irrigation. The economy of most of the people depends on agriculture. Although the basin comprises of 12% area of Afghanistan, it has 35% population of Afghanistan (Vick, 2014). Major portion of the population live in the area and, mostly, the population is concentrated along the main courses of the river and adjacent areas where water is available for agriculture the production of agriculture because the main source of income in that area is agricultural production. Most of the agriculture is depends on the river and, thus, is located along the adjacent areas where river water is accessible for irrigation purposes. The greater amount of water supplied in the summers results in intensive cropping in the area while some of the areas is also rain fed. In this study, MODIS data were used for calculating the snow cover area. MODIS 8- days snow cover product with a spatial resolution of 500 meters, i.e. MOD10A2V6, is available at the National Snow and Ice Data Center NSIDC with a spatial resolution of 500 meters. The 8-daily product is generated from the daily snow cover product MOD10A1V6, which in turn is derived from Normalised difference snow index NDSI (equation 1). Snow cover has a high reflectance rate in visible bands and low reflectance rate in shortwave infrared SWIR. The Normalised difference snow index reveals those differencing high values show snow covered areas. Cells having NDSI>=0.4 are considered as the snow covered cells. Different Remote Sensing and GIS softwares (Table 1) were used for the processing of imagery and estimation of the snow cover area. NDSI= Band 4- Band 6/Band 4+ Band 6 (1) 83 Qamar Zaman and Shahid Nawaz Khan/ GEOSI Vol 5 No 1 (2020) 80-91 Table 1. Tools and data used for SCA assessment Data/Tools Details MODIS Snow cover data MOD10A2V6 (500 meters) TRMM data 3B43 Monthly data mm/hr. ArcMap 10.1.4 Collecting, Analyzing, Processing, and manipulating satellite imagery Modis Reprojection Tool Spectrally subsetting and reprojecting MODIS data 2.1. Reprojection MODIS data were processed and generated as a gridded output in the sinusoidal SIN projection in the HDF format. The MODIS reprojection tool is used for a number of pre- processes on MODIS data, including the spatial sub setting, spectral sub setting, format conversion, and geographic transformation to a different coordinate system. MRT was used to spectrally subset the products and retain only those bands which were needed for further processing in the study. MRT was available online for processing MODIS datasets, in which the data were converted from sinusoidal to projected coordinate system to enable the measurement and then were converted to TIFF format so that the other softwares can conveniently read it. 2.2. Image processing The aim of image processing using geospatial softwares is to extract the meaningful information from the images. Many open sources and commercial geospatial image processing softwares are available to use to process the different types of satellite imagery and extract the meaningful information from it. Some of the available softwares are Erdas Imagine, ENVI, IDRISI, and ArcMap. Although ArcMap is a GIS software having more capabilities for vector data processing, still it is used for different kinds of image processing .Images preprocessed through MRT were processed in ArcMap, where snow cover areas and non-snow cover areas were separated and the final snow cover areas from those images were extracted. 84 Qamar Zaman and Shahid Nawaz Khan/ GEOSI Vol 5 No 1 (2020) 80-91 2.3. Precipitation data acquisition and processing Tropical Rainfall Measuring Mission TRMM monthly product 3B43 version 7 was used for quantifying liquid precipitation in the study area. These data combined the daily data from TRMM and other satellites, and estimated the best possible estimate from the data at temporal resolution of one month. The data acquired from NASA was converted, reprojected, and extracted to study area using different GIS softwares. The data were rescaled to 500 meters spatial resolution. Many image processing and statistical techniques were applied on the data. TRMM 3B43 monthly data iswere available in unit of mm/hr,. which was then converted to mm/day, and subsequently to mm/month. Annual accumulated data as well as annual mean were generated using those monthly data. 2.4. Snow cover area (SCA) and precipitation quantification Snow cover and non-snow cover areas were separated in the MODIS product. Moreover, the snow cover area was calculated for each image, which was further averaged over the month by using all the data acquired and processed. Similarly, rainfall was quantified for the study area by using monthly and annually accumulated and mean data, using different techniques. 2.5. Snow cover area (SCA) and precipitation Maps The snow cover areas were estimated using 8-day MODIS snow cover product and the maps were also generated at the same interval. Four maps for each month were generated, i.e. the 1st to 8th day, 9th-17th day, 18th -25th day, and last from 25th to 3rd day of next month, which was the 3rd of February. The precipitation data was averaged manually and annual accumulated maps were produced by ArcMap. 3. Results and Discussion The results show that during the initial years i.e. 2001 onwards, there was an increase in the snow cover area which could be seen up to 2006 and 2007. There were also minor increase and decrease in the snow cover area in the successive years up to 2011. From 2012 onwards, a gradual decrease could be seen in the snow cover area which was 65,285 sq. kilometers in 2012, 63,244 sq. kilometers in 2013, 56716 sq. km in 2014, and 50,935 sq. km in 2015. Researchers pointed out that the decrease in snow cover area in the next 10-15 years in Pakistan will be considerable (Zahid & Rasul, 2011). It might not be considerably noted in the initial years, but it was observed during the last 4 years, from 2012 to 2015. From 2008 up to 2011, an irregular pattern was observed in the snow cover change, which was neither 85 Qamar Zaman and Shahid Nawaz Khan/ GEOSI Vol 5 No 1 (2020) 80-91 decreasing gradually nor increasing gradually but increased somewhere and decreased at other times. As shown in Table 1 and Figure 1, the snow over area gradually increased from 50,856 sq. km in 2001 to 61,406 sq. km in 2006. The snow cover changed irregularly during 2007 up to 2011, which sometimes increased or decreased. This might also be attributed to the accuracy of the data used for the study. The snow cover area was recorded the highest in 2012, which was 65,285 sq. km, from where it went on decreasing gradually towards 2015. Table 2. Accumulated snow cover area estimated from MODIS snow cover product Year Snow cover area (km2) Year Snow cover area (km 2) 2001 50856 2009 57812 2002 50721 2010 51286 2003 54307 2011 45517 2004 55629 2012 65285 2005 57201 2013 63244 2006 61406 2014 56716 2007 61221 2015 50925 2008 49599 Table 2 shows that there are fluctuations in the area covered by snow. This was due to changes in the rainfall and significant temperature differences. In 2012 and 2013, the area covered was very large due to the influence of changes in the duration and pattern of the rainy season. Table 3. Mean Annual Rainfall of Kabul Basin Estimated from TRMM Year Mean Annual rainfall (mm) Year Mean Annual rainfall (mm) 2001 260.00 2009 599.34 2002 388.73 2010 676.43 2003 535.31 2011 584.91 2004 472.43 2012 556.61 2005 562.78 2013 648.75 2006 521.37 2014 654.67 2007 518.39 2015 642.01 2008 566.19 86 Qamar Zaman and Shahid Nawaz Khan/ GEOSI Vol 5 No 1 (2020) 80-91 The mean annual rainfall varied from 260 mm to 676 mm as shown in Table 3. In the initial years, just like the snow cover, less rainfall was recorded and was restively larger in the following years. The year 2010 received the largest amount of rainfall in all 15 years which caused Pakistan to suffer devastating floods in areas of Khyber Pakhtunkhwa, where large amount of crops were destroyed and suffered from around 1100 causalities (Sayama et al., 2018). the average rainfall in Table 3 does not directly affect the snow cover,. This is because the intensity of the rain also greatly influences the existence of run off. In areas with high rainfall intensity, rain will erode soil material quickly with higher exogenous power. 87 Qamar Zaman and Shahid Nawaz Khan/ GEOSI Vol 5 No 1 (2020) 80-91 Figure 1. The expansion of the area covered by snow Figure 1 shows the expansion of the area covered by snow over time. It shows that there are human interventions on the environment that cause climate change and snow expansion. Areas that have steep slopes will be a source of melt snow that flows with high intensity to other areas; hence, there is a need for conservation efforts to overcome this. Snow Non snow 0 km 200 km 88 Qamar Zaman and Shahid Nawaz Khan/ GEOSI Vol 5 No 1 (2020) 80-91 When the rainy season is too short, snowmelt can cause changes in landforms with different impacts on the environment. The result of the previous research that can be compared to the results of this study is the need for water that comes from snowmelt when summer is very high, as it shows that snow has an impact on people's lives during the summer (Amstrong et al., 2019). Microclimate greatly influences snow characteristics in an area, so the is a need for differing planning to cope with too high rainfall (Plach et al., 2019). Meltsnow is the most influential source of irrigation in the area before the rainy season (Biemans., 2019). Meltsnow can increase run off so that it will significantly cause changes in the landform (Il Jeong & Sushama, 2018). Higher soil erosion occurs when there is snow when compared to the rainy season, so there needs to be a different conservation effort for the two seasons (Wu et al., 2018). Compared to the results of the previous studies, the advantage of this study lies on the temporal comparison in the low latitude region. With different rainfall, it is necessary to have different conservation in each watershed in this region. Areas with higher sun exposure will affect snow characteristics, especially the ones which exposed to erosion and run off. Watershed conditions that have a varied microclimate will have an impact on the diversity of vegetation and human activities. 4. Conclusion Remote Sensing satellites have the capabilities to sense different phenomenon on the surface of earth from space. Their data availability, reliability, and cost have defeated many conventional techniques to acquire data for analysing different processes on the surface of earth. It is also Ssimilar to the case of Snow Cover Area Aassessment and Rainfall Quantification, The conventional techniques may be accurate and useful, yet they are quite expensive and, sometimes, unreliable because of human induced factors or some other effects. Although satellite images are free of charge and are available on the daily basis, still, the choice of selecting the specific source of data needs much analysis and study on the analyst’s side. Thise study found no considerable change in the snow cover area in the initial years, but later on in the last 3-4 years, it found a gradual decrease in the snow cover which might be due to the changing snow cover accumulation periods. Related to rainfall, variations were found in which less rainfall were recorded in the initial years and, later on in 2010, the highest rainfall was recorded in which Pakistan experienced devastating floods. 89 Qamar Zaman and Shahid Nawaz Khan/ GEOSI Vol 5 No 1 (2020) 80-91 Conflict of Interest The authors declare that there is no conflict of interest with any financial, personal, or other relationships with other people or organizations related to the material discussed in the article. Acknowledgements We would like to express our sincere gratitude to Dr. Javed Iqbal, Professor and the Head of Institute of Geographical Information System Department, SCEE, NUST for his continuous support and guidance throughout the study. We would also like to thanks Mr. Junaid Aziz Khan, the lecturer, of Institute of Geographical Information Systems, NUST for his support and guidance throughout the period of this study. References Armstrong, R. L., Rittger, K., Brodzik, M. J., Racoviteanu, A., Barrett, A. P., Khalsa, S. -. S., . . . Armstrong, B. (2019). Runoff from glacier ice and seasonal snow in high asia: Separating melt water sources in river flow. Regional Environmental Change, 19(5), 1249-1261. doi:10.1007/s10113-018-1429-0 Atif, I., Mahboob, M. A., & Iqbal, J. (2015). Snow cover area change assessment in 2003 and 2013 using MODIS data of the Upper Indus Basin, Pakistan. Journal of Himalayan Earth Sciences, 48(2), 117-128. Aziz, A. (2014). Rainfall-Runoff Modeling of the Trans-Boundary Kabul River Basin Using Integrated Flood Analysis System (IFAS). Pakistan Journal of Meteorology, 10(20). Beato Bergua, S., Poblete Piedrabuena, M. Á., & Marino Alfonso, J. L. (2019). Snow avalanches, land use changes, and atmospheric warming in landscape dynamics of the atlantic mid-mountains (cantabrian range, NW spain). Applied Geography, 107, 38- 50. doi:10.1016/j.apgeog.2019.04.007 Biemans, H., Siderius, C., Lutz, A. F., Nepal, S., Ahmad, B., Hassan, T., . . . Immerzeel, W. W. (2019). Importance of snow and glacier meltwater for agriculture on the indo- gangetic plain. Nature Sustainability, 2(7), 594-601. doi:10.1038/s41893-019-0305-3 Daly, S. F., Vuyovich, C. M., Deeb, E. J., Newman, S. D., Baldwin, T. B., & Gagnon, J. J. (2012). Assessment of the snow conditions in the major watersheds of Afghanistan using multispectral and passive microwave remote sensing. Hydrological Processes, 26(17), 2631-2642. Ershadi, A., Khiabani, H., & Lorup, J. K. (2005). Applications of remote sensing, GIS and river basin modelling in integrated water resource management of Kabul River Basin. In ICID 21st European Regional Conference. Hall, D. K., Riggs, G. A., & Salomonson, V. V. (1995). Development of methods for mapping global snow cover using moderate resolution imaging spectroradiometer data. Remote sensing of Environment, 54(2), 127-140. 90 Qamar Zaman and Shahid Nawaz Khan/ GEOSI Vol 5 No 1 (2020) 80-91 Hall, D. K., Riggs, G. A., & Salomonson, V. V. (2001). Mapping Global Snow Cover using Moderate Resolution Imaging Spectroradiometer (MODIS) Data. Glaciological Data, 33. Il Jeong, D., & Sushama, L. (2018). Rain-on-snow events over north america based on two canadian regional climate models. Climate Dynamics, 50(1-2), 303-316. doi:10.1007/s00382-017-3609-x Kumar, S., Srivastava, P. K., & Snehmani. (2018). Geospatial modelling and mapping of snow avalanche susceptibility. Journal of the Indian Society of Remote Sensing, 46(1), 109-119. doi:10.1007/s12524-017-0672-z Kummerow, C., Barnes, W., Kozu, T., Shiue, J., & Simpson, J. (1998). The tropical rainfall measuring mission (TRMM) sensor package. Journal of atmospheric and oceanic technology, 15(3), 809-817. Larson, K. M., Gutmann, E. D., Zavorotny, V. U., Braun, J. J., Williams, M. W., & Nievinski, F.G. (2009). Can we measure snow depth with GPS receivers?. Geophysical Research Letters, 36(17). Lashkaripour, G. R., & Hussaini, S. A. (2008). Water resource management in Kabul river basin, eastern Afghanistan. The Environmentalist, 28(3), 253-260. Miao, F., Wu, Y., Li, L., Liao, K., & Zhang, L. (2019). Risk assessment of snowmelt -induced landslides based on GIS and an effective snowmelt model. Natural Hazards, 97(3), 1151-1173. doi:10.1007/s11069-019-03693-2 Nakamura, K. (2019). Implementation and demonstration of a system for the forecasting of surface avalanche potential caused by snowfall from a cyclone. Journal of Disaster Research, 14(9), 1201-1226. doi:10.20965/jdr.2019.p1201 New, M., Todd, M., Hulme, M., & Jones, P. (2001). Precipitation measurements and trends in the twentieth century. International Journal of Climatology, 21(15), 1889-1922. Plach, J., Pluer, W., Macrae, M., Kompanizare, M., McKague, K., Carlow, R., & Brunke, R. (2019). Agricultural edge-of-field phosphorus losses in ontario, canada: Importance of the nongrowing season in cold regions. Journal of Environmental Quality, 48(4), 813- 821. doi:10.2134/jeq2018.11.0418 Pyankov, S. V., Shikhov, A. N., Kalinin, N. A., & Sviyazov, E. M. (2018). A GIS-based modeling of snow accumulation and melt processes in the votkinsk reservoir basin. Journal of Geographical Sciences, 28(2), 221-237. doi:10.1007/s11442-018- 1469-x Sayama, T., Ozawa, G., Kawakami, T., Nabesaka, S., & Fukami, K. (2012). Rainfall–runoff– inundation analysis of the 2010 Pakistan flood in the Kabul River basin. Hydrological Sciences Journal, 57(2), 298-312. Vick, M. J. (2014). Steps towards an Afghanistan–Pakistan water-sharing agreement. International Journal of Water Resources Development, 30(2), 224-229. Wu, Y., Ouyang, W., Hao, Z., Yang, B., & Wang, L. (2018). Snowmelt water drives higher soil erosion than rainfall water in a mid-high latitude upland watershed. Journal of Hydrology, 556, 438-448. doi:10.1016/j.jhydrol.2017.11.037 91 Qamar Zaman and Shahid Nawaz Khan/ GEOSI Vol 5 No 1 (2020) 80-91 Zahid, M., & Rasul, G. (2011). Frequency of extreme temperature and precipitation events in Pakistan 1965–2009. Sci. Int.(Lahore), 23(4), 313-319. Zhang, X., & Srinivasan, R. (2010). GIS-based spatial precipitation estimation using next generation radar and raingauge data. Environmental Modelling & Software, 25(12), 1781-1788.