A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering, School of Environment, Resources and Development

Thesis (M. Sc.) - Asian Institute of Technology, 1998

Nepal is a mountainous country, ranging from an altitude of 100 meter in terai (lower prut) to the highest elevation of Mt. Everest. Nepal is blessed with an enormous capacity of hydropower and tapping of the country's potential of the hydropower becomes an utmost concern for the country's development. Soil erosion, which is a serious problem in hilly areas, becomes especially significant in a country like Nepal, because it is a countly that is geologically young, endowed with 'slopey' fragile mountains, and built with rugged surface topography. The erosion process in the rugged terrain watershed transports a lot of sediment to the reservoir and is the most important factor for reducing the net storage capacity of hydropower. Therefore, monitoring and prediction of siltation of the reservoir become important for the formulation of the strategies for the sustainable management of hydropower generation. This research deals with the ways and means of predicting the soil erosion in Andhi Khola watershed in Nepal, using Landsat TM and SPOT satellite images, aerial photographs and GIS. The Landsat and SPOT images of 1991 have been used in this study for predicting 1991 soil loss and aerial photographs were used as data for 1997. The other data sets used in this study are scale 1 :25,000, 1 :50,000 topography maps, respectively, and a land system map of scale 1 :50,000. The study has been carried out using USLE model, which relates the phenomena of soil erosion to topography, soil and vegetation. High-resolution satellite data have been found to be very effective in monitoring land cover in hilly area, but require special consideration in classification under shading effect. Similarly, The use of GIS, by overlaying various maps related to factors affecting soil erosion, has been found to be very effective in analyzing the soil loss. Finally, with the parameters RKLSCP for soil erosion, the productive life span of the reservoir was estimated.