This research is an attempt to develop and test a methodology for evaluating the land use characteristics and assessing the range of soil loss values for a typical watershed on mountainous area in the northern Thailand, on the basis of Universal Soil Loss Equation, with the application of Remote Sensing and Geographic Information Systems. Digital image processing is employed for the spatial and temporal analysis of land use, extensive use of digital elevation model is made for analyzing the watershed landscape, and both the raster and vector based GIS is utilized for the spatial and temporal analysis of land use and soil erosion. The result shows that the rate of soil erosion in this watershed is very high, for exceeding acceptable limits in most areas. Further more, the trend in the twenty year period from year 1978 to 1999 shows that the watershed condition is further degrading with increasing rate of erosion, a phenomenon which is primarily explained by the degradation of forest. The study concludes that the high rate of erosion is mainly because of the cultivation in the steep slope. An analysis of land use planning with considerations for soil loss tolerance showed that about one third of the total cultivated land need to be converted into forest of tree crops in order to have sustainable land management. High-resolution satellite data have been found to be very effective in monitoring land use/land cover of a mountain watershed. It is concluded that special consideration in classification under shading effects (use of band ratios, NDVI, DEM) seems very useful for monitoring the land use land cover in such areas. Similarly the use of GIS has been found to be very effective in analyzing soil erosion spatially by overlaying various maps related to factors affecting soil erosion. Both raster and vector overlay was canied out and it is found that raster overlay is more powerful and effective than vector overlay It is also concluded that use of ARC VIEW spatial analysts for the estimation of soil erosion seem to be very useful using of high resolution grid cell data along with other factor in order to estimate soil loss. Specially estimating the topographic factor in the ARC VIEW GIS environment with in the capability of hydrological modeling proved to be very useful. On the applicability of the USLE, the research concludes that through more reliable results can be obtained with the revised topographic factors, still the soil loss values under steep slopes are not reliable enough. This calls for more research on developing appropriate approach for correctly assessing soil erosion rates under complex mountainous area such as those in Upper Nam Wa Watershed in Thailand. However, the methods is very effective in finding the range of soil erosion rates at different areas, and thus evaluating the susceptibility of the watershed to soil erosion, and finally in developing the general watershed management plan.

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

Thesis (M.Sc.) - Asian Institute of Technology, 2001