Fuzzy class membership approach to soil erosion modelling

To optimize the use of the available land and water resources of a watershed for sustainable agricultural production, soil erosion assessment and conservation are essential. The Universal Soil Loss Equation (USLE) is a widely accepted model (Wishmeier and Smith, 1965. Predicting Rainfall Erosion Losses from Cropland (Agricultural Handbook No. 282). USDA. Washington, D.C.) for assessing soil erosion and to account for the most important factors. Conventional approaches to classi®cation are designed to assign a given area element (pixel) to a single erosion class. However, the soil and other physical parameters might vary spatially within a pixel and it may not correspond entirely to a single erosion class. To determine the loss of information on the susceptibility to erosion a fuzzy class membership approach was used to assign partial grades to the erosion classes. In the present studies, a spatially distributed approach was used to consider the in ̄uence of the spatial variation in the soil and other physical parameters in soil erosion assessment in the study area, Kalyanakere subwatershed Karnataka, India, using the USLE model. The emphasis of the studies was laid on the use of a fuzzy class membership approach in soil erosion classi®cation, and to develop a criteria table specifying the erosion parameter values related to erosion susceptibility classes from available literature to apply fuzzy class membership approach for the classi®cation. Salient features of the approach and the results of the study are presented in this paper. # 2000 Elsevier Science Ltd. All rights reserved.