Abstract:The Water Erosion Prodiction Project(WEPP) watershed scale model was developed by the USDA for soil erosion assessment and conservation planning.Serveral sensitivity analyses and evaluation studies have been conducted for the WEPP hillslope version, but few efforts have been made to evaluate the accuracy of WEPP watershed model predictions by comparing predicted results to measured data.The purpose of this study was to give a brief introduction of WEPP watershed model and to evaluate the WEPP watershed model applicability and prediction accuracy for smal l watersheds under different climates,topographies,soils,and manag ements.Data from 15 watersheds of 6 locations in the United States were compared to runoff and sediment yield estimates using WEPP95.The result shows that the WEPP predicted reasonable average annual and event runoff and sediment yield with default parameter estimation procedures for most of the sites and individual watersheds.Model efficiencies for storm by storm runoff and sediment yield were quite reasonable for general conditions.The distributions of measured and predicted events were similar.This study also pointed out the potential areas for model improvement,including better and more specific silage routines,inaddition of a weed component to the plant growth and management model,and better linkage of the sediment from contoured hillslopes to the watershed channel system.