Application of the wepp model to hillside landsin The Tuxtlas, Veracruz, México

In attempts to better manage soil and water resources, soil erosion models have received increasing attention. Until recently, the Universal Soil Loss Equation (USLE) has been used as a soil erosion model on the field scale. Since then, more process-based soil erosion models have been developed such as the Water Erosion Prediction Project (WEPP) model. This new technology represents the state-of-the-art in modeling the effects of changes in surface characteristics on soil erosion. In this study, the capability of the WEPP hillslope interface in version 2002.07 for simulating runoff and soil loss, applied to selected runoff plots, was tested in the Tuxtlas, Veracruz, Mexico by comparing runoff and soil loss from live tree hedgerow terraces (TMV-TA 1.0 m) as an erosion control method with a conventional tillage (CT) practice. The breakpoint climate data generator (BPCDG) was used to generate the climate file. When compared to measured data, the CT WEPP estimated runoff and soil loss were low, whereas the TMV-TA 1.0 m WEPP estimated soil loss was high. For soil loss, the model over-predicts small events and under-predicts the larger ones. The Nash-Sutcliffe model efficiencies for predicting runoff and soil loss using an optimization method were 0.84 and 0.86 for the TMV-TA 1.0 m and 0.88 and 0.84 for the CT. The sensitivity analysis and calibration process show that the model is very sensitive in the effective hydraulic conductivity value and erodibility parameters. This study showed that in order to apply the WEPP model, revision of data collection system is necessary for studying the hydrologic and erosion soil processes with reliable estimates of soil erodibility parameters, effective hydraulic conductivity, and crop and managements parameters, in order to provide realistic estimates of runoff and soil loss.