Predicting Soil Erosion by Water using a Spatial Distributed Data GIS Model

Spatially distributed models coupled with Geographic Information System allow to identify areas sensitive or susceptible of soil loss.The objective of this study was mapping areas more prone to soil erosion in a forest catchment using the Revised Universal Soil Loss Equation. Patterns of spatial variability of RUSLE parameters were evaluated through geostatistical methods. The study area (about 1.4 km2) was a forest watershed (South Italy). To calculate the soil erodibility factor (K), 110 soil samples were collected from superficial pedological horizon and were analyzed for soil texture and organic matter content. Three raingauges, located inside the watershed, were used to compute the rainfall-runoff erosivity factor (R). Coupling a geostatistical approach with a spatial distributed model in a GIS, produced a good representation of the erosion process. The annual soil loss was in accordance with field measurements and other studies carried out using different approaches