Evaluation of the sensitivity of the RUSLE erosion model to rainfall erosivity: a case study of the Ksob watershed in central Algeria

Water erosion is a serious challenge in Algeria, because it affects ecosystems, contributes to soil degradation, and leads to the silting of dams; moreover, this process is complex and needs costly field equipment and trip reconnaissance. The main objective of this paper is to evaluate the sensitivity of the Revised Universal Soil Loss Equation (RUSLE) model to rainfall erosivity in the Ksob watershed in Algeria, using six empirical formulas of rainfall erosivity. The RUSLE model with the Geographic Information System produces highly variable specific soil loss ranging from 11.35 to 22.85 t ha(-1) year(-1) throughout the watershed, depending on the erosivity R factor. To validate the results, the soil loss is compared with the silting volume data of the Ksob dam, which is located at the outflow of the basin. The best result is obtained using the Diodato R factor formula with a relative error of 21% and a specific soil loss of 11.35 t ha(-1) year(-1). Moreover, over 81% of the watershed area is exposed to low erosion (<20 t ha(-1) year(-1)), and less than 5% are affected by high erosion (>50 t ha(-1) year(-1)).