APPLYING THE USLE FAMILY OF MODELS AT THE SPARACIA (SOUTH ITALY) EXPERIMENTAL SITE

Soil erosion is a key process to understand the land degradation, and modelling of soil erosion will help to understand the process and to foresee its impacts. The applicability of the Universal Soil Loss Equation (USLE) at event scale is affected by the fact that USLE rainfall erosivity factor does not take into account runoff explicitly. USLE-M and USLE-MM, including the effect of runoff in the event rainfall– runoff erosivity factor, are characterized by a better capacity to predict event soil loss. The specific objectives of this paper were (i) to determine the suitable parameterization of USLE, USLE-M and USLE-MM by using the dataseries of Sparacia experimental site and (ii) to evaluate their performances at both event and annual scale. The measurements allowed to establish the relationships for calculating the factors of USLE, USLE-M and USLE-MM usable at the Sparacia experimental area. At first, for slope-length values greater than 33 m, the calibration of USLE model at event scale pointed out that sediment delivery processes, that is processes involving deposition of the transported eroded soil particles, occur. The analysis showed that USLE and USLE-M tend to overestimate low event soil losses, while for USLE-MM, this tendency is less pronounced. However, the USLE-MM performed better than USLE and USLE-M and was able to reproduce better than other two models the highest soil loss values that are the most interesting from a practical point of view. The results obtained at annual scale were generally consistent with those obtained at event scale.