Using Be-7 measurements to explore the performance of the SEDD model to predict sediment yield at event scale

The SEdiment Delivery Distributed Model (SEDD) is used worldwide to predict sediment yield in small- and medium-sized catchments. Its ability to provide estimates at different timescales was also demonstrated in several cases. However, in order to provide reliable predictions of sediment yield, this model requires calibration of the sediment delivery component that depends on the estimation of the calibration parameter beta for which no direct approaches are available so far. In this contribution, Be-7 measurements were carried out in a small catchment (1.38 ha in size) representative of marginal areas in Southern Italy that have been subjected to afforestation. The Be-7 measurements were used to provide estimates of the catchment sediment delivery ratio (SDR) at event-scale. These estimates resulted very well correlated with the values of the calibration parameter beta calculated using the analytical approach based on the recursive solution of the sediment budget equation. Based on this correlation and following a distributed approach, the final predictions of sediment yield were compared with the independent measurements of sediment output available at the catchment outlet. The analysis showed satisfactory agreement between measured and predicted values of sediment yield and suggested that the SEDD model, if properly calibrated, can be considered a useful tool to predict erosion risk in marginal areas. In this respect, the use of the Be-7 technique, in view of its ability to provide independent estimates of soil redistribution rates, proved to be very useful in such calibration exercises. Further work is however required to test the performance of the model in different geomorphic contexts.