Impacts of sustainable management on the spatial distributions of erosion susceptibility and probable sediment yield in a mixed-forested watershed

Forest management practices play multifaceted roles in enhancing the geophysical properties that affect raindrop erosion in the watershed, and consequently, sediment deposition in the reservoir. The current work attempts to integrate empirical and physically-based modeling approaches to quantify the impacts of forest conservation on erosion risk and potential sediment accumulation in the mixed-forested Ogouchi Dam watershed in Japan. The reliability of the empirical model for estimating the total erodibility coefficient (TEr), as a function of various forest properties, was evaluated by applying the mathematical expression to multiple forest conditions and comparing the values to field-measured soil erosion rates. The spatial distribution of the empirically derived values showed that about 25.8% of the Government-managed forests and 45.1% of the private forests have higher risks of raindrop splash erosion compared to natural forests. The TEr value in each small Government-divided forest land (less than 5 ha) was then corresponded to the MUSLE management practice factor (MUSLE P) input in each hydrologic response unit (HRU) in the Soil and Water Assessment Tool (SWAT) model to create a sediment yield distribution map and to predict the amounts of sediment accumulation for different management scenarios. The spatial distribution of sediment yield for the base condition showed that 20.9% of the Government-managed forests and 61.6% of the private forests have higher probable amounts of sediment yield relative to the value simulated in the natural forest. A maximum cumulative sediment reduction of about 14.4% is likely attainable upon the complete control of the Government in the entire planted forest area. Overall, this study effectively utilized the field survey datasets to develop a robust empirical model for quantifying erosion risk and was able to couple the results to a GIS-based model that predicts the amounts of sediment yield under varying environmental conditions.