Predicting the effect of land use/land cover change on catchment soil erosion risk using swat model

Abstract

The Manafwa River sub-catchment in eastern Uganda has undergone significant changes in land use and land cover (LULC) over the past two decades, primarily driven by agricultural expansion. Such transformations have profound effects on hydrology, sediment dynamics, and flood risk. This study applied the Soil and Water Assessment Tool (SWAT) to evaluate the impacts of LULC changes on streamflow, soil erosion, and catchment hydrology from 2004 to 2024. LULC trends were first quantified, followed by SWAT calibration and validation using daily flow data, and scenario analyses to assess hydrological responses under varying land use scenarios. Results indicate that cropland expanded by 5%, while wetlands and forested areas declined by 5.5% and 2%, respectively. Flood hazard mapping revealed that built-up areas and commercial farmlands are highly vulnerable, whereas natural ecosystems exhibited low to moderate vulnerability. Hydrological simulations demonstrated strong model performance (NS= 0.86 for calibration and 0.77 for validation), with LULC changes increasing streamflow (from 0.49 to 0.54), surface runoff (from 0.26 to 0.33), and reducing evapotranspiration (from 0.49 to 0.44). Scenario analyses revealed that low-flow conditions are more sensitive to land-use changes than high-flow events. The risk of soil erosion increased markedly in areas converted from grassland, forest, and wetland to subsistence farmland, particularly on steep slopes. The findings underscore the critical need for integrated land management strategies, including agroforestry, contour banding, and soil conservation measures, to mitigate runoff, enhance infiltration, and reduce erosion. This study demonstrates the utility of SWAT for predicting the hydrological and ecological consequences of LULC change and provides actionable insights for sustainable watershed management in tropical catchments.