Spatiotemporal Analysis of the Hydrological Responses to Land-Use Land-Cover Changes in the Manafwa Catchment, Eastern Uganda

Abstract

This study evaluated the impact of land-use land-cover (LULC) change (LULCC) on the hydrologic processes of the Manafwa Basin in eastern Uganda using the Soil and Water Assessment Tool (SWAT) model. Model inputs included the 1995, 2008, and 2017 LULC maps; digital elevation model; and soil, climatological, and hydrologic data. The Calibration and Uncertainty Program Sequential Uncertainty Fitting algorithm was used to calibrate and validate the SWAT model. The model was validated for 2011 to 2013 using daily river discharge. Model statistical measures obtained during calibration and validation ranged from 0.79 to 0.94 for coefficient of determination, from 0.65 to 0.79 for Nash–Sutcliffe coefficient and between −12.0 and − 30.2 for percentage bias. Spatial coverage of commercial farmland and built-up areas increased by 405 percent and 337 percent, respectively, while wetland and tropical high forests decreased 82 percent and 41 percent, respectively, during the period from 1995 to 2017. There is increased annual surface runoff (67 percent), sediment yield (109 percent), and water yield (72 percent) and decreased potential evapotranspiration (32 percent) and actual evapotranspiration (31 percent). Therefore, implementation of integrated basin management strategies such as soil and water conservation and afforestation could reverse the negative impacts and ensure effective water resources management.