Estimation of soil loss and sediment yield by using the modified RUSLE model in the Indus River basin, including the quantification of error and uncertainty in remote-sensing images

Context Indus River is the cradle of Pakistani lifeline, and its lower reaches are prone to soil loss owing to bank erosion. Aims The aim was to investigate the sediment yield in the Lower Indus River Basin (LIRB), while addressing challenges related to error or uncertainty in remote-sensing data. Methods We employed a modified revised universal soil loss equation (RUSLE) model, integrating high-resolution digital elevation model (DEM) and calibrated Climate Hazards Group InfraRed Precipitation with station data (CHIRPS). Additional data layers, including land use, soil and cropping data, were also utilised. Key results The extent of actual soil erosion ranges from minimum to maximum erosion; 38.9% area lies in the range >50 Mg ha‒1 year‒1, whereas 23.2% area lies in the range of 0–10 Mg ha‒1 year‒1, and 18.1% area lies in the range of 10–20 Mg ha‒1 year‒1. Conclusions The study identifies critical erosion areas and tackles uncertainties in remote-sensing data. The spatial analysis showed that higher distribution sediment erosion along the channel flow direction from the northern part of LIRB to the Arabian Sea. Implications The findings have provided critical information for policymakers and water managers to implement effective measures to reduce erosion, maintain soil integrity and promote the sustainability of the Indus River system.