<p>Soil erosion is a critical environmental process driving land degradation, sedimentation, and non-point source pollution in river systems. In tropical regions undergoing rapid urbanisation, intensified rainfall, land-use change, and vegetation loss have amplified erosion-related hazards affecting both terrestrial and aquatic ecosystems. This study integrates the Revised Universal Soil Loss Equation (RUSLE) with Geographic Information Systems (GIS) and remote sensing to quantify and map soil erosion vulnerability in a representative tropical river basin in Malaysia. The model incorporates rainfall erosivity (R), soil erodibility (K), slope length and steepness (LS), cover management (C), and conservation practice (P) factors, derived from multi-source spatial datasets including rain gauge records, land use maps, soil surveys, and a digital elevation model (DEM). Spatial analysis indicates annual soil loss ranging from 0 to 300 t·ha⁻¹·yr⁻¹, with low to moderate erosion dominating most of the basin and localized hotspots in steep, sparsely vegetated, and rapidly urbanising zones. These high-risk areas contribute significantly to sediment loads and downstream water quality deterioration. The findings highlight the importance of integrating erosion vulnerability mapping into watershed-scale hazard management and pollution mitigation strategies. The methodological framework is adaptable to other tropical regions facing similar challenges of land degradation, flood risk, and sediment-driven pollution.</p> Graphical Abstract <p></p> <p>In this graphical abstract, the Klang River Basin is prominently displayed to establish the geographical context of the study and situate the environmental challenges being assessed. Surrounding the study area map, the graphical abstract highlights the primary datasets used in the analysis, including rainfall erosivity records from national hydrological stations, land-use and land-cover data derived from Sentinel-2 imagery, soil classifications from global soil databases, and a Digital Elevation Model (DEM) obtained from the Shuttle Radar Topography Mission. These inputs collectively form the empirical foundation of the study. The central section of the graphical abstract illustrates the five key RUSLE factors, which are R (rainfall erosivity), K (soil erodibility), LS (slope length–steepness), C (cover management), and P (support practices), each represented as an individual spatial layer. This arrangement visualizes the analytical workflow used to derive erosion susceptibility, demonstrating how multi-source geospatial data were integrated within a GIS environment to characterize the drivers of soil loss. The final panel presents the annual soil loss map, where erosion intensity is visualized spatially to highlight patterns and hotspots across the basin. This synthesis of model outputs reveals substantial heterogeneity, with certain steep and sparsely vegetated areas exhibiting significantly higher erosion risks. A key quantitative outcome, an estimated 476,134.1 tons of annual soil loss, is emphasized to underscore the scale of sediment yield and its potential contribution to downstream pollution.</p>