<p>The Rarh region of West Bengal, India, faces a critical challenge to its agricultural sustainability and environmental well-being due to soil erosion. This study aims to develop a scalable model for estimating soil erosion rates and identifying vulnerable areas within the region. By employing the Revised Universal Soil Loss Equation in conjunction with GIS and remote sensing techniques, the research integrates factors such as rainfall erosivity, soil erodibility, topography, land cover, and conservation practices to create a comprehensive soil erosion risk map. Findings reveal that annual soil loss in the study area ranges from 0.53 to 955.39 t ha⁻¹&#xa0;yr⁻¹, with an average of 13.014 t ha⁻¹&#xa0;yr⁻¹, exceeding the tolerable soil loss limit of 11.2 t ha⁻¹&#xa0;yr⁻¹. The erosion map classifies the region into zones of varying soil loss severity from very low to very severe. Approximately 2.67% of the area falls under severe to very severe categories, 10.63% is classified as high to very high risk, while 48.92% and 17.60% are in moderate and low-risk zones, respectively. The western and central parts of the Rarh region, characterized by steep slopes, intensive agriculture, and high rainfall intensity, are identified as areas with high soil loss potential. The model's performance was validated using an Area under the Curve (AUC) analysis, achieving a score of 0.832, indicating good predictive capability. This study's results offer valuable insights for policymakers and land managers, enabling the implementation of targeted soil conservation measures and sustainable land use practices in the Rarh region of West Bengal.</p>