Sub-watershed Prioritization for Soil Erosion Management using Entropy- MAIRCA: A morphometric and LULC – based approach for the Noyyal River Basin, India

Watershed prioritization for soil erosion assessment requires an objective framework that integrates drainage characteristics with land-use changes. While morphometric analysis and Land Use/Land Cover (LULC) classification have been independently applied in prior studies, their combined use with Multi-Criteria Decision Making (MCDM) remains underexplored at river watershed scales. This study addresses that gap by proposing a hybrid Entropy-MAIRCA (Multi-Attribute Ideal Real Comparative Analysis) framework to classify sub-watersheds into high, medium and low vulnerability zones within the Noyyal River Basin. The Entropy Weighting Method (EWM) was adopted to objectively compute parametric weights, minimizing subjective bias in criterion importance. The basin was segmented into 21 sub-watersheds using 30 m Shuttle Radar Topography Mission (SRTM) Digital Elevation Model (DEM). Eighteen morphometric parameters were reduced to nine statistically independent variables through multicollinearity analysis. LULC data extracted from Sentinel-2A imagery were validated with an overall accuracy of 94% (Kappa = 0.925) and integrated with morphometry indicators to provide a comprehensive assessment. Model stability was verified through One-at-a-time (OAT) sensitivity analysis and the percent change technique. Consistency checks against the established TOPSIS method yielded strong base rank agreement exceeding 0.9, further validating the framework's reliability. Findings reveal that 43% of the basin falls within high-priority sub-watersheds requiring immediate conservation intervention, while 32% falls under medium priority, requiring targeted land management. Limitations include the 30 m DEM spatial resolution, single-year static LULC data and the lack of sediment loss records. The proposed framework serves as a robust, replicable decision support tool for long term soil erosion management. • This study ranks watersheds by integrating morphometry and land-cover data. • Statistical screening ensures that nine parameters provide independent results. • Sentinel-2 spatial data at ten-meter resolution improves mapping accuracy. • Sensitivity tests confirm the ranking remains stable across different scenarios. • High Spearman correlation with existing methods validates the model’s robustness.