International Journal of Environment and Climate Change

Landslides are a major geomorphic hazard in the hilly terrain of Northeast India, intensified by monsoon rainfall, active tectonics and unscientific road-cutting. This study assesses landslide susceptibility in Pherzawl District, Manipur, using an integrated geospatial framework combining Frequency Ratio (FR) and Weight-of-Evidence (WOE) methods. Eleven causative factors elevation, slope, aspect, curvature, lithology, geomorphology, soil, drainage density, land use/land cover, rainfall and distance to roads were analysed using high-resolution satellite data and GIS techniques. A landslide inventory of 60 historical events was prepared and divided into training and validation datasets. Model performance was evaluated using Receiver Operating Characteristic (ROC) curve analysis, yielding an Area Under Curve (AUC) value of 0.84, indicating high predictive accuracy. The susceptibility map classifies the district into three zones: low susceptibility covering 24.29% (555 km²), moderate susceptibility 49.77% (1,137 km²) and high susceptibility 25.94% (592 km²). High-risk zones are concentrated along steep slopes and road-cut hill sections in the eastern and southern parts of the district, while low-susceptibility areas are mainly located in valley floors and regions with dense forest cover. The results provide reliable hazard information for policymakers, planners and disaster management authorities. The integrated FR–WOE approach improves statistical reliability in data-limited mountainous regions and represents the first hybrid landslide susceptibility model for Pherzawl District. By delineating high-risk clusters, the study supports tri-zonal disaster mitigation, climate-resilient transportation planning, slope management and the prioritization of bio-engineering measures in vulnerable corridors.