Comparative Assessment of Rockfall Susceptibility in the Potohar Plateau Using Frequency Ratio, Analytical Hierarchy Process, and Weights of Evidence Models

Rockfall is an extensively researched geomorphic process with notable socioeconomic consequences, especially in mountainous regions worldwide. Potohar Plateau, characterized by challenging terrain, seismic activity, and monsoon rains, faces various geo-hazards. However, limited research has been conducted on rockfall susceptibility in these areas. The current study aims to estimate and compare the effectiveness of three methods: frequency ratio (FR), analytical hierarchy process (AHP), and weights of evidence (WOE) in developing rockfall susceptibility models. Initially, a rockfall inventory was constructed using NASA’s broader datasets (Cooperative Open Online Landslide Repository (COOLR)), remote sensing (RS) images, and field observations. Fourteen causative factors comprising slope, elevation, aspect, Normalized Difference Vegetation Index (NDVI), land cover, curvature, rainfall, lithology, distance from roads, distance from faults, seismic activity, distance from streams, Topographic Wetness Index (TWI), Stream Power Index (SPI), were derived from a spatial database. Rockfall susceptibility maps were prepared using FR, AHP, and WOE in ArcGIS. The accuracy of each model was assessed by the area under the receiver operating characteristic (AUC-ROC) curve. The results indicated that the FR had the maximum accuracy with an AUC value of 0.762, followed by AHP with 0.733 and WOE with 0.639. The model with the highest performance highlights a high susceptibility to rockfall occurrences in the southern part of the study area, encompassing the Salt Range and M2-Motorway. The research findings can be valuable for policymakers and planners to mitigate the financial, human, and transportation costs associated with rockfall risk.