Flood susceptibility mapping of the Subarnarekha River Basin using multicriteria flood conditioning factors in GIS-AHP integrated framework
Abstract
Floods are one of the most damaging natural disasters in the world, causing significant socioeconomic disruptions, especially in countries like India that are controlled by the monsoon. This study used an integrated method that included remote sensing (RS), Geographic Information Systems (GIS), and the analytical hierarchy process (AHP) to identify regions in the Subarnarekha River Basin (SRB) that are susceptible to flooding. Seventeen flood-conditioning factors including elevation, slope, drainage density, distance from river, rainfall, land-use land cover, normalized difference vegetation index, soil texture, geomorphology, and topographic indices were used for comprehensive analysis. The AHP-derived weights achieved a consistency ratio of 0.0089, indicating highly consistent expert judgments. The map was validated using the area under the curve (AUC) method against an independent Sentinel-1 synthetic aperture radar (SAR)-derived flood inventory. The AUC value of 0.770 was obtained, which indicates a moderate-to-good level of prediction performance. The most significant conditioning elements were found to be elevation (17%) and slope (15%), followed by drainage density (11%), distance from river (11%), and rainfall (9%). A significant area of the basin is classified as moderate to high susceptibility, with high-risk zones primarily located in low-lying downstream areas, according to spatial analysis. Unlike previous GIS–AHP flood susceptibility studies in India that typically employed fewer conditioning factors or relied mostly on expert-based validation, this study integrates 17 flood-conditioning factors at a uniform 30 m spatial resolution with independent Sentinel-1 SAR-derived flood inventory validation. In the SRB and other monsoon-dominated river basins, the suggested GIS-RS-AHP framework improves the dependability of basin-scale flood susceptibility mapping and offers a strong, transferable decision-support tool for flood-risk management, land-use planning, and disaster preparedness.