Asian Journal of Geographical Research

Flood susceptibility mapping is essential for disaster risk reduction in monsoon-dominated river basins. This study applies an integrated Geographic Information System (GIS) and Analytical Hierarchy Process (AHP) approach to assess flood susceptibility in the Shilabati River basin of West Bengal, India (area ~3,029 km²). Eight criteria - elevation, slope, distance from river, drainage density, flow accumulation, Topographic Wetness Index (TWI), rainfall intensity, and Land Use/Land Cover (LULC) - were selected based on field observations and expert judgment. Pairwise comparison matrices were constructed using Saaty’s 1-9 scale, and criterion weights were derived through AHP. The consistency ratio (CR = 0.0376) confirmed the reliability of the weight assignments, with elevation (32%) and slope (22%) emerging as the dominant controls. Weighted overlay analysis classified the basin into five susceptibility zones. This susceptibility map identifies the relative likelihood of inundation; it does not provide hydraulic parameters such as flood depth, velocity, or duration. The Most Susceptible zone (MoSFA) covers 0.16 % (4.98 km²), representing critical flood foci in the Ghatal confluence area. The Highly Susceptible zone (HFA) occupies 14.28% (432.50 km²), while the Moderately Susceptible zone dominates 73.21 % (2,217.38 km²). Less and Not Susceptible zones together account for 12.35% of the basin. Community-based validation using primary data from 120 local residents (including the researcher as an insider) confirmed that 92.7 % of reported flood points fell within the MoSFA and HFA zones, strongly supporting the model’s accuracy. The spatial distribution aligns with documented chronic flood-prone areas, demonstrating the utility of this approach. The results provide a scientific basis for targeted flood management, land-use zoning, and early warning system development. This AHP-GIS framework offers a cost-effective and replicable methodology for flood susceptibility assessment in data-scarce regions, particularly when combined with participatory validation.