Abstract
Abstract Flooding is one of the most severe environmental hazards affecting riverine urban centers, particularly in developing countries characterized by rapid urbanization, inadequate drainage systems, and climate variability. Makurdi metropolis, located along the floodplain of the River Benue in North-Central Nigeria, experiences recurrent flood events that threaten lives, infrastructure, agricultural land, and socio-economic activities. This study assessed flood susceptibility in Makurdi metropolis using an integrated Geographic Information System (GIS) and Analytical Hierarchy Process (AHP) approach. Ten flood-conditioning factors, including elevation, slope, rainfall, drainage density, distance to streams, Topographic Wetness Index (TWI), Normalized Difference Vegetation Index (NDVI), geomorphology, curvature, and land use/land cover (LULC), were analyzed within the ArcGIS Pro 3.4.0 environment using datasets derived from ASTER DEM, Landsat 8 imagery, Sentinel-2 land cover data, rainfall records, and field observations. The AHP analysis identified elevation and slope as the most influential conditioning factors, with normalized weights of 19.43% and 18.21%, respectively, followed by distance to streams (14.22%), drainage density (13.43%), precipitation (8.73%), LULC (8.25%), TWI (6.98%), NDVI (4.89%), geomorphology (3.34%), and curvature (2.52%).The resulting flood susceptibility map classified Makurdi metropolis into four susceptibility classes comprising low (44.07 km²; 5.34%), moderate (546.02 km²; 66.13%), high (229.62 km²; 27.81%), and very high susceptibility (5.96 km²; 0.72%). Moderate susceptibility occupied the largest proportion of the study area, while very high susceptibility was concentrated within the River Benue floodplain and adjacent low-lying communities. Highly susceptible zones were concentrated around low-lying floodplain communities and areas adjacent to the River Benue. Validation using an independent binary validation dataset comprising historical flood-occurrence points and non-flood reference points yielded an Area Under the Receiver Operating Characteristic Curve (ROC–AUC) value of 0.94, suggesting good predictive performance of the GIS–AHP model. Although this result demonstrates a strong discriminatory capability of the GIS–AHP model due to the limited data points considered, Hence, validation using larger independent datasets and complementary uncertainty analyses would provide additional confidence in the model’s robustness.