Integrated RS–GIS–AHP approach for mapping groundwater potential zones in the Barind Tract of Bangladesh
Abstract
The availability of groundwater resources varies spatially due to hydrogeological, geomorphological, and environmental factors, representing a significant threat to agricultural sustainability in the drought-prone Barind Tract of Bangladesh. This study aimed to identify groundwater potential zones (GWPZ) in Naogaon district using an integrated approach combining remote sensing (RS), geographic information system (GIS), and the analytic hierarchy process (AHP). Eleven thematic layers including geology, lineament density, topographic wetness index (TWI), drainage density, rainfall, land use and land cover (LULC), soil type, slope, curvature, distance from rivers (DFR), and elevation were weighted and integrated using weighted overlay analysis. Multicollinearity assessment indicated no significant multicollinearity among the parameters, and sensitivity analysis identified lineament density, geology, and drainage density as the most influential factors. The resulting GWPZ map classified the district into very low (9.07%), low (23.36%), moderate (27.81%), high (26.10%), and very high (13.66%) potential zones, with high and very high zones concentrated in the central and eastern parts and low potential zones in the northwestern and southwestern regions. Model validation using the ROC curve yielded an AUC of 0.821, indicating very good predictive performance. The GWPZ map can support groundwater management by identifying suitable locations for new wells, prioritizing artificial recharge sites, and guiding sustainable irrigation planning in the Barind Tract.