The study aims at delineating groundwater potential zones using geospatial technology and analytical hierarchy process (AHP) techniques in mining impacted hard rock terrain of Ramgarh and part of Hazaribagh districts, Jharkhand, India. Relevant thematic layers were prepared and assigned weight based on Saaty’s 9-point scale and normalized by eigenvector technique of AHP to identify groundwater prospect in the study area. The weighted linear combination method was applied to prepare the groundwater potential index in geographic information system. Final groundwater prospects were classified as excellent, very good, good, moderate, poor and very poor groundwater potential zones. Study thus revealed that the excellent, very good and good groundwater potential zones, respectively, cover 148.3, 373.66 and 438.86 km² of the study area, whereas the poor groundwater potential zone covers 180.05 km². Validation was done through a receiver operating characteristic curve, which indicated that AHP had good prediction accuracy (AUC = 75.45%).

Groundwater vulnerability and risk assessment is an essential step for preventing and controlling contamination in any area. DRASTIC model with modification incorporating land use (LU) widely used for assessing groundwater vulnerability worldwide. The present study used this modified model termed as DRASTIC-LU considering land use as a significant parameter in addition to other conventional DRASTIC parameters. Geographic information system (GIS) employed to prepare and integrate different parameter maps led to demarcate the risk zones in a catchment of Damodar River, Jharkhand, India. Final DRASTIC-LU map was categorized as low (19.50%), moderate (34.02%), high (29.90%) and very high (16.58%) risk zones of groundwater contamination. Validation of models done through Spearman’s rank correlation coefficient (ρ) and receiver operating characteristic (ROC) curve using nitrate (NO₃) concentration in groundwater, which indicated DRASTIC-LU (ρ = 0.893; AUC =71.65%) had a better agreement than DRASTIC model (ρ = 0.7818; AUC =67.36%). Thus, such modification in DRASTIC improves the performance of the model.