Sustainable groundwater management is crucial for achieving the Sustainable Development Goals (SDGs), particularly in arid and semi-arid regions. Identifying high-potential zones is key to optimal exploitation and mitigating water crises. However, the available frameworks are inherently subjective due to the influence of expert judgment. This study develops an innovative formulation based on Bayesian Merging (BM) and utilizes Fuzzy Catastrophe (FC) for groundwater potential mapping. The formulations were implemented on an unconfined aquifer in the northwest of Iran. Six key data layers were considered: drainage density, distance from rivers, land use, lithology, Normalized Difference Vegetation Index (NDVI), and slope. FC utilizes fuzzy logic and catastrophe theory for layer normalization and weighting, while BM estimated groundwater occurrence suitability based on existing well data. The results showed that BM achieved an Area Under Curve (AUC) of 0.74, outperforming FC (AUC = 0.68) according to the receiver operating characteristic, and identified about 60% of successful well locations within 40% of the study area, compared to 50% for FC as per cumulative success rate. These results demonstrate that BM not only outperforms FC in identifying promising high-potential zones but also provides a more objective basis for decision-making in groundwater exploration. Combining the results of both approaches promises a more comprehensive and sustainable model for groundwater resource management.
