Comparative analysis of groundwater potentiality zone using fuzzy AHP, frequency ratio and Bayesian weights of evidence methods

Abstract

Groundwater resource management in the urban area is one of the important aspects because of growing population demand and having inadequate water supply. So, proper information is needed to manage the future urban planning for such kinds of areas. This study emphasizes groundwater potentiality zone (GPZ) assessment in the Asansol urban agglomeration (AUA) region, West Bengal, India. For this purpose, we have incorporated eight conditioning factors namely LULC, lithology, slope, elevation, rainfall, drainage density, lineaments density, and soil map using conventional and remote sensing data in GIS software. All these conditioning factors have been reclassified in ArcGIS and processed by the fuzzy analytical hierarchy process (FAHP), frequency ratio (FR), and Bayesian weights of evidence (BWOE) statistical methods. Then, the groundwater potential index has been formulated, and finally, GPZ maps are generated based on of selected three models. The result shows that very high area of GPZ, e.g. 9.13%, 11.62% and 7.43% are under BWOE, FR and FAHP models, respectively. The receiver operating characteristic validation curves show that FR method (AUC = 96.4%) is well obtained for GPZ in comparison with both BWOE (AUC = 83.8%) and FAHP methods (AUC = 82.9%). Therefore, this statistical method is highly recommended for the study of groundwater potential assessment and this outcome is very suitable for the groundwater resources management in future land use planning. Precautionary works in low potential areas should be given priority for long-term planning. Thus, this study can be considered as a good document for decision support in water exploitation planning and sustainable groundwater management in AUA region.