Delineation of groundwater potential zones using a parsimonious concept based on catastrophe theory and analytical hierarchy process

• Main Authors: Shahinuzzaman, M., Haque, M. N., Shahid, S.
• Format: Article
• Published: Springer Science and Business Media Deutschland GmbH 2021
• Subjects: TA Engineering (General). Civil engineering (General)
• Online Access: http://eprints.utm.my/id/eprint/94347/; http://dx.doi.org/10.1007/s10040-021-02322-2

A cost-effective “parsimonious” approach to delineating groundwater potential zones is proposed, based on catastrophe theory (CT) and analytical hierarchy process (AHP) in a geographic information system (GIS). Eleven indicators that influence groundwater storage (slope, drainage density, surface-water body (proximity), soil permeability, aquitard thickness, aquifer thickness, hydraulic conductivity, specific yield, recharge, aquitard resistivity and aquifer resistivity) were prepared. The suitable weights of the factors and the index values of the features of the factors were normalized using AHP and CT multicriteria decision analysis (MCDA) techniques for the development of a groundwater potential index (GWPI) map. Finally, the relative sensitivity of the factors was evaluated to develop a parsimonious groundwater potential index (P-GWPI) map using the most sensitive themes. GWPI and P-GWPI maps were validated using 14-year average annual post-monsoon depth to groundwater level data of 36 monitoring wells in a study area in Bangladesh. The generated GWPI map classified the study area as moderately good, good and very good groundwater potential covering an area of 19.5, 40.3 and 40.2% respectively. Subsequently, a modified GWPI map was developed using effective weights derived from single-parameter sensitivity analysis. The P-GWPI map developed using the most sensitive factors categorized the groundwater potential zones as moderately good (13.0%), good (38.2%) and very good (48.8%). The results of this study can serve as guidelines for future groundwater exploration, planning and management of the area, and the methodology used can also be easily adopted in other similar and data-scarce areas.