Groundwater Potential Zone Identification in A Coastal Region Using AHP and ML Technique

Abstract

In coastal regions, due to the over-extraction of groundwater, the regions are facing the issue of seawater intrusion, which causes depletion of fresh water and degradation of groundwater quality. Therefore, the availability of water for drinking and irrigation is reduced. In the present study, the Analytical Hierarchical Process (AHP) and Machine Learning technique are used to identify the potential zones for groundwater which will be useful for the development of effective strategies for effective water conservation and recharge as well as ensuring a reliable and uncontaminated water source for local communities. The study was conducted in the Baleswar district which is a coastal region of Odisha. The study integrates multiple hydrogeological, topographical, and climatic factors, including rainfall, land use/land cover (LULC), drainage density, slope, soil type, geology, and groundwater level. AHP is employed to assign weights to the influencing factors, while the machine learning (ML) model Random Forest (RF) is used for automated classification and prediction of groundwater potential zones. The results are validated using field data and the statistical index correlation coefficient. The comparative study will be useful for sustainable management of water resources.