Groundwater is a vital freshwater resource that supports domestic, agricultural, and industrial activities in many regions worldwide. Accurate groundwater potential mapping (GPM) is essential for sustainable water resource management; however, traditional empirical and statistical approaches often struggle to capture the complex, nonlinear relationships among hydrogeological variables. In recent years, machine learning (ML) has emerged as a powerful data-driven approach for improving GPM accuracy and efficiency. This review synthesizes findings from 83 peer-reviewed studies published between 2015 and 2025, focusing on widely used ML algorithms such as Random Forest, Support Vector Machines, Artificial Neural Networks, and hybrid models. The review evaluates key methodological aspects, including input parameter selection, data partitioning, integration with GIS and remote sensing, and model justification techniques. It also discusses common challenges such as data limitations, regional variability, and model interpretability. The results indicate that ML-based approaches can significantly enhance groundwater prediction when supported by appropriate data and validation strategies. Future research directions include explainable artificial intelligence, uncertainty quantification, multi-source data integration, and improved model transferability. This review provides a comprehensive reference for advancing reliable and sustainable groundwater potential mapping.
Ali et al. (Wed,) studied this question.