Comparative Assessment of Weights of Evidence and Statistical Index Models for Groundwater Potential Mapping in the Dehgolan Plain, Iran

Groundwater behavior and related processes, such as production, movement, and the effects of extraction, occur in complex subsurface formations, making direct observation and analysis challenges. Quantitative assessment, management, and utilization of groundwater resources pose significant challenges. This study aims to map the groundwater potential of the Dehgolan Plain in Kurdistan Province, Iran. It compares the performance of two methods: Weight of Evidence (WoE) and Statistical Index (SI), to identify the most suitable model for groundwater potential assessment. In this research, field data from wells and groundwater sources were collected, and a random algorithm split the data into 70% (103 wells) for training and 30% (45 wells) for validation. The groundwater potential map was validated using the ROC curve in SPSS software. The ROC curve analysis yielded Area Under the Curve (AUC) values of 79.5% for the WoE model and 85.9% for the SI model, indicating that both models are suitable for groundwater potential mapping, with the SI model showing higher accuracy. Ultimately, the study concluded that the SI method is more effective for groundwater potential assessment in the Dehgolan Plain.