Groundwater Potential Mapping using the weights of evidence and statistical indexes in Najafabad, Iran

Groundwater resources are one of the most valuable sources of water for various uses. In this study, the potential of groundwater resources in Najafabad study area was evaluated using statistical index and weight of evidence. For this purpose, 4484 wells in the area with a discharge of more than 11 liters per second were randomly divided into two groups (70%) and validation (30%). Then digital layers of variables affecting groundwater potential (Elevation, Slope, Slopedirection, Surface curvature, Topographic wet index (TWI), Land use, Soil, Lithology, River distance and Drainage density) were constructed in ArcGIS10.5 environment. And then, using the weighted overlay method, the final groundwater potential map was prepared. Finally, the ROC curve of the groundwater potential map was prepared to determine the accuracy of the final map using SPSS19 software. Accuracy based on this curve was 93.2% for the statistical index model and 74.1% for the evidence weight model, indicating high accuracy of the statistical index model compared to the evidence weight model for this region. Therefore, it can be said that using GIS-based models for potential groundwater prone areas are appropriate and reliable.