Delineation of groundwater potential zones in Torbate Jam district using maximum entropy and AHP methods

With the increase in population, the importance of groundwater resources as one of the most important sources of drinking water in the arid regions becomes apparent. In this research, in order to determine the areas with groundwater potential in the city of Torbate Jam and prioritizing the effective factors, hierarchical analysis methods and maximum entropy method using MaxEnt model and the factors of distance from fault and fault density, lithology, slope, slope direction, distance from the waterway and drainage density, elevation, land use, slope curvature, topographic humidity index and topographic position indicator was used. Also, for assessing these two methods, the Receiver Operating Characteristic (ROC) was used. From 220 sources, 30% were randomly assigned as validation data and 70% were categorized as test data in maximum entropy method. Results showed that 29.6% of the watershed had high groundwater potential according to the maximum entropy method. Based on Jack-Knife Diagram, DEM, slope, distance from fault and lithology were the most important factors affecting groundwater potential, respectively. The Area Under the Curve (AUC) in the maximum entropy method indicated a precision of 91% (excellent) at the training period and 80% (very good) in the validation period to determine areas with potential for groundwater. Based on AHP method, 34.4% of the area has groundwater potential, and the slope layers, lithology, elevation and distance from the faults were the most important factors, respectively and accuracy of this method was 73%. The results showed that applying AHP and maximum entropy methods, while saving time and cost, have a good ability to predict the potential of groundwater and the maximum entropy method has more superiority than the hierarchical analysis method.