Fuzzy analytical hierarchy process for groundwater potential mapping in a Mediterranean catchment: the case of the Medjerda catchment in northeast Algeria

In this study, groundwater potential was mapped for a Mediterranean catchment using remote sensing techniques, geographical information systems (GIS), and the fuzzy analytical hierarchy (FAHP) process. Seven environmental factors were considered to influence groundwater potential, including soil type, lithology, precipitation, land use/cover, slope, lineament density, and drainage density. A FAHP method was implemented to calculate the weight of each factor and each class in the groundwater potential. The FAHP confusion matrix was established by eliciting opinions of local experts on factors that drive groundwater potential in this area. A weighted overlay was made in Arc-GIS™ for producing the final groundwater potential map. Three groundwater potential classes were considered (low, moderate, strong). Areas of very high potential represent 107.09 km2 (8%) of the total surface of the basin and are located in alluvial and calcareous formations of a high density of lineament and in areas with a low slope which promotes rapid infiltration. About 1022.124 km2 (71%) of the surface of the sub-catchment is characterized by a moderate to strong potential, while the remainder 305.945 km2 (21%) is characterized by a low potential. The results of the classification are validated by establishing the receiver operating characteristic (ROC) curve using 61 data from the wells located in the study area. The area under curve (AUC) yields 75.5% demonstrating the effectiveness of the method in identifying groundwater potential areas.