Archives

The water deficit in Nzhelele-Makhado area calls for an efficient assessment to delineate and map existing alternative water resources for future extraction. We are of a view that the identification of groundwater potential zones (GWPZ) in an area is as equally important as determining the most effective groundwater mapping models for the efficient production of reliable GWPZ maps. Consequently, relative frequency ratio (RF), with predictor value, and analytic hierarchy process (AHP) models, including two RF-AHP hybrid models (HM-A and HM-B) were applied and compared to evaluate their effectiveness and accuracy in groundwater potential mapping of Nzhelele-Makhado area. HM-A model is based on the ranks of classes of groundwater controlling factors that were computed by RF model and weights of controlling factors computed by AHP model. Meanwhile, for HM-B model it was vice versa. Eight groundwater controlling factors (slope, geomorphology, lithology, lineament density, soil, rainfall and land use land cover) were considered for this assessment. These were derived and extracted from conventional and remotely sensed datasets. The resultant GWPZ maps computed by the four models were classified into five categories: very low, low, moderate, high and very high groundwater potential zone. The RF, AHP, HM-A and HM-B has an area coverage of 13.62%, 21.50%, 14.56% and 19.62%, respectively, in very high potential zone. The area under curve (AUC) was used to assess the accuracy of the four models and evaluate the resultant maps using 37 geosites in the area. The RF, AHP, HM-A and HM-B show the AUC value of 72.47%, 60.55%, 71.96% and 49.85%, respectively. The models based on RF-derived ranks show better accuracies than models based on AHP-derived ranks. This suggest that the ranks have more influence in the performance of models than weights. The weights derived by AHP and RF models were generally consistent; thus, confirming that the ranks have more influence on the resultant maps. The maps generated from RF-derived ranks can therefore be used for further ground-based investigation of groundwater potential sources in the area.