Groundwater prospectivity modeling over the Akatsi Districts in the Volta Region of Ghana using the frequency ratio technique

To investigate the groundwater prospects over the Akatsi Districts of the Volta Region of Ghana, ten predictors (slope, topographic wetness index, stream power index, lineament density, drainage density, elevation, slope aspect, geology, land cover and soil type) that has been extracted from geological and remote sensing datasets have been integrated to generate a groundwater prospectivity model (GPM). The production of the model that characterises various potential zones of groundwater within the study area was carried out by examining the coherence between the classes of evidential layers (or predictors) and groundwater occurrences by applying the frequency ratio approach. Assessment of the GPM produced over the area indicates that, the soil type evidential layer greatly influenced the GPM the most; in comparison with the other nine evidential layers, whereas the predictor with the least influence on the model produced was the slope aspect. Using the area under the receiver operating characteristics curve, the performance of the GPM produced was determined to have a score of 0.876; depicting that the accuracy of the model generated is good. To further evaluate the performance of the model, the prediction-area plot was also employed on the GPM generated, with results obtained indicating that 76% of the known location of groundwater occurrence can be found within the prospectively delineated zones of groundwater (24% of the study area) over the study area.