Application of Geoinformatics with Frequency Ratio (FR) Model to Delineate Different Groundwater Potential Zones in Ken Basin, India

The goal of present study was to make groundwater potential zones map using the frequency ratio (FR) model (i.e., probabilistic based bivariate statistical approach) and Geoinformatics technique in Ken Basin, India. Very few studies have been done yet to demarcate the groundwater potential zones using FR model in India. This research contains the analysis of spatial relationships between groundwater and its contributing factors viz., geology, geomorphology, lineament density, land use/land cover, soil texture, rainfall, slope, drainage density, depth to water level – pre monsoon and depth to water level – post monsoon. The ten groundwater contributing factors were collected from different resources and prepared in 10 m spatial resolution in ArcGIS 10.8. There are 425 observation wells located in the study area, out of these, 296 (70%) observation wells were randomly selected as training datasets for the model and the remaining 129 (30%) were used as a training dataset for the validation purpose. The final output of groundwater potential zones map was classified into five different zones as very high, high,moderate,low, and very low. Amore significant portionof the Ken Basin, nearly 30.66%, falls with in the moderate ground water potential zone followed by poor and good potential zones. For quantitative validation, ROC curve analysis was executed through the IBM SPSS software by comparing existing groundwater well locations in the validation datasets (well yield data) with the groundwater potential map obtained by the FR model. The validation results illustrated that the area under the curve (AUC) for the frequency ratio model is 78.1% which achieved fair satisfactory prediction accuracy. The outcome of this study is highly reliable which can serve as a guideline for the planning of exploration, exploitation, and sustainable management of groundwater resources in the study area.