Validation of Groundwater Potential Models Using Statistical Techniques: Insights from Electrical Resistivity Method

Groundwater exploration is crucial for ensuring sustainable water supply, particularly in regions with increasing population and surface water scarcity. This study focuses on the application and correlation of Multi-Criteria Decision Analysis (MCDA) techniques in the evaluation of groundwater potentiality in the study area (Ilere) of Ondo state, Nigeria. The integration of geophysical, geological and remote sensing data was carried out using the Preference Ranking Organization METHod for Enrichment Evaluations (PROMETHEE-II) and the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) models. The CRITIC and Entropy methods were employed for objective weight determination of seven key groundwater conditioning factors: Lithology, hydraulic conductivity, lineament density, transverse resistance, drainage density, elevation and transmissivity. The study utilized Python for data processing and modelling and thematic maps were generated within a GIS environment. The validation of the groundwater potential models was conducted using correlation analysis, Area Under The Curve (AUC) and p-values, achieving a “very good” performance rating with an AUC of 0.89 and a statistically significant p-value of 3.96e-7, indicating a strong relationship between the models. The study’s findings provide a robust framework for groundwater resource management in crystalline basement complex terrains, offering valuable insights for sustainable water development in similar regions.