Assessment of aquifer vulnerability using a developed GODL method (modified GOD model) in a schist belt environ, Southwestern Nigeria

Developing a predictive decision model for assessing the vulnerability of hidden groundwater reservoir formation to contamination risk via unavoidable anthropogenic activities is a key to managing water resources looming security crisis globally. This study explored multiple and robust methodologies including GIS, analytical hierarchy process (AHP)-based data mining, statistical and geophysical techniques for developing a novel GODL vulnerability method: a modified GOD model to ameliorate these challenges. The input for the modeling was based on the 65 located depth sounding geophysical data occupied in a schist belt environ, Southwestern Nigeria. From the geophysical data interpreted results, four factors, namely, groundwater hydraulic confinement (G), aquifer overlying strata (O), depth to water table (D), and longitudinal conductance (L), regarded as aquifer vulnerability causative factors (AVCFs) were derived. The GIS-based produced AVCFs' themes were synthesized by employing the conventional GOD and the AHP-driven GODL algorithms. Based on these algorithms applied results, the GOD-based aquifer vulnerability prediction zone map and GODL-based aquifer vulnerability prediction zone (AVPZ) map were produced in GIS environment. The produced AVPZ maps were validated by applying the statistical model evaluation to the water chemistry correlation results. The validation result exhibits 70% prediction accuracy for the developed GODL model compared with 66% for the GOD model. The GODL model demonstrated better performance than the GOD model. The AVPZ maps produced in this study can be used for precise decision-making process in environmental planning and groundwater management.

Automated summary

Developing a predictive decision model for assessing vulnerability of hidden groundwater reservoir formation to contamination risk is crucial for managing global water resources security crisis. This study utilized multiple methodologies including GIS, AHP-based data mining, and statistical and geophysical techniques to develop a novel GODL vulnerability method. The validation results showed that the GODL model performed better than the GOD model, demonstrating 70% prediction accuracy compared to 66% for the GOD model.