Comparative evaluation of groundwater vulnerability to pollution using DRASTIC and modified DRASTIC models: case study in Woleka sub-basin, upper Blue Nile, Ethiopia

Abstract

The increasing use of agricultural inputs has raised concerns about groundwater pollution. Therefore, assessing the vulnerability of groundwater systems to pollution is essential in agricultural watersheds either using specific or intrinsic vulnerability. This research investigates the intrinsic vulnerability of groundwater resources to pollution in the Woleka sub-basin. To achieve this, DRASTIC and modified DRASTIC (DRASTIC_LU) models were applied. The GIS software was used to prepare the required parameter maps and to execute overlay analysis. A sensitivity analysis was conducted by employing the single-map removal method. To assess model performance, 33 groundwater samples’ nitrate concentrations were used, and verified using the ROC and AUC curve methods. The results indicated that, based on the DRASTIC model, only 12.68% of the study area demonstrated a very high vulnerability to groundwater pollution. However, when accounting for anthropogenic influences, the percentage of areas classified as very highly vulnerable increased to 56.86%. The model’s sensitivity showed that hydraulic conductivity is a significant factor affecting groundwater pollution using both methods. Both models’ performance fell within a similar range; the modified DRASTIC_LU model demonstrated higher predictive accuracy compared to the standard DRASTIC method in the study area. These findings provide valuable insight for decision-makers by serving as an early warning mechanism for groundwater contamination risks in the study area.