Enhanced groundwater vulnerability assessment to nitrate contamination in Chongqing, Southwest China: Integrating novel explainable machine learning algorithms with DRASTIC-LU

Abstract

Groundwater vulnerability to nitrate assessment serves as a measure of potential groundwater nitrate pollution in a target area. The primary objective of this study is to utilize the traditional DRASTIC-land use assessment framework, groundwater nitrate distribution data, and three machine learning models (random forest (RF), XGBoost, and support vector machine) to classify whether groundwater nitrate exceeds a threshold (10 mg/L as nitrogen) in Chongqing, southwest China. Model evaluation is conducted using accuracy and F1 score metrics, and ultimately, the classification probabilities are employed as the groundwater vulnerability to nitrate index. The results indicate that the RF model outperforms the other two models, achieving the highest accuracy (92.9% for testing), kappa value (0.857 for testing), and area under the curve (0.948 for testing). Furthermore, the SHapley Additive exPlanations (SHAP) interpreter revealed that aquifer conductivity, lithology, agricultural activities, areas with high-intensity development, and groundwater recharge are the most influential indicators of groundwater vulnerability. The final groundwater vulnerability level distribution map, with a resolution of 1 km × 1 km, reveals that high and extremely high vulnerability levels are concentrated in areas with high-intensity urban development and karst trough valleys in the southeastern, northeastern, and central urban areas. This work represents the first attempt at using machine learning models for groundwater vulnerability assessment in Chongqing.