Drinking Water Resources Suitability Assessment Based on Pollution Index of Groundwater Using Improved Explainable Artificial Intelligence

Abstract

The global significance of fluoride and nitrate contamination in coastal areas cannot be overstated, as these contaminants pose critical environmental and public health challenges across the world. Water quality is an essential component in sustaining environmental health. This integrated study aimed to assess indexical and spatial water quality, potential contamination sources, and health risks associated with groundwater resources in Al-Hassa, Saudi Arabia. Groundwater samples were tested using standard methods. The physiochemical results indicated overall groundwater pollution. This study addresses the critical issue of drinking water resource suitability assessment by introducing an innovative approach based on the pollution index of groundwater (PIG). Focusing on the eastern region of Saudi Arabia, where water resource management is of paramount importance, we employed advanced machine learning (ML) models to forecast groundwater suitability using several combinations (C1 = EC + Na + Mg + Cl, C2 = TDS + TA + HCO3 + K + Ca, and C3 = SO4 + pH + NO3 + F + Turb). Six ML models, including random forest (RF), decision trees (DT), XgBoost, CatBoost, linear regression, and support vector machines (SVM), were utilized to predict groundwater quality. These models, based on several performance criteria (MAPE, MAE, MSE, and DC), offer valuable insights into the complex relationships governing groundwater pollution with an accuracy of more than 90%. To enhance the transparency and interpretability of the ML models, we incorporated the local interpretable model-agnostic explanation method, SHapley Additive exPlanations (SHAP). SHAP allows us to interpret the prediction-making process of otherwise opaque black-box models. We believe that the integration of ML models and SHAP-based explainability offers a promising avenue for sustainable water resource management in Saudi Arabia and can serve as a model for addressing similar challenges worldwide. By bridging the gap between complex data-driven predictions and actionable insights, this study contributes to the advancement of environmental stewardship and water security in the region.

Water, Resources, and Resilience: Insights from Diverse Environmental Studies

“…By leveraging AI's capabilities, researchers can develop predictive models, optimize resource allocation, and enhance decision-making processes. AI-driven monitoring systems can provide real-time insights into water quality, usage patterns, and network vulnerabilities, thereby promoting resilience and sustainability [64][65][66][67][68][69][70][71][72][73].…”

Insightful Analysis and Prediction of SCOD Component Variation in Low-Carbon/Nitrogen-Ratio Domestic Wastewater via Machine Learning

“…Shapley additive explanations (SHAP) analysis was performed to enhance the reliability and validity of the results calculated by the ML algorithm [49]. Therefore, the SHAP value, determined by the optimum Shapley value from game theory, offers a rational way to allocate payoffs among coalition members [50]. A subset of the eigenvalues and predictions of the subset represent the coalition and payoffs of the coalition, respectively.…”

Application of machine learning techniques to predict groundwater quality in the Nabogo Basin, Northern Ghana

“…In recent years, Water Quality Index (WQI) integrated with Machine Learning (ML) and Deep Learning, as many innovative techniques [ 18 ]. Deshpande et al [ 19 ], Goodarzi et al [ 20 ], Patel et al [ 21 ], Siriwardhana et al [ 22 ],Abba et al [ 23 ] and Xiong et al [ 24 ] assessed water, and groundwater quality in their various study areas, where their findings from WQI methods proved significant for water quality management, and it recommended the replication of these methods in different territories of the world.…”