Groundwater Quality and Health Risk Assessment Using Indexing Approaches, Multivariate Statistical Analysis, Artificial Neural Networks, and GIS Techniques in El Kharga Oasis, Egypt

The assessment and prediction of water quality are important aspects resource management. Therefore, the groundwater (GW) Nubian Sandstone Aquifer (NSSA) in El Kharga Oasis was evaluated using indexing approaches, such as drinking index (DWQI) health (HI), supported with multivariate analysis, artificial neural network (ANN) models, geographic information system (GIS) techniques. For this, physical chemical parameters were measured for 140 GW wells, which indicated Ca–Mg–SO4, mixed Ca–Mg–Cl–SO4, Na–Cl, Ca–Mg–HCO3, Na–Ca–HCO3 facies under influence silicate weathering, rock–water interactions, ion exchange processes. had high levels heavy metals, particularly iron (Fe) manganese (Mn), average concentrations above limits recommended by World Health Organization (WHO) water. DWQI categorized most samples not suitable (poor to very poor class), while some fell good class. results HI a potential risk due ingestion water, being higher children only one location. However, both adults, there low dermal exposure all locations. contaminants could be from natural sources, minerals leaching rocks soil, or human activities. Based on ANN modeling, ANN-SC-13 accurate model, since it demonstrated strongest correlation between best characteristics DWQI. example, this model’s thirteen extremely predicting R2 value training, cross-validation (CV), test data 0.99. ANN-SC-2 model measuring adults. CV, 1.00 models. at detecting adults (R2 = 0.99, 0.99 sets, respectively). Finally, integration physicochemical parameters, indices (WQIs), models can help us understand its controlling factors, implement necessary measures that prevent outbreaks various water-borne diseases detrimental health.