Geochemical evolution, geostatistical mapping and machine learning predictive modeling of groundwater fluoride: a...

Around 2.6 billion people are at risk of tooth carries and fluorosis worldwide. Quetta is the worst affected district in Balochistan plateau. Endemic abnormal groundwater fluoride ( $$ ) lacks spatiotemporal studies. This research integrates geospatial distribution, geochemical signatures, and data driven method for evaluating $$ levels and population at risk. Groundwater $$ ranged from 0 to 3.4 mg/l in (n = 100) with 52% samples found unfit for drinking. Through geospatial IDW tool hotspot areas affected with low and high groundwater $$ levels were identified. Geochemical distribution in geological setups recognized sediment variation leads to high $$ (NaHCO₃) and low $$ (CaHCO₃) water types in low elevation (central plain) and high elevation (mountain foot) respectively. Results of the modified water quality index identified 60% samples to be unsuitable for drinking. Support vector machine (SVM), random forest regression (RFR) and classification and regression tree (CART) machine learning models found $$ , Salinity and $$ as important contributing variables in groundwater $$ prediction. CART model with R² value of 0.732 outperformed RFR and SVM in predicting $$ . Noncarcinogenic health risk vulnerability from $$ increased from Adults < Teens < Children < Infants. Infants and children with hazard quotient values of 11.3 and 4.2 were the most vulnerable population at risk for consuming $$ contaminated groundwater. The research emphasizes on both nutritional need and hazardous effect of $$ , and development of desirable limit for $$ .