Spatiotemporal Dynamics of Fluoride and Nitrate Contamination Driven by Groundwater Depletion and Climatic Stress: A Decadal Analysis Using GRACE and Machine Learning

High Resolution Image Download MS PowerPoint Slide Groundwater quality models often neglect dynamic water quantity, a gap amplified by the climate-driven hydrological change. This study examines the spatiotemporal evolution of geogenic fluoride (F − ) and anthropogenic nitrate (NO 3 − ) contamination across India (2008–2018) using an integrated modelling framework combining Gravity Recovery and Climate Experiment (GRACE)-derived groundwater storage anomalies with the Standardized Precipitation Index (SPI) and the Standardized Precipitation Evapotranspiration Index (SPEI) to quantify climatic stress. Hydrochemical facies identified using Piper diagrams and mineral Saturation Indices (SI) elucidate water−rock interactions and thermodynamic controls on contaminant mobilization. Results indicate widespread groundwater decline and increasing dryness by 2018, with contamination hotspots concentrated in western India. Groundwater became progressively undersaturated with respect to fluorite, enhancing dissolution of fluoride-bearing minerals. Cluster analysis grouped provinces into agroclimatic zones, supporting regional groundwater-quality assessment. A Random Forest (RF) model driven by hydroclimatic indicators showed strong predictive skill for contaminant exceedance, achieving a high Area Under the Curve (AUC) and balanced precision and recall (each 0.87). Overall, the study links hydroclimatic variability to geochemical processes, enabling quantity-informed groundwater-quality prediction.