Elevated uranium (U) (>WHO limit of 30 μg L −1 ) in Indian groundwaters is primarily considered geogenic, but the specific mineralogical sources and mechanisms for U mobilization are poorly understood. In this contribution, statistical and geochemical analyses of well-constrained metadata of Indian groundwater quality (n = 342 of 8543) were performed to identify key parameters and processes that influence U concentrations. For geochemical predictions, a unified speciation model was developed from a carefully compiled and updated thermodynamic database of inorganic, organic (Stockholm Humic model), and surface complexation reactions and associated constants. Critical U contamination was found at shallow depths (<100 m) within the Indo-Gangetic plain, as determined by bivariate nonparametric Kendall's Tau b and probability-based association tests. Analysis of aquifer redox states, multivariate hierarchical clusters, and principal components indicated that U contamination was predominant not just in oxic but mixed (oxic−anoxic) aquifers under high Fe, Mn, and SO 4 concentrations, presumably due to U release from dissolution of Fe/Mn oxides or Fe sulfides and silicate weathering. Most groundwaters were undersaturated with respect to relevant U-bearing solids despite being supersaturated with respect to atmospheric CO 2 (average pCO 2 of reported dissolved inorganic carbonate (DIC) data = 10 −1.57 atm). Yet, dissolved U did not appear to be mass limited, as predicted solubilities from reported sediment concentrations of U were ∼3 orders of magnitude higher. Integration of surface complexation models of U on typical aquifer adsorbents, ferrihydrite, goethite, and manganese dioxide, was necessary to explain dissolved U concentrations. Uranium contamination probabilities with increasing dissolved Ca and Mn exhibited minima at equilibrium solubilities of calcite [∼50 mg L −1 ] and rhodochrosite [∼0.14 mg L −1 ], respectively, at an average groundwater pH of ∼7.5. A potential indirect control of such U-free carbonate solids on U mobilization was suggested. For locations (n = 37) where dissolved organic carbon was also reported, organic complexes of U contributed negligibly to dominant U speciation at the groundwater pH. Overall, the unified model suggested competitive dissolution−precipitation and adsorption−desorption controls on U speciation. The model provides a quantitative framework that can be extended to understand dominant mobilization mechanisms of geogenic U in aquifers worldwide after suitable modifications to the relevant aquifer parameters.
