Mineralogic controls on the composition of natural waters dominated by silicate hydrolysis

A detailed mineral-solute mass-balance spreadsheet model has been developed to improve understanding of mineralogic controls on water composition in surface waters and groundwater systems dominated by silicate lithologies. The spreadsheet-based approach is used to investigate mass-balance solutions over a range of mineral compositions for predominant, multi-element phases such as plagioclase, biotite, chlorite, amphibole, and dioctahedral smectite. The mass-balance technique has been applied to a variety of hydrogeologic environments including calibrated watersheds, springs and groundwater systems. Lithologic settings include plutonic, metamorphic, minor volcanic, and sedimentary (alluvial and glacial deposits) materials. Sites investigated, most of which allowed comparison with earlier work, include ephemeral and perennial springs of the central Sierra Nevada Mountains, California; the Loch Vale, Colorado high-altitude watershed; four catchments in the Inyo Mountains, California and Nevada; watersheds in the Catoctin Mountain, Maryland area; and ground-water How systems in Vekol Valley, Arizona and in the Trout Lake area, Wisconsin. Analytical mass balances were derived with a novel approach that permits graphic analysis of the functional dependency of mineral mass-transfer coefficients on variations in major silicate composition. A 10 x 10 matrix of mineral-solute reaction equations was used to solve for the best mass balances. Alternative methods were developed to investigate cases where the number of mineral phases exceeds the number of solute constraints (10) that are commonly found in clastic, silicate-dominated systems with diverse sources and/or phase mixtures of variable chemistry (such as an assortment of plagioclase compositions). Techniques were also developed to allow for important structurally controlled compositional variation in clay minerals. Final mass-balance choices were tempered by thermodynamic stability calcudations for various minerals under near-surface weathering conditions. Mineral mass-transfer ratios were also found to be useful in restricting mass balances. Additional constraints were provided by establishing consistent sets of similar mineral assemblages and mineral compositions for a series of wells along a ground-water flow path, or through comparison of multiple nearby watersheds subject to similar climatic and vegetative conditions. Whereas unique analytical solutions from spreadsheet mass balances were not always possible, a surprisingly limited range was obtained for mineral assemblages and compositions satisfying observed mineral occurrences and thermodynamic stability conditions. Our results demonstrate the importance of having well-characterized mineral compositions for both reactant and product phases. The mass-balance results are particularly sensitive to the compositions of plagioclase, dioctahedral smectite, and a few primary mafic mineral groups (trioctahedral mica, amphibole, and pyroxene). In addition to mineral controls, the specific examples studied demonstrate effects of dolomite and calcite overprints on silicate weathering, biomass fluctuations, road salt contamination, and on buried evaporite salts. A comparison of the mass-balance results from surface- and groundwater systems indicates strong similarities, suggesting that similar processes dominate the mineral controls in both settings, and that, under soil weathering conditions ion exchange might have been overemphasized, except under conditions where pre-existing waters are displaced by solutions of significantly different composition.