Evolution of groundwater chemical composition by plagioclase hydrolysis in Norwegian anorthosites

The Precambrian Egersund anorthosites exhibit a wide range of groundwater chemical composition (pH 5.40–9.93, Ca2+ 1.5–41 mg/L, Na+ 12.3–103 mg/L). They also exhibit an evolutionary trend, culminating in high pH, Na-rich, low-Ca groundwaters, that is broadly representative of Norwegian crystalline bedrock aquifers in general. Simple PHREEQC modelling of monomineralic plagioclase–CO2–H2O systems demonstrates that the evolution of such waters can be explained solely by plagioclase weathering, coupled with calcite precipitation, without invoking cation exchange. Some degree of reaction in open CO2 systems seems necessary to generate the observed maximum solute concentrations, while subsequent system closure can be invoked to explain high observed pH values. Empirical data provide observations required or predicted by such a model: (i) the presence of secondary calcite in silicate aquifer systems, (ii) the buffering of pH at around 8.0–8.3 by calcite precipitation, (iii) significant soil gas CO2 concentrations (PCO2 > 10−2 atm) even in poorly vegetated sub-arctic catchments, and (iv) the eventual re-accumulation of calcium in highly evolved, high pH waters.