Abstract:
Most models of cave formation in limestone that remains near its depositional environment and has not been deeply buried (i.e. eogenetic limestone) invoke dissolution from mixing of waters that have different ionic strengths or have equilibrated with calcite at different pCO2 values. In eogenetic karst aquifers lacking saline water, mixing of vadose and phreatic waters is thought to form caves. We show here calcite dissolution in a cave in eogenetic limestone occurred due to increases in vadose CO2 gas concentrations and subsequent dissolution of CO2 into groundwater, not by mixing dissolution. We collected high-resolution time series measurements (1 year) of specific conductivity (SpC), temperature, meteorological data, and synoptic water chemical composition from a water table cave in central Florida (Briar Cave).We found SpC, pCO2 and calcite undersaturation increased through late summer, when Briar Cave experienced little ventilation by outside air, and decreased through winter, when increased ventilation lowered cave CO2(g) concentrations.We hypothesize dissolution occurred when water flowed from aquifer regions with low pCO2 into the cave, which had elevated pCO2. Elevated pCO2 would be promoted by fractures connecting the soil to the water table. Simple geochemical models demonstrate that changes in pCO2 of less than 1% along flow paths are an order of magnitude more efficient at dissolving limestone thanmixing of vadose and phreatic water.We conclude that spatially or temporally variable vadose CO2(g) concentrations are responsible for cave formation becausemixing is too slow to generate observed cave sizes in the time available for formation. While this study emphasized dissolution, gas exchange between the atmosphere and karst aquifer vadose zones that is facilitated by conduits likely exerts important controls on other geochemical processes in limestone critical zones by transporting oxygen deep into vadose zones, creating redox boundaries that would not exist in the absence of caves.