COUPLING DATA AND MODELS TO BETTER UNDERSTAND KARST DRIP WATER CHEMISTRY

Karst systems are useful for examining spatial and temporal variability in Critical Zone processes. They provide a window into the subsurface where waters have interacted with vegetation, soils, and bedrock across a range of length and time scales. Furthermore, these hydrologic pathways frequently include the formation of speleothems, which provide long-term archives of these interactions. Through sampling waters from soils and drip sites within a cave on a multi-year timescale, it is possible to observe how environmental signals are translated and modified along water flow paths. However, interpreting time-averaged speleothem records requires integration of observational data and simulation studies that provide a temporal bridge between long and short-term processes.

To address this problem, we are adapting the reactive transport model CrunchTope for karst systems. To parameterize the model, we use the chemical and mineralogic properties of the soils and host rocks at Blue Spring Cave in Tennessee. We validate the model using data from coordinated measurements of elemental and isotopic signatures of waters and gases moving along different flow paths within and between the soil, epikarst, and cave. For 3 years we have monitored surface and soil temperature, precipitation and soil moisture, cave temperature and pCO₂, drip rate, and drip elemental and isotopic chemistry. The range of variability in drip water δ¹⁸O indicates some drips are fed by fracture flow from the surface, while others are fed by diffuse flow. For both drip types, δ¹³C_DIC is inversely related to monthly rainfall. Cave air pCO₂ suggests seasonal ventilation driven by surface air temperature change. Drip water Sr/Ca and Mg/Ca indicate prior carbonate precipitation occurs in the epikarst, but do not appear to reflect cave ventilation. Initial model runs reproduce drip water Ca concentrations, with fracture-fed drips matching faster flow rates, and diffuse-fed drips matching slower flow rates due to increased precipitation of secondary calcite along the flow path. However, further modifications are needed to match drip water Mg concentrations. This coupled measurement/modeling approach shows great promise for using simulations of the long-term effects of climate change on cave waters to improve our interpretation of speleothem records.