HIGH-RESOLUTION SPELEOTHEM GEOCHEMISTRY: CONTINENTAL LITHOSPHERE TRACKING THE ATMOSPHERE

Freshwater carbonate systems offer settings where the atmosphere, via the hydrosphere, destroys the lithosphere, but creates new lithosphere as a by-product. Carbonate dissolution is the most active weathering reaction on the planet, but the geochemistry of the resulting solutions offer complexities which can be recast as opportunities (best presented in the form of speleothems in karst caves) to read the history of atmosphere-lithosphere interaction in response to climate, and even weather.

Calcite dissolves appreciably more rapidly than dolomite and so in a mixed dolomite-limestone system, solution Mg/Ca can therefore track residence time. In turn water residence time in an aquifer reflects recharge and hence the seasonal distribution of atmospheric precipitation. This tendency is accentuated by any soil processes (freezing or evaporation) or aquifer degassing of CO2 causing CaCO3 reprecipitation, since residual solutions have raised Mg/Ca. As a result of such processes, many dripwaters display inverse relationships between Mg/Ca (and sometimes Sr/Ca) and drip rate.

From speleothems of shallow karst cave from seven European sites, it is normal to find an annual cyclicity of trace element variation, as determined by ion microprobe analysis. This is most strongly expressed by P, a nutrient element which provides a link to the seasonal temperature cycle by the growth and die-back of vegetation, but which is also influenced by hydrology. Na and H tend to vary in parallel and are influenced by variations in crystal growth rate, as is Sr. The partition coefficient for Mg in calcite at constant temperature depends primarily on solution Mg/Ca and hence on hydrology as discussed above.

A theoretical and observational forward modelling approach to understanding the controls on speleothem geochemistry is complemented by an empirical backward modelling approach whereby elemental records are compared with instrumental data. The first transfer function from speleothem geochemistry to hydrologically-effective atmospheric precipitation, based on a speleothem from the Alps of NE Italy, is currently being developed.