Adsorbed silica in stalagmite carbonate and its relationship to past rainfall

Despite considerable work on other trace elements, the incorporation of dissolved silicon from cave waters into speleothems has not been previously investigated. In this study, the controls on dissolved Si in cave waters and on adsorbed Si in resulting speleothems are therefore investigated. Bedrock (dolomite), soil water, dripping water, and cave carbonates were retrieved from Heshang Cave situated in the central Yangtze valley of China and were subjected to analysis of dissolved Si content (plus accompanying Ca and Fe analyses). Soil waters have Si/Ca of 45.5 mmol/mol, compared to only 3.2 mmol/mol in the dolomite bedrock, demonstrating that >80% of the dissolved Si must come from dissolution of silicate minerals in the soil. Drip waters have a dissolved Si concentration of ≈4.2 μg/mL, similar to that in the overlying soil water. Actively growing cave carbonates have a Si/Ca of 0.075 mmol/mol suggesting a partition coefficient for incorporation of dissolved silicon of 0.0014, in good agreement with previous laboratory studies. Extrapolating the results of these laboratory studies to the cave environment suggests that changes in Si/Ca in cave carbonates are likely to be primarily controlled by changes in drip-water Si/Ca. The drip-water Si/Ca will, in turn, be controlled by the rate of wind-blown silicate supply; by soil weathering rates; by rainfall dilution; and by precipitation of calcite. The general expectation is that these effects combine to produce high Si/Ca in speleothems during times of low rainfall. A δ18O record from a Heshang Cave stalagmite which grew between 20 and 11 thousand years ago allows these controls to be tested. Correlation of high Si/Ca with high δ18O demonstrates that regional rainfall exerts significant (but not complete) control on speleothem Si/Ca. With further understanding, speleothem Si/Ca may provide a proxy for past rainfall to complement existing proxies such as δ18O and Mg/Ca.