The Effects of Freshwater Dissolution on Coral Geochemistry and Morphology

The aragonite skeletons of massive reef-building corals provide an invaluable high-resolution archive of past climate variability. However, studies have repeatedly shown that alteration of the coral skeleton (diagenesis), occurring as secondary cements and/or dissolution, is fairly prevalent among both modern and fossil corals. While the effects of secondary cements on bulk coral geochemistry and morphology have been extensively documented, the impacts of dissolution remain relatively unconstrained. Given that aragonite is metastable and that most fossil corals are exposed to rainfall for long periods, it follows that dissolution-related artifacts in fossil coral paleoclimate records merit further study. To date, the only study on coral dissolution suggests that dissolution does not significantly impact the oxygen isotopic ratios (δ18O), but leads to an increase of up to +0.06mmol/mol in coral Sr/Ca (equivalent to sea-surface temperature (SST) cooling artifacts of -1.2°C using standard Sr/Ca paleo-temperature conversions) [Hendy et.al., 2007]. Here we investigate the effects of freshwater dissolution on coral geochemistry and morphology by exposing a 2.5cm x 5cm segment of a modern coral from Palmyra Island (6°N, 162°W) to a constant freshwater drip for one week in order to simulate the effects of rainfall. Scanning electron microscope (SEM) images and coral δ18O and Sr/Ca measurements were taken before and after the coral was dissolved to assess the impacts of dissolution. We observe that dissolution occurs both on the surface and within the coral skeleton. Surface coral dissolution results in a "bumpy" surface, while interior dissolution targets centers of calcification. In the heavily-dissolved portions of the coral, nearly 60% of the skeleton surface is dissolved. In these areas δ18O shows a significant decrease of -0.2-0.4‰, which would correspond to an SST increase of +1-2°C in paleotemperature [Epstein, 1953]. Due to the relatively large range of Sr/Ca variability in the undissolved coral sample, we cannot detect any signature of dissolution on coral Sr/Ca. We estimate that dissolution-related coral Sr/Ca offsets must be < 0.1mmol/mol in our dissolved coral sample (corresponding to a SST cooling artifacts of <1°C using the Palmyra Sr/Ca-SST calibration [Nurhati et.al.,2009]). In characterizing the geochemical and morphological signatures of coral skeletal freshwater dissolution, we provide coral paleoclimatologists with 1) a means of identifying dissolution in fossil coral sample via SEM, and 2) estimates of dissolution-related biases in the corresponding paleoclimate reconstructions. References 1. Epstein, S., Mayeda T. (1953), Geochimica et Cosmochimica Acta 4, 213-224. 2. Hendy, E.J., et.al. (2007), Paleoceanography 22, PA4101. 3. Nurhati, I.S., et.al. (2009), Geophysical Research Letters 36, L21606.