Influence of pH and Temperature on Elemental and Isotopic Composition of Cultured Scleractinian Corals

Scleractinian corals are known for strong "vital effects". Nevertheless, they have successfully been used as proxy recorders for a variety of environmental parameters. Understanding the processes inducing "vital effects" is essential for a successful application of elemental and isotopic proxies. In order to investigate the effects of pH and temperature on isotope and element partitioning, a branching scleractinian coral of the genus Pocillopora was grown under controlled laboratory conditions in natural seawater (i) at a temperature range from 22 to 28°C at constant pH; and (ii) at a pH range from 7.8 to 8.3 at constant temperature. Samples were analyzed for boron (δ11B), oxygen (δ18O), carbon (δ13C), and calcium (δ44/40Ca) isotopes, as well as Sr/Ca and Mg/Ca ratios. The temperature influence on δ18O in Pocillopora sp. (-0.14‰/°C) is in the lower range of sensitivities observed in corals (e.g., Reynaud-Vaganay et al., 1999, MEPS 180, 121-130). The temperature response of Sr/Ca in Pocillopora sp. (-0.07mmol/mol/°C) compares well with the range of observed sensitivities of corals (-0.03 to -0.09, Reynaud et al. 2007 GCA 71, 354-362) and inorganic aragonite (-0.045, Dietzel et al. 2004 Chem. Geol. 203, 139-151). The Mg/Ca-temperature sensitivity (0.15 mmol/mol/°C) agrees with the one determined in culture experiments of Acropora sp. (0.14, Reynaud et al. 2007, GCA 71, 354-362). The pH shows a clear control on δ11B in Pocillopora sp., increasing systematically with pH by about +10 ‰/pH-unit. Data comply with isotopic signatures of other branching corals (Hönisch et al. 2004 GCA 68, 3675-3685; Reynaud et al. 2004, Coral Reefs 23, 539-546). However, they are still systematically enriched in 11B relative to the empirical values for aqueous borate and inorganic calcite (Klochko et al. 2006 EPSL 248, 261-270). Oxygen isotopes also demonstrate a weak response to pH (-0.7 ‰/pH-unit), similar to the dependence found in planktic foraminifera by Spero et al. (1997 Nature 390, 497-500). Finally, Mg/Ca in Pocillopora sp. shows a weak dependence on pH (0.52 mmol/mol/pH-unit). Sr/Ca and Ca isotopes show no significant variation. Covariation of coral Mg/Ca and Sr/Ca with ambient carbonate ion concentration was recently explained by Cohen et al. (2009 Geochem. Geophys. Geosys. 10, 1-12) with a Rayleigh distillation model where the efficiency of cation utilization in a calcifying fluid increases with CaCO3 saturation. This model implicitly predicts a significant covariation of the skeletal calcium isotopic composition with seawater pH. The Rayleigh model quantitatively explains the relatively high Ca isotope ratio observed in coral aragonite. However, our data show only 10% of the Mg/Ca variations predicted by the Rayleigh model for the investigated pH range. Additionally, no significant change in Sr/Ca and Ca isotope ratios was observed. Consequently, our data suggest that the Rayleigh model, which was developed using data from coral recruits is difficult to apply to adult coral colonies. The latter may be less saturation dependent in their efficiency of using cations compared to juvenile corals.