The influence of temperature and vital effects on magnesium isotope variability in Porites and Astrangia corals

Abstract

The magnesium isotope composition (δ26Mg) of sub-annually and bulk-sampled Porites corals is presented and suggests that the coral–seawater fractionation factor (∆26Mg) is temperature-dependent with a sensitivity of ~0.03‰°C−1. Although the apparent temperature dependence of ∆26Mg in Porites corals could be used empirically to reconstruct paleoclimate, its temperature sensitivity is about three times that of abiogenic aragonites, which indicates that kinetic and/or biologically-induced processes (collectively known as vital effects) are also important. Comparison with data from abiogenic aragonite precipitation experiments suggests that ∆26Mg vital effects vary seasonally in Porites. The ∆26Mg of skeleton deposited in winter is within the analytical error of abiogenic aragonite, but that of skeletal aragonite grown during summer exhibits significantly smaller fractionations that may be associated with light-enhanced calcification. The observed vital effect is consistent with <0.03% of coral Mg occurring as fluid inclusions, possibly associated with amorphous calcium carbonate (ACC). Other mechanisms such as surface entrapment and/or pH-induced changes in Mg speciation may also affect ∆26Mg, but cannot be evaluated with current data. However, the observed vital effect is not consistent with the isotope fractionation expected from Rayleigh processes, ion attachment/detachment kinetics or diffusion. Furthermore, the ∆26Mg of symbiotic and asymbiotic Astrangia corals is within analytical error, and indicates that vital effects cannot be attributed to the presence of photosymbionts. Some data suggest that the offset between coral and abiogenic aragonite ∆26Mg may be proportional to that for oxygen isotope fractionations, raising the possibility that vital effects in both isotope systems may be influenced by a common mechanism. Although the presence of vital effects and analytical limitations will complicate the development of a ∆26Mg-based paleothermometer, it may still be a valuable tool for reconstructing seawater δ26Mg or investigating biomineralization processes, especially when applied as part of a multi-proxy approach.