Application of calcite Mg partitioning functions to the reconstruction of paleocean Mg/Ca

Calcite Mg/Ca is usually assumed to vary linearly with solution Mg/Ca, that a constant partition coefficient describes the relationship between these two ratios. Numerous published empirical datasets suggests that this relationship is better described by a power function. We provide a compilation of these literature data for biotic and abiotic calcite in the form of Calcite Mg/Ca = F(Solution Mg/Ca)H, where F and H are empirically determined fitting parameters describing the slope and deviation from linearity, respectively, of the function. This is equivalent to Freundlich sorption behavior controlling Mg incorporation in calcite. Using a power function, instead of a partition coefficient, lowers Phanerozoic seawater Mg/Ca estimates based on echinoderm skeletal material by, on average, 0.5 mol/mol from previous estimates.These functions can also be used to model the primary skeletal calcite Mg/Ca of numerous calcite phases through geologic time. Such modeling suggests that the Mg/Ca of all calcite precipitated from seawater has varied through the Phanerozoic in response to changing seawater Mg/Ca and that the overall range in Mg/Ca measured among various calcite phases would be greatest when seawater Mg/Ca was also high (e.g., “aragonite seas”) and lowest when seawater Mg/Ca was low (e.g., “calcite seas”). It follows that, during times of “calcite seas” when the seawater Mg/Ca is presumed to have been lower, deposition of calcite with low Mg contents would have resulted in a depressed drive for diagenetic stabilization of shelfal carbonate and, in turn, lead to greater preservation of crystal and skeletal microfabrics and primary chemistries in biotic and abiotic calcites.