Elucidating the controls on the Mg isotopic composition of marine pore fluids

The Sr and Mg isotopic composition of pore fluids and carbonates from the Neogene section of Ocean Drilling Project Site 806B are reported (⁸⁷Sr/⁸⁶Sr and δ²⁶Mg, measured using a Thermo Scientific Neptune Plus multi-collector ICP-MS). Site 806B, located on the northern margin of the Ontong Java Plateau, hosts a thick (776 m cored, depth to basement ~ 1200 m), relatively continuous, carbonate-rich section (between 83 and 96% CaCO₃). Our goal in the current study is to use the Sr and Mg isotope data of pore fluids and carbonates to address open questions regarding (1) the extent to which the pore fluid chemistry is overprinted by calcite recrystallization, (2) the effects of diagenesis on bulk carbonate chemistry, and (3) the likelihood of preserving secular seawater δ²⁶Mg trends in pore fluids. Accordingly, the current study compares and contrasts the isotopic and elemental data between adjacent ODP Sites 806B and 807A, which have similar depositional histories, carbonate contents, and pore fluid chemistries. The measured ⁸⁷Sr/⁸⁶Sr ratios of pore fluids at 806B range smoothly from 0.70914 at 4.45 mbsf to 0.70851 at 509.3 mbsf, similar (though offset relative) to the bulk carbonate trend (0.70918 to 0.70877 between 1.11 and 501.94 mbsf). The δ²⁶Mg_DSM3 of 806B pore fluids generally increases from -0.86‰ at 4.45 mbsf to -0.17‰ at 679.0 mbsf. The overall trend is consistent with previously collected δ²⁶Mg data at 807A [1]; there is, however, a significant difference in pore fluid δ²⁶Mg between the two sites at depths of 300 to 600 mbsf. At these depths, 806B pore fluid δ²⁶Mg values are +0.2 to 0.3‰ relative to 807A at similar depths [1]. The application of a depositional reactive transport model to the Sr isotope data suggests that bulk carbonate recrystallization rates at 806B are similar to those at 807A (<2%/Ma) [2]. An iterative model construct is employed to evaluate the dominant controls on the δ²⁶Mg of marine pore fluids; specifically the relative influence of carbonate recrystallization, diffusion, and the upper and lower boundary conditions. The numerical model is also used to investigate specific hypotheses regarding the influence of a depth-variable fractionation factor, and variable growth histories of the sedimentary column (the latter is suggested by the differences in δ²⁶Mg trends at depth). Preliminary model results suggest that diffusive communication with an isotopically distinct lower boundary does not explain the features of the δ²⁶Mg data, but that a depth variable fractionation factor during sediment-fluid exchange can. Ultimately, the objective of understanding the controls on pore fluid δ²⁶Mg is twofold: (1) to provide accurate constraints on the leverage to alter isotopic composition during diagenesis, and (2) to evaluate the extent to which secular changes in seawater influence, or can be deconvolved, from pore fluid δ²⁶Mg. References: [1] Higgins and Schrag, 2012. Earth Planet. Sci. Lett., 357-358, 386-396 [2] Fantle and DePaolo, 2006. Geochim. Cosmochim. Acta. 70, 3883-3904