Quantifying the effect of diagenetic recrystallization on the Mg isotopic composition of marine carbonates

Abstract

The Mg and Sr isotopic compositions (δ 26 Mg and 87 Sr/ 86 Sr) of pore fluids and bulk carbonates from Ocean Drilling Project Site 1171 (South Tasman Rise; 2148.2 m water depth) are reported, in order to evaluate the potential of diagenesis to alter carbonate-based geochemical proxies in an open marine system. Given the trace amounts of Mg in marine carbonates relative to coexisting pore fluids, diagenesis can alter carbonate δ 26 Mg, a promising proxy for seawater δ 26 Mg that may help elucidate long-term changes in the global Mg cycle. Constraints on the effect of diagenetic recrystallization on carbonate δ 26 Mg are therefore critical for accurate proxy interpretations. This study provides context for assessing the fidelity of geochemical proxyreconstructions using the primary components (i.e., foraminiferal tests and nannofossils) of bulk carbonate sediments.We find that pore fluid δ 26 Mg values (on the DSM3 scale) at Site 1171 increase systematically with depth (from -0.72‰ to -0.39‰ in the upper ~260 m), while the δ 26 Mg of bulk carbonates decrease systematically with depth (from -2.23‰ to -5.00‰ in the upper ~260 m). This variability is ascribed primarily to carbonate recrystallization, with a small proportion of the variability due to down-hole changes in nannofossil and foraminiferal species composition. The inferred effect of diagenesis on bulk carbonate δ 26 Mg correlates with downcore changes in Mg/Ca, Sr/Ca, Na/Ca, and 87 Sr/ 86 Sr. A depositional reactive-transport model is employed to validate the hypothesis that calcite recrystallization in this system can generate sizeable shifts in carbonate δ 26 Mg. Model fits to the data suggest a fractionation factor and a partition coefficient that are consistent with previous work, assuming calcite recrystallization rates of ≤7%/Ma constrained by Sr geochemistry. In addition, either partial dissolution or a distinctly different previous diagenetic regime must be invoked in order to explain aspects of the elemental chemistry and 87 Sr/ 86 Sr of relatively deep sediments from Holes A and C. This study indicates that the dynamics of a given sedimentary system can significantly alter bulk carbonate geochemistry, and presents a framework for considering the potential impact of such alteration on picked archives such as foraminiferal tests and nannofossils. Ultimately, this study contributes to the development of δ 26 Mg as a proxy for seawater δ 26 Mg by quantifying the susceptibility of carbonate δ 26 Mg to diagenetic alteration, particularly in sediments in open marine systems. This study suggests that because of the sensitivity of carbonate δ 26 Mg to diagenetic recrystallization, it can, in certain systems, be used to quantify the impact of diagenesis on carbonate-based geochemical proxies.

The role of marine sediment diagenesis in the modern oceanic magnesium cycle

“…2b). These findings are consistent with studies of net fractionation of magnesium isotopes during diagenesis 24,34 . A simple binned regression of sites (Fig.…”

“…All measured magnesium isotope analytical data are provided in the Supplementary Table 4. Additional pore water magnesium isotope values from ODP Sites 925, 984, 1012, 1082, 1086, 1171, and 1219, and IODP Sites U1378 and U1395 were obtained from published sources 24,34,63 . Other sites with available pore water magnesium isotope data are excluded that are either sparse near the sediment-water interface, affected by brine diffusion, or associated with unquantified pore water advection in the upper sediment section, including ODP Sites 807, 1003, 1052, 1196, 1265, and IODP Site U1403 17,63,64 .…”

“…Authigenic carbonates often form with relatively high magnesium contents due to the high Mg/Ca ratio and low inhibitory sulfate concentrations in the pore water, particularly deeper within the sulfate reduction zone or at the sulfate-methane transition zone 18–22 . In addition to precipitation of primary authigenic carbonate, magnesium may be depleted in pore waters during the recrystallization of low-magnesium biogenic carbonate to forms of carbonate that have a greater magnesium content than the original biogenic calcite 11,18,23,24 . Authigenic aluminosilicate formation is also a potentially important sink for magnesium in marine sediments, where it can be incorporated into the authigenic phase in a greater stoichiometric ratio than in the original primary silicate from which it formed, resulting in the release of calcium and other cations to pore waters 25 .…”

Variation in the diagenetic response of aragonite archives to hydrothermal alteration

“…Jonas et al (2017) demonstrated that, at elevated temperatures, rockbuffered alteration is generally diffusion-driven, with the host-rock dominating the petrographic and geochemical signature of the secondary phase. Many systems have shown to vary in their behavior between given proxies, with more fluid-buffered behavior for one (i.e., O) and more rock-buffered for another (i.e., C, Ca, or Mg; Pingitore, 1982;Geske et al, 2012;Chanda and Fantle, 2017), adding further complexity to the discussion. We suggest that internal fluids and organics can have a large effect on the diagenetic progression of carbonate archives, and play a much larger role in more internally rock-buffered systems, with increasing solid-to-fluid ratios at a local level.…”

Exploring the importance of authigenic clay formation in the global Li cycle

“…Before reconstructing secular trends in silicate weathering using any proxy archive, be it siliciclastic or carbonate, it is crucial to consider the effect of post-depositional (or postformational) alteration of the archive (i.e., diagenesis). Bulk carbonates have been shown to react and recrystallize relatively rapidly (Fantle and DePaolo, 2007;Fantle, 2015;Gorski and Fantle, 2017;Huber et al, 2017), impacting strontium, oxygen, and magnesium isotope compositions and Mg/Ca ratios (Richter and DePaolo, 1987;Schrag et al, 1992;Fantle and DePaolo, 2006;Chanda and Fantle, 2017). In contrast to elements such as Ca and Sr, and similar to Mg, Li has a low concentration in carbonates (1-2 ppm;Hoefs and Sywall, 1997;Burton and Vigier, 2012) compared to its concentration in siliciclastics (50-500 ppm; Hein et al, 1979;Zhang et al, 1998).…”