EXCERPT:
INTRODUCTION
Fluoride is a significant dissolved constituent in the oceans, having a concentration of 68 µM in seawater with salinity of 35 (e.g., Greenhalgh and Riley, 1961). Despite its high concentration, the large-scale geochemical cycling of F remains problematic (Rude and Aller, 1994). Rivers contribute nominally 16.8 x 1010 mol/yr of F to the oceans (Carpenter, 1969). Presumably, this input balances (or nearly balances) outputs to the seafloor. Various authors have identified and quantified a series of major F sinks, including authigenic carbonate fluorapatite (CFA), biogenic calcium carbonate, terrigenous aluminosilicate alteration, biogenic opal, and hydrothermal activity (e.g., Carpenter, 1969; Edmond et al., 1979; Froelich et al., 1983; Rude and Aller, 1994). However, these seafloor outputs amount to <6.5 x 1010 mol/yr of F, and some unidentified process seems to remove substantial F from the ocean (Rude and Aller, 1994). Enormous quantities of methane reside as dissolved gas, gas hydrates, and free gas in the pore space of marine sediment along continental margins (Kvenvolden, 1999; Dickens, 2001). Crucially, this CH4 is dynamic with clearly established (though poorly constrained) carbon fluxes between the ocean and sediment (Dickens, 2003). One ubiquitous flux is through anaerobic oxidation of methane (AOM). Above all methane-charged sediment sequences examined to date (e.g., Borowksi et al., 1999; D'Hondt et al., 2002), upward migrating CH4 reacts with downward diffusing SO42– at a sulfate–methane transition (SMT) to produce HCO3– and H2S (Fig. F1). The SMT can occur at or beneath the seafloor. Importantly, removal of SO42– and production of HCO3– leads to precipitation of authigenic carbonate minerals, especially Mg-rich phases such as high-Mg calcite (HMC), protodolomite, and dolomite (e.g., Ritger et al., 1987; Aloisi et al., 2000; Rodriguez et al., 2000; Greinert et al., 2001; Moore et al., 2004). The exact depth relationship between the SMT and the various authigenic Mg-rich carbonate minerals remains unclear and may vary from one location to another. Nonetheless, pore water profiles above methane-charged sediment systems invariably show concave-downward inflections in dissolved Ca2+ and dissolved Mg2+ (and inflections in alkalinity and Mg/Ca) consistent with Mg-rich carbonate formation near the SMT (Fig. F1). The overall process is intriguing to studies of the marine F cycle because F– complexes with dissolved Mg2+ in seawater (e.g., Rude and Aller, 1991) and Mg-rich carbonates preferentially incorporate F (Akaiwa and Aizawa, 1979; Rude and Aller, 1991). Rude and Aller (1991) show that dissolution of biogenic Mg-rich carbonate releases F– to surrounding pore waters. To our knowledge, however, uptake of F– into authigenic Mg-rich carbonates has not been considered, excepting a brief note by Schulz et al. (1994). In this study, we measure F– concentrations of pore waters collected at nine drill sites that contain abundant methane, an SMT in shallow (<20 meters below seafloor [mbsf]) sediment, and evidence for methane-driven authigenic Mg-rich carbonate precipitation (D'Hondt, Jørgensen, Miller, et al., 2003; Tréhu, Bohrmann, Rack, Torres, et al., 2003). Sediments clearly remove F– from pore water near the SMT, although additional work is needed to confirm that the F– is entering authigenic Mg-rich carbonate. Fluoride and methane gas
Tréhu, A.M., Bohrmann, G., Torres, M.E., and Colwell, F.S. (Eds.)
Proceedings of the Ocean Drilling Program, Scientific Results Volume 204
16. DISSOLVED FLUORIDE
CONCENTRATIONS IN METHANE-CHARGED
SEDIMENT SEQUENCES1
Gerald R. Dickens,2 Catherine M. Donohue,2 and Glen T. Snyder2
ABSTRACT
Dissolved fluoride was determined for pore waters at eight sites
drilled on Hydrate Ridge during Ocean Drilling Program (ODP) Leg 204
and one site drilled in the Peru Trench during ODP Leg 201. All nine
sites contain a shallow (<20 m) sulfate–methane transition (SMT) above abundant methane including gas hydrate. For Sites 1248, 1249, and 1250 on the crest of Hydrate Ridge, F– concentrations are significantly lower than that of seawater in the shallowest samples (<50 µM), rise to a broad maximum, and generally decrease with depth. The low values at the top are consistent with rapid F– removal at or near the seafloor, and the relatively smooth F– profiles are consistent with high upward fluid fluxes. In contrast, Sites 1244, 1245, 1247, 1251, and 1252 from the flanks and slope basins of Hydrate Ridge and Site 1230 from the Peru Trench have F– profiles apparently characterized by two lows with an intervening high. Processes involving sediment components appear to consume F– at shallow depth, release F– at intermediate depth, and consume F– again at deeper depth. The upper low in F– concentrations consistently lies near the SMT where pore water alkalinity and Mg2+ profiles suggest precipitation of Mg-rich carbonate. A similar pattern occurs at other sites drilled into methane-charged sediment. We speculate that Mg-rich carbonates (e.g., high-Mg calcite, protodolomite, and dolomite) remove F– from pore water near the SMT but, with burial and recrystallization, return F– to pore waters at depth. Authigenic Mg-rich carbonates conceivably represent a major sink of F– from the ocean, although additional work is needed to confirm this idea. 1Dickens, G.R., Donohue, C.M., and Snyder, G.T., 2006. Dissolved fluoride concentrations in methane-charged sediment sequences. In Tréhu, A.M., Bohrmann, G., Torres, M.E., and Colwell, F.S. (Eds.), Proc. ODP, Sci. Results, 204, 1–22 [Online]. Available from World Wide Web:
[Cited YYYY-MM-DD]
2Department of Earth Sciences, Rice
University, Houston TX 77005, USA.
Correspondence author:
jerry@rice.edu
Initial receipt: 23 March 2005
Acceptance: 8 March 2006
Web publication: 29 September 2006
Ms 204SR-118
