HgCl2 addition to pore water samples from cold seeps can affect the geochemistry of dissolved inorganic carbon ([DIC], δ13CDIC)

The conventional use of mercuric chloride (HgCl2) as an antimicrobial agent in water samples for dissolved inorganic carbon (DIC) analysis poses health and environmental risks related to its handling and disposal. Even though there is an increasing interest in quantifying pore water DIC contribution to the ocean carbon cycle and deep-sea acidification, the paucity of comparative studies on marine pore waters prevents the modification of sampling protocols and HgCl2 still remains widely used. Here, we compared DIC concentrations and δ13CDIC composition in pore water samples from methane seepage areas in the Barents Sea and offshore N. Svalbard. Samples were extracted using 0.15 μm rhizon filters and split into two aliquots with 2–3 replicates each. Only one aliquot was treated with 10 μL saturated HgCl2(aq) and all samples were stored in the dark at 4 °C, prior to measurements ∼30 days later. The samples yielded extremely wide ranges of DIC concentrations, from 1.8 mM to 20.1 mM, and δ13CDIC values, from −36.0‰ to −1.6‰ (VPDB), due to variable contributions of methane-derived DIC to the pore water system. Overall, we obtained lower [DIC] (77% samples; N = 26) and heavier δ13C values (79% samples; N = 42) in Hg-treated samples. Isotopic and concentration differences larger than the uncertainties on the mean of replicated measurements (±0.2–0.5‰; ± 0.5 mM) and analytical precision (0.15‰; 0.71 mM) represent the 38% and 19% of the samples, with offsets of up to 7.4‰ and 1.9 mM, respectively. The largest offsets are in agreement with our CO2-degassing model, suggesting an interaction between mercuric chloride and dissolved hydrogen sulfide released by sulfate-driven methane oxidation. We therefore caution against further use of HgCl2 for DIC studies of marine pore waters from cold seeps.