Northeastern Section - 47th Annual Meeting (18–20 March 2012)

Paper No. 3
Presentation Time: 2:10 PM

MODELING MERCURY PARTITIONING AND BIOAVAILABILITY IN COASTAL SEDIMENTS


MASON, Robert P.1, HOLLWEG, Terill A.1 and GILMOUR, Cynthia C.2, (1)Department of Marine Sciences, University of Connecticut, 1080 Shennecossett Rd, Groton, CT 06340, (2)Smithsonian Environmental Research Center, 647 Contees Wharf Road, Edgewater, MD 21037, robert.mason@uconn.edu

The most important process linking mercury (Hg) inputs to the environment and the health risks of seafood consumption is the process wherebyinorganic Hg is converted to methylmercury (MeHg). Many factors influence the net reaction; Hg is methylated in the environment but MeHg is also demethylated. The factors controlling the extent of net Hg methylation in sediments are complex and many studies have found contrasting relationships between Hg methylation and ancillary parameters. We report here speciation modeling that was developed through comparison with observational data, mostly from the Chesapeake Bay and the mid-Atlantic continental shelf, to examine and better understand the biogeochemical factors that control Hg(II) bioavailability to methylating bacteria in sediments. This study supports the idea that Hg(II) partitioning between the solid-phase and pore water is strongly controlled by organic matter content (NOM) and dissolved inorganic sulfide concentration. Association of Hg with sulfide in the presence of NOM alters bioavailability through the formation of dissolved and nanoparticulate phases. We suggest that the fraction that was previously thought to represent dissolved HOHgSH0 actually represents a more complex suite of entities in the filtered fraction of both truly dissolved and nanoparticle Hg-S-NOM entities; other data also support this contention Our modeling incorporates the new information on the formation of nanoparticles in the presence of Hg, reduced S and NOM. Overall, measured MeHg production was related to the concentration of this so-called neutral fraction in the filtered solution, and the presentation will highlight the modeling results in light of our data and those of other recent studies, and suggest directions for future work.