Total and Methylated Mercury in the Beaufort Sea: The Role of Local

Oct 1, 2012 - Total and Methylated Mercury in Arctic Multiyear Sea Ice ... Mercury in the Black Sea: New Insights From Measurements and Numerical ...
5 downloads 0 Views 403KB Size
Article pubs.acs.org/est

Total and Methylated Mercury in the Beaufort Sea: The Role of Local and Recent Organic Remineralization Feiyue Wang,†,‡,* Robie W. Macdonald,†,§ Debbie A. Armstrong,† and Gary A. Stern†,⊥ †

Center for Earth Observation Science, Department of Environment and Geography, University of Manitoba, Winnipeg, MB, Canada R3T 2N2 ‡ Department of Chemistry, University of Manitoba, Winnipeg, MB, Canada R3T 2N2 § Institute of Ocean Sciences, Department of Fisheries and Oceans, Sidney, BC, Canada V8L 4B2 ⊥ Freshwater Institute, Department of Fisheries and Oceans, Winnipeg, MB, Canada R3T 2N6 S Supporting Information *

ABSTRACT: Mercury is a major contaminant in the Arctic marine ecosystem. While extensive studies have been conducted on mercury in the Arctic’s atmosphere and biota, far less is known about the distribution and dynamics of mercury species in the Arctic Ocean. Here, we present vertical profiles for total mercury (HgT) and total methylated mercury (MeHgT, sum of monomethylmercury and dimethylmercury) from the Beaufort Sea of the Arctic Ocean at locations with differing sea ice conditions. The concentration of HgT ranged from 0.40 to 2.9 pM, with a surface enrichment that can be attributed to a combination of sea ice-modified atmospheric deposition and riverine input. The concentration of MeHgT ranged from 80%) found in higher trophic levels in the foodweb. In freshwater and estuarine systems, MMHg in the water column is thought to originate primarily from sediments where the oxic−anoxic transition zone favors microbial Hg methylation process.12,13 In contrast, the discovery of a subsurface maximum of MMHg in seawater,14 and its apparent association with nutrient maxima in the Pacific Ocean,15,16 Southern Ocean,9 and Mediterranean Sea17,18 suggests that Hg methylation occurs within oxic ocean domains where organic matter (OM) is undergoing remineralization. Relatively high MMHg and dimethylmercury (DMHg) concentrations have also been reported in deep waters in Hudson Bay and the Canadian Arctic Archipelago (CAA).19,20 However, low vertical resolution in these studies precludes a determination of the role OM remineralization may 11822

dx.doi.org/10.1021/es302882d | Environ. Sci. Technol. 2012, 46, 11821−11828

Environmental Science & Technology

Article

Table 1. Details about the Sampling Stations and the Associated MeHgT Maximum latitude (deg)

longitude (deg)

water depth (m)

date of sampling

ice condition

D29 D33 1208 421 6006 2010

71.0385 71.0650 71.0717 71.4700 72.6630 75.1223

−123.9103 −121.7867 −126.0767 −133.9063 −128.3783 −120.4380

327 188 402 1184 224 424

03/10/2008 03/27/2008 06/28/2008 06/30/2008 07/03/2008 07/05/2008

D34

71.0772

−121.8200

182

07/13/2008

ice covered ice covered open water open water open water ice covered; melting ice covered

station

a

a

depth of PML (m)

depth of MeHgT maximum (m)

peak [MeHgT] (pM)

[HgT] at MeHgT maximum (pM)

[MeHgT]/[HgT] at the MeHgT maximum

40 40 45 52 54 50

140 50−140 106 140 54 106

0.451 0.098 0.140 0.418 0.284 0.590

0.922 0.966 0.673 1.072 0.897 1.545

0.49 0.10 0.21 0.39 0.32 0.38

41

160

0.314

b

“D” denotes a drift ice station. No data.



b

MATERIALS AND METHODS Study Area. The Beaufort Sea in the western Arctic Ocean is predominantly influenced by water from the Pacific Ocean entering through Bering Strait (Figure 1). The surface circulation (ice and water) in the Beaufort Sea, where we conducted our measurements, is dominated by westward drift associated with the Beaufort Gyre. At depth, the Beaufort Undercurrent transports water eastward along the continental slope. 23 Sea ice forms in around mid-October and begins to melt in late June leaving predominantly open water by late September (Figure S1 of the Supporting Information). The water masses in the region comprise three layers:24−26 the surface 40−50 m, the polar mixed layer (PML) forms in winter when the surface is cooled to freezing and brine is added from sea-ice growth. In summer, this layer becomes stratified by the combined addition of runoff and sea-ice melt. Below the PML is a cold Pacific halocline (∼ 50−200 m), which is formed over the Chukchi Shelf from Pacific water entering the Arctic Ocean. This water is associated with a nutrient maximum and oxygen minimum centered around a salinity of 33.1. Below this halocline, water of Atlantic origin prevails, first as a second, deeper component of the cold halocline formed in the Barents Sea, and beneath this as the Atlantic Layer, recognized by temperatures above 0 °C. A system of midshelf flaw leads frequently expands into the Cape Bathurst Polynya throughout winter and spring producing an important location for physical and biological processes.22 The Mackenzie River provides an important regional source of fresh water to the shelf, especially between mid-May and June. During the summer months, the upper 20−30 m of the water column becomes strongly stratified due to freshening by sea ice melt and inflow from the Mackenzie River.27 Sampling and Analysis of Total and Methylated Mercury. Seawater profile samples for HgT and MeHgT were collected from 7 stations from March 3 to July 7, 2008 (Figure 1 and Table 1). These stations varied in water depth (182 to 1184 m) and sea-ice conditions (ice covered to open water) at the time of sampling. Two of the stations, D33 (March 27) and D34 (July 13), were close to one another but 3.5 months apart; both stations were covered with drifting ice at the time of sampling. Water samples were collected between a depth of 10 m and the bottom onboard the Canadian Research Icebreaker CCGS Amundsen using 20 L Niskin bottles mounted on a General Oceanics 24-bottle rosette equipped with a SeaBird conductivity−temperature-depth (CTD) sensor. Rosette bottles used to collect samples for HgT and MeHgT were tested at the Class 100 cleanroom, Portable In situ Laboratory for Mercury

Speciation (PILMS) onboard the icebreaker for acceptable background Hg concentrations. Surface seawater samples (0− 10 m) were collected manually using a precleaned Teflon Niskin bottle (General Oceanics) from a zodiac for open water stations or through an augered hole (50−100 cm diameter) for ice-covered stations. The Niskin bottle was precleaned with acid at PILMS and tested for acceptable background Hg concentrations. Sampling and sample handling for HgT and MeHgT followed the clean-hands-dirty-hands protocol.28 Duplicate HgT samples were collected, without filtration, into 50 mL new polypropylene tubes (BD Falcon) and preserved with 0.5% (v:v) ultrapure HCl (CMOS grade, JT Baker) immediately after sampling. Extensive testing in our lab shows that new Falcon tubes can be used for ultratrace sampling of HgT provided bottles are not reused; additionally, each new batch of tubes was randomly tested for acceptable HgT background before use. Samples for MeHgT were collected in 50 mL hot acid-cleaned Teflon bottles (tested for acceptable Hg background before use) and preserved with 0.2% H2SO4 (OmniTrace Ultra High Purity, EMD). Under such acidic conditions, DMHg is known to break down to MMHg;29 therefore, MeHgT reported here represents the sum of MMHg and DMHg in unfiltered seawater. A recent study has shown that MMHg and DMHg are present at similar concentrations in seawater from the Canadian Arctic Ocean.20 Field blanks were taken whenever HgT and MeHgT samples were collected. The samples were double bagged and stored in the dark at 4 °C. Surface water was sampled near the mouths of the Mackenzie and Horton Rivers (Stations MK and HT, respectively; Figure 1). Unfiltered samples were collected from the edge of the rivers using a helicopter to reach the mouth of the river. HgT and MeHgT were analyzed at the PILMS usually within 48 h of sampling. To our knowledge, this is the first study in which HgT and MeHgT were both analyzed onboard the ship in a cleanroom laboratory shortly after sampling. HgT was analyzed using cold vapor atomic fluorescence spectroscopy (CVAFS) on a Model 2600 Hg analyzer (Tekran) following U.S. EPA 1631.30 The samples were spiked with 0.5% BrCl (freshly prepared from KBr and KBrO3, JT Baker) for at least 6 h to destroy OM before being neutralized with NH2OH·HCl (Fluka) and reduced with SnCl2 (JT Baker). The detection limit (DL), determined as three times the standard deviation of analyses of seven replicate samples of a laboratory blank, was 0.25 pM. Certified reference waters BCR579 (coastal seawater; [HgT] = 9.47 pmol kg−1, Institute for Reference Materials and Measurements, Belgium) and ORMS-3 (river water; [HgT] = 62.8 pmol kg−1, National Research Council Canada) were 11823

dx.doi.org/10.1021/es302882d | Environ. Sci. Technol. 2012, 46, 11821−11828

Environmental Science & Technology

Article

Figure 2. Depth profiles of total mercury (HgT) and methylated mercury (MeHgT), along with salinity, temperature, δ18O, dissolved phosphate (PO43‑T), dissolved oxygen (DO), and dissolved inorganic carbon (DIC), in the water column of the Beaufort Sea.

Ancillary Data. Temperature, salinity, and density were obtained from the CTD instrument mounted on the rosette. Dissolved oxygen (DO) was measured by an oxygen probe attached to the CTD, which was frequently calibrated against Winkler titrations. The concentrations of nutrients (dissolved phosphate, nitrate, silica) and dissolved inorganic carbon (DIC) were measured on samples from the rosette bottles using sampling and analytical techniques described elsewhere.26 Samples of δ18O analysis were collected in 20 mL glass scintillation vials and analyzed at the University of Ottawa.

analyzed following the same procedures to ensure data quality. [HgT] in field blanks was always