Article pubs.acs.org/est
Mercury Stable Isotopes in Ornithogenic Deposits As Tracers of Historical Cycling of Mercury in Ross Sea, Antarctica Wang Zheng,† Zhouqing Xie,*,‡ and Bridget A. Bergquist*,† †
Department of Earth Sciences, University of Toronto, 22 Russell Street, Toronto, Ontario M5S 3B1, Canada Institute of Polar Environment, School of Earth and Space Sciences, University of Science and Technology of China, Hefei, Anhui 230026, China
‡
S Supporting Information *
ABSTRACT: Production of methylmercury (MeHg) in ocean waters and its bioaccumulation in marine organisms are critical processes controlling the fate and toxicity of mercury (Hg). However, these processes are not well understood in the Antarctic, where high levels of MeHg are observed in the subsurface ocean (100−1000 m). We explored the use of Hg stable isotope compositions in historical and modern biological deposits as a new approach for discerning Hg sources and tracing MeHg cycling in the ocean and bioaccumulation in marine biota. We found similar mass independent isotope fractionation (MIF) of Hg between a sediment profile containing historical penguin and seal feces deposits from coastal Antarctica and modern penguin and seal feces, suggesting that penguin and seal feces were the dominant sources of Hg to the sediments at different time periods. Furthermore, sediments dominated by seal feces displayed a significantly lower MIF slope (Δ199Hg/Δ201Hg) than those dominated by penguin feces despite similar extents of MIF. Since seals forage at greater depths (>400 m) than penguins (70%).45 Two seal feces samples were also collected from Byers peninsula (62°38′00″ S, 61°05′00″ W) and Zhongshan Station (69°22′24″ S, 76°22′40″ E), respectively (SI Figure S1). The seal species are identified to be Southern elephant seal (Mirounga leonine) in both locations. The locations and species of modern and ancient animals are different, and how these differences affect our interpretation will be discussed in “Results and Discussion”. The sediment core was directly sectioned in the field at 0.6 cm intervals and stored at −20 °C. Before analysis, sediment subsamples were air-dried and grinded after removal of large rock fragments and visible biological remains such as bone pieces, seal hairs, eggshells and feathers. The final powder samples were passed through a 74 μm mesh sieve. Fresh feces samples were freeze-dried and prepared similarly as the sediment. Mercury Concentration. Total Hg (THg) in powdered sediments and fresh feces was extracted by modified aqua regia digestion based on previously described methods (details in SI Method S2).46 THg concentration was measured by cold vapor atomic fluorescence spectroscopy (CVAFS, Tekran 2600). Digestion recovery and reproducibility were checked against an in-house Hg standard (JTBaker) and two procedural standards, NIST SRM 1944 (marine sediment reference material) and NIST SRM 1646a (estuarine sediment reference material). The average recoveries are 100.9 ± 1.3% (2SD, n = 4) for JTBaker Hg standard, 103.0 ± 19.0% (2SD, n = 6) for NIST 1944 and 107.7 ± 7.1% (2SD, n = 5) for NIST 1646a. Mercury Isotope Measurements. Mercury in digested samples was purified by transferring Hg into ∼20 mL 0.2% KMnO4 (w/w) with 5% H2SO4 (v/v) trapping solution using a purge and trap method (SI Method S2). The mean recovery of Hg during purification for sediments and feces was 96.8 ± 12.0% (2SD, n = 47), and for digestion standard and reference
■
MATERIALS AND METHODS Site Description and Sample Collection. The sediment core was collected from a currently dried-out meltwater pond in an active Adélie penguin (Pygoscelis adeliae) colony on the north side of Cape Bird (77°12′47.5″ S, 166°26′44.4″ E), Ross Island, on the east of the southern Ross Sea, Antarctica. More B
DOI: 10.1021/acs.est.5b00523 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology
Figure 1. Mercury isotope compositions, (a) δ202Hg, (b) Δ199Hg, (c) total Hg (THg), and (d) Hg flux in the sediment along the depth. The Roman numbers represent different periods.
materials was 98.0 ± 4.4% (2SD, n = 11). Total procedural blanks including digestion and purification were
