1d-2: Mercury cycling, bioaccumulation and health impacts in polar regions

The Antarctic plateau ice-covered area of 7 million km2 is a highly photochemically active area during the sunlit period with oxidant concentrations approaching those of tropical or urban mid-latitude environments. Earlier studies carried out on the high-altitude Antarctic plateau with modern instruments also suggested, based on short-term observations (a few weeks) in summer, an intense reactivity of mercury on the plateau at the airsnow interface.

Recent monitoring studies of atmospheric gaseous mercury Hg(0) at Concordia Station revealed the occurrence of multi-day to weeklong depletion events of Hg(0) in ambient air in summer, that are not associated with depletion of O3, and likely result from a stagnation of air masses on the plateau triggering an accumulation of oxidants in the shallow boundary layer. This behavior is radically different from what is usually observed in the Arctic where only mercury depletion events that were associated with O3 depletion (and with a Hg(0)/O3 correlation) have been highlighted so far. Such reactivity was also observed at Dome A, and between Concordia Station and Vostok Station.

According to observations at coastal (or near-coastal) Antarctic stations (DDU, Troll, Neumayer, Terra Nova), the reactivity observed on the plateau in spring and summer is transported on a continental scale by strong katabatic winds. This is well demonstrated by looking at data set of east Antarctica. Hg(II) species can be thus rapidly brought to coastal ecosystems and could provide a source of Hg to coastal and sea-ice organisms. This may explain higher values of Hg that are retrieved in bioindicators, in sea-ice and surface waters. The recent identification of microbial methylation of Hg in Antarctic sea-ice could exacerbate the influence of this atmospheric source for introducing Hg into Southern Ocean ecosystems.

This presentation will give an overview of the functioning of the Hg chemistry in central Antarctica and its potential influence on coastal environments.

Surface snow samples were collected over three sampling campaigns during the Austral summer from the clean sampling area upwind of Concordia station on the Antarctic plateau. During the first campaign a snow pit was also dug to gain an idea of the stability and temporal profile of mercury in the snow pack in the high Antarctic. In the last campaign (Austral summer 2015-16) a 72 hour experiment was carried out that collected surface snow and precipitation on an hourly basis for comparison with atmospheric measurements to evaluate surface recycling in connection with diurnal solar radiation changes.

Sample analysis was carried out for 202Hg and 79Br using a Thermo Electron Element 2 sector field ICP-MS and an Agilent Technologies 7500cx ICP-MS. Samples were left unacidified for Br, I, Ca, K, Na, and were acidified with 100 µL of ultrapure HCl before mercury analysis.

We conducted three sampling campaigns at Dome C, in the austral summers of 2013-2014, 2014-2015, and 2015-2016, during which snow and atmospheric mercury concentrations (TGM) we measured. The three campaigns had different metereological conditions that heavily affected mercury deposition processes. In the absence of snow deposition (including diamond dust) the surface mercury and atmospheric concentration remained stable without any particular trend. During and after snow deposition, mercury concentrations undergo marked changes. In the 2015-2016 field campaign, several snow deposition events occurred strongly affecting the surface snow Hg abundance. Fresh snow deposits were collected from a bench elevated above ground level to reduce wind blown snow collection, and the concentrations detected (around 100 pg g-1) suggest/confirm the capacity of snow as a mercury atmospheric scavenger. These high concentrations decreased rapidly to 10-20 pg g-1 on average within days of being deposited, influencing the atmospheric TGM concentrations. Apart from this variation, during the high resolution experiment, no diurnal variations in surface snow concentrations ware seen.

Preliminary results from this sampling campaign give us important indications for evaluating mercury cycling on the central Antarctic plateau. The connection between Br and Hg in snow does not show any particular correlation or trend during the high resolution experiment (72h). The connection is more pronounced over time. We found a correlation of R2=0.59 in the snow pit samples. Higher deposition of mercury in surface snow seems connected with an increase in Br suggesting an intrusion of humid air from the coast bringing reactive Br that interacted with mercury.

In cooperation with the Long Term Ecological Research (LTER) project along the West Antarctic Peninsula (WAP), this study aims to characterize mercury (Hg) speciation in this region's continental shelf waters and Hg accumulation in the WAP food web. Seawater was sampled using a towfish surface sampler and a trace metal clean rosette at depths of 2 m to 2000 m at northern (64.0-64.9°S), mid-latitude (67.1-68.1°S), and southern (~69°S) stations. In surface waters, total dissolved Hg (THg) and dissolved elemental mercury (DEM) were highest at northern and mid-latitude nearshore stations, but THg at depth was generally higher at mid-latitude and southern stations. Profiles of dissolved total methylated mercury (MeHgT) revealed subsurface maxima in offshore (230 m, 0.61 pM) and coastal (500 m, 0.92 pM) waters at northern stations, while MeHgT at mid-latitude and southern stations (~69°S) was generally lower. At the southern shelf break, MeHgT was highest in 2°C, low O2, Upper Circumpolar Deep Water (UCDW). High unfiltered MeHgT in 1 m sea ice sub-cores relative to surface seawater reveal that sea ice may be an important source of Hg in WAP surface waters for grazers such as the Antarctic Krill Euphausia superba. Accumulation of THg and monomethylmercury (MMHg) was higher in juvenile (7.8 to 20 ng g-1 and 0.7 to 2.9 ng g-1, resp.) than adult (4 to 13 ng g-1 and 0.3 to 1.6 ng g-1, resp.) krill which feed near the coast and under sea ice during overwintering. Nearshore or under sea ice feeding may support MMHg accumulation in E. superba, which was higher in coastal than offshore krill at northern stations over multiple years. Trophic enrichment (~10^2) of mercury was observed between E. superba and feathers of all three Pygoscelis penguins, Adélie (Pygoscelis adeliae), gentoo (P. papua), and chinstrap (P. antarctica) living on neighboring islands to U.S. Antarctic research station Palmer. The mean concentration of Hg in chinstrap (0.77 mg kg-1) was higher than in gentoo (0.15 mg kg-1) or Adélie (0.09 mg kg-1) penguins, likely attributed to higher consumption of krill, specifically juveniles with higher MMHg, suggesting that MMHg may be used as a tracer of food web connectivity in this highly productive, remote marine ecosystem.

Leadership and community members in the Dehcho region of the Northwest Territories, Canada, are concerned about mercury concentrations ([Hg]) in food fishes such as Northern Pike (Esox lucius), Walleye (Sander vitreus), and Lake Whitefish (Coregonus clupeaformis). Mercury levels in food fishes vary widely across lakes in a relatively small geographic area, with some lakes having high fish [Hg] and associated consumption advisories. Previous research has been unable to elucidate the main drivers of among-lake differences in fish [Hg] in this region. From 2013-2015, 8 remote Dehcho lakes were sampled for fish, benthic invertebrates, zooplankton, sediment, and water by a collaborative team that included First Nations and academic researchers. Fish mercury concentrations were related to fish stable isotope ratios, age, size, and growth rates, as well as to a suite of water chemistry and sediment variables. Interim results indicate that size-standardized differences in fish [Hg] among lakes were best explained by concentrations of chlorophyll-a for Walleye (R2=0.9), and by dissolved organic carbon concentration (R2=0.6) for Lake Whitefish. For Northern Pike, size-standardized differences in [Hg] were best explained by lake chloride concentrations and age-at-size (R2=0.93). Intercepts of Hg-del15N relationships for each lake were negatively related to pH, and positively related to concentrations of dissolved methyl mercury in water. These results are discussed in the context of causal mechanisms, in-lake versus catchment controls, consumption advice for northern fishers, and a resulting mercury mitigation strategy initiated by the Dehcho First Nations.

Arctic marine ecosystems are threatened by new risks of Hg contamination under the combined effects of climate change and human activities. Rapid change of the cryosphere might for instance release large amounts of Hg trapped in sea-ice, permafrost and terrestrial glaciers over the last decades. Sea-ice disappearance is opening new shipping areas to polluting human industries. The general warming of ocean water masses is expected to affect the cycle of Hg, thereby increasing exposure of marine organisms. Hence, Hg could have high impacts on Arctic organisms, biodiversity and ecosystems and is still a source of major environmental concerns. In that context, providing a large-scale and comprehensive understanding of the Arctic marine food-web contamination is essential tobetter apprehend impacts of anthropogenic activities and climate change on the exposure of Arctic species and humans to Hg. In 2015, an international sampling network (ARCTOX) has been established, allowing the collection seabird samples all around the Arctic. Seabirds are indeed good indicators of Hg contamination of marine food webs at large spatial scale. Gathering researchers from 10 countries, ARCTOX allowed the collection of >5000 samples from twelve seabird species at >40 Arctic sites in 2015 and 2016. These different species have different trophic ecologies (diets and habitats) and will therefore provide information on Hg contamination for the different compartments of Arctic marine ecosystems (i.e benthic, pelagic, epontic, coastal, oceanic). By relying on this new network and by combining Hg analyses with biotelemetry, we aim at (1) monitoring spatio-temporal variations of Hg in Arctic biota. (2) Defining Arctic hotspots of Hg contamination and highlighting sensitive areas that require particular attention and protection. (3) Identifying non-Arctic sources of Hg contamination for migratory Arctic predators.

We investigated the influence of climate variability on total mercury concentrations (THg) in eggs of thick-billed murres and northern fulmars from Prince Leopold Island in the Canadian High Arctic. Climate variables were compiled from a variety of ground-based and satellite-derived sources, and included climate data: air temperature, wind speed, sea level pressure, total precipitation, snowfall, rainfall, Arctic Oscillation Index (AO) and North Atlantic Oscillation (NAO) indices; and oceanic data: sea surface temperature, primary productivity, and sea ice concentrations. Mercury monitoring data for the seabird eggs spanned 40 years (1975-2014); and climate influence time lags of 0 to 10 years were evaluated. The main effects of climate variables on THg concentrations were assessed for each time lag using General Linear Models. Models with the most parsimonious fit for both species of seabirds were selected using Akaikes Information Criteria (AIC). The most parsimonious models for northern fulmars included climate variables such as AO, precipitation, and temperature and, for the thick-billed murres, the most parsimonious models included sea ice concentration, precipitation and temperature. Truncated data sets for the past 10 years, when THg concentrations were relatively constant (2005-2014), were reasonably correlated with some climate variables such as precipitation. Overall, the results suggest a small, but significant effect of climate variables on THg concentrations in Arctic seabirds.

The presence of mercury (Hg), a transboundary pollutant, is increasing in Arctic region and the Arctic sea may take short time to reflect changes in atmospheric Hg levels. The geochronological and pollution analysis of sedimentary deposits of Arctic fjords can indicate the deposition profile of mercury in this region. The sediment deposit can also explain the environmental changes occurred in the past and will help to understand the present situation as it is influenced by the changes in the glacial inputs. The recent studies have indicated the retreat of glaciers due to climate change and subsequent increase in melt water outflow with high content of sediments. The present study thus focused on the monitoring of mercury content and deposition profile in the sedimentary environment of the Kongsfjorden, an Arctic Fjord in Svalbard. The surface and core sediment samples were collected during Indian Arctic expedition 2014-15 and 2015-16. Mercury concentration was detected by using Cold Vapour Atomic Florescence Spectrophotometer and Direct mercury analyser. Mercury fractionation was also carried out for selected samples. The sedimentation rate in inner Kongsfjorden, Ny-lesund, Svalbard using 210Pb/210Po dating technique was also done. The sedimentation rate ranged between 0.22-0.37 cm/year during the last 112 years. The average sedimentation rate obtained was 0.28 cm/y. The rate has been increased during the last 20 years and it might be due to the increased influx of melt water due to climate change. High variation in mercury deposition was observed. The results showed that mean concentration of THg in sediment was 0.198mg/kg dry weight. The results of fractionation indicated the mobility and bioavailability of mercury in the Kongsfjord sediments. The highest percentage of Hg was in fourth fraction (F4) followed by initial fractions (F2>F1>F3) and final fraction (F5). The high Hg concentration in the initial fractions indicated the availability of mercury for chemical and biological transformations and transport in the Fjord. Hence potential toxic effects are possible in the system.

Remote regions such as the Antarctic have become increasingly important for investigations into far-reaching anthropogenic impacts on the environment, most recently in regard to the global mercury (Hg) cycle. Pygoscelis penguins are ideal model organisms to track Hg through the Antarctic marine food web as they are long-lived, broadly distributed, and are susceptible to biomagnification due to foraging at relatively high trophic levels. However, using these species as biomonitors requires a solid understanding of the degree of species-specific variation in foraging behaviors act to mediate their dietary exposure to Hg. We combined stomach content analysis along with Hg and stable isotope analyses of eggshell membrane, blood, feathers and common prey items to help explain inter and intra-specific patterns of dietary Hg exposure. Breeding colonies were sampled from four of the major regions of the Antarctic Peninsula to address spatial variation in Hg exposure and diet. In addition, a long-term data set from a single colony where all three species breed sympatrically was used to examine inter-annual variation in Hg and diet. Hg concentrations did not vary significantly among regions of the Antarctic Peninsula and remained fairly consistent over the eight year time period investigated. Hg concentrations did differ significantly among species; Hg concentrations in Chinstrap (Pygoscelis antarctica) penguins were significantly higher than concentrations in Gentoo (P. papua) and Adeli (P. adeliae) penguins. Stable isotope analysis revealed diets of all three species to be dominated by Antarctic krill (Euphausia superba) and to a lesser extent fish. Integrating isotopic approaches with stomach content analysis allowed us to identify species-specific preferences for prey fish insufficiently explained by stable isotope proxies for trophic level. Chinstrap penguins were found to forage on a larger proportion of higher Hg mesopelagic prey fish relative to their congeners targeting epipelagic or benthic prey species. While the potential for adverse effects due to Hg exposure is currently low in Pygoscelis penguins in the Antarctic Peninsula, this approach allowed us to substantiate for the first time, the relatively higher risk of Hg exposure for Chinstrap penguins.