Mercury in the Canadian Arctic: a review of aquatic mercury conditions
A review of current knowledge on mercury concentrations in the Canadian arctic, including effects of anthropogenic (human) activity, biotic changes, and remediation efforts
Beginning in the late 1800s, concentrations of mercury in the tissues of many fish and shellfish species eaten by humans began to rise significantly (Chételat et al., in press), to the point where today the U.S. Environmental Protection Agency advises against the consumption of shark, swordfish, tilefish, and mackerel because they contain potentially toxic levels of mercury (U.S. EPA 2014). Similar increases have been observed in non-food fish species (Scheuhammer et al. 2014), high-trophic-level organisms (Dietz et al. 2013), and many aquatic birds (Braune et al. 2014). These increases, while geographically variable (Tran et al., in press), are not geographically restricted: mercury increases have been reported in the Atlantic (Chahid et al. 2014), Pacific (Ferriss and Essington 2014), Arctic (Braune et al., in press), and Indian Oceans (Carravierei et al. 2014). Over 90% of the increase in mercury concentrations over the past century can be attributed to anthropogenic mercury emissions (Dietz et al. 2009).
Mercury increases in Arctic aquatic ecosystems are difficult to both accurately measure and mediate. Our understanding of mercury fluxes in the Arctic is complicated by the complexity of the Arctic ecosystem: not only is the ecosystem vast, but it exhibits more seasonality in sunlight and primary production than any other ecosystem, waters span a salinity gradient from freshwater to salt water, and waters can be found as liquid, snow, ice, or permafrost (Braune et al., in press). An understanding of Arctic mercury fluxes, however, is of particular importance, as fish comprise the majority of the local Inuit diet (Saint-Amour et al. 2006). In humans, mercury ingested from contaminated fish can lead to sensory impairment, slow or impaired neurological development, diminished reaction time, neural degeneration, and, in high enough concentrations, death (Hansen and Gilman 2005). Increases in fish mercury levels in the Arctic may consequently lead to an increase in cases of mercury poisoning, raising troubling implications for human health (Hansen and Gilman 2005).
This review summarizes current knowledge about mercury concentrations in the Canadian Arctic, defined as the region of Canada north of tree line. I first discuss the sources of Arctic mercury before briefly describing the different chemical forms of mercury, how they are cycled through the ecosystem, and where they are deposited. Current studies on mercury concentrations in aquatic species, and the physiological effects of mercury poisoning, are then addressed, with attention to geographic variation in mercury levels and observed effects on human health. I conclude with discussions of current international efforts to reduce mercury emissions and of the theoretical complications to the mercury cycle arising from climate change.
SOURCES OF ARCTIC MERCURY
Mercury enters the atmosphere exclusively as gaseous elemental mercury (GEM), chemically identified as Hg0, from either natural, anthropogenic, or “re-emitted” sources (Pirrone et al. 2010). Anthropogenic sources account for 32% of mercury entering the Arctic annually. Coal and oil contain trace amounts of mercury, and their combustion was historically the predominant source of anthropogenic mercury (Pirrone et al. 2010). Since 2010, gold mining has produced a greater annual tonnage of mercury than fossil fuel combustion; the two industries are now responsible for 37% and 24% of anthropogenic emissions, respectively (Kessler 2013). Other anthropogenic sources of mercury include mercury released during the production of iron, steel, and cement, and from the inappropriate disposal of these products (Pirrone et al. 2010).
Natural sources account for 10% of the annual atmospheric mercury load and include volcanic eruptions and geothermal activity (Pirrone et al. 2010). “Re-emission” sources, when mercury is evaporated from marine and terrestrial surfaces or released by decaying organic matter, account for the remaining 58% of atmospheric mercury (Pirrone et al. 2010); however, the majority of the mercury being re-emitted was likely originally deposited from anthropogenic sources (Wilson et al. 2010).
Based on mercury concentrations measured in sediment cores, atmospheric mercury has increased seven-fold and oceanic mercury has increased six-fold since anthropogenic mercury emissions began around 2000 b.c. (Kessler 2013), with most of that increase within the past 150 years (Chételat et al., in press). Analysis of more recent sediments indicates that atmospheric mercury levels began to increase steadily in the late 1800s after the Industrial Revolution, presumably due to the development of mercury-intensive mining activities and the combustion of fossil fuels for electricity and heat (Dietz et al. 2009). Coal and oil combustion during the 20th century alone is believed to be responsible for 35% of the current total Arctic mercury load, with mercury waste from gold mining responsible for another 30% (Pirrone et al. 2010).
Because there are no point sources of mercury in the high Arctic, all Arctic mercury originates outside of the Arctic and is transported into the ecosystem by the atmosphere (Dietz et al. 2009). Studies of mercury transport in global wind and ocean currents estimate that 30-40% of Arctic mercury can be attributed to Asian sources, another 10-20% to North American sources, and less than 10% from European sources (Cole et al. 2013). The remainder comes from Africa and South America or cannot be accurately sourced. China is the predominant mercury source, responsible for one-third of annual global mercury emissions (Kessler 2013). The United States and India are the second- and third-highest mercury-emitting countries, respectively, but combined emit only one-third as much mercury as China (Sharma 2014). Annual mercury emissions from Asia have increased steadily in the past two decades, from approximately 800 metric tons in 1990 to over 1200 metric tons in 2005, while emissions from Europe and North America have remained constant at 300 metric tons (Wilson et al. 2010). The many, geographically distant sources of Arctic mercury suggest that mercury contamination is a global rather than a regional concern.
THE MERCURY CYCLE
Once mercury has entered the Arctic ecosystem as atmospheric GEM, it is transported through the environment along a complex series of photo- or bio-mediated pathways collectively referred to as the mercury cycle. GEM is highly volatile yet chemically stable, and therefore can be transported long distances and can remain in the atmosphere for 6-24 months before falling as precipitation (Cole et al. 2013). To fall as precipitation, GEM must be oxidized to form reactive gaseous mercury (RGM), chemically identified as Hg2+. GEM oxidation is catalyzed by sunlight in the presence of chloride or bromide radicals, like those found in seawater (Poulain et al. 2007). In the Arctic, massive GEM oxidation and precipitation events, called atmospheric mercury depletion events (AMDEs), occur each spring, when melting ice cover and increased incident angle of sunlight leads to vaporization of seawater (Braune et al., in press). Gaseous halogen radicals from the vaporized seawater interact with GEM in the atmosphere to form RGM. RGM subsequently condenses to form liquid Hg2+, which associates with water molecules or other particles in the atmosphere and then falls as rain or snow (Cole et al. 2013).
Following deposition, particle-bound Hg2+ is either re-emitted to the atmosphere or retained in soil, vegetation, surface water, sea ice, snowpack, or sediments (Chaulk et al. 2011). Because particle-bound Hg2+ is not volatile, re-emission occurs only when Hg2+ is reduced to Hg0 and vaporized in the presence of sunlight. This photoreduction reaction is favored at high UV intensities and low chloride concentrations, which are more common in the freshwater environment than the marine or terrestrial environments (Braune et al., in press). Much of the Hg2+ that is deposited in freshwater during AMDEs is exported to the ocean through the vast network of Arctic rivers. River deltas provide an important sink for freshwater mercury species: the Mackenzie River delta captures 19% of the mercury entering it, primarily due to the settling of particle-bound mercury as the river current slows, and consequently removes that mercury from exposure to biota (Emmerton et al. 2014).
Hg2+ floating in the water column can be accessed by aquatic microorganisms capable of methylating it to form methylmercury (MeHg) (Pucko et al. 2014). MeHg is the mercury species that causes mercury poisoning and is therefore the species of toxicological concern (Braune et al, in press). Mercury methylation occurs primarily in the anoxic conditions found deeper in the water column (Kirk et al. 2008), but methylation in surface waters has been documented (Cole et al. 2013). Vertical mixing patterns, driven by water temperature and salinity, bring MeHg to surface water. Demethylation is catalyzed by sunlight and therefore possible only in surface waters (Kirk et al. 2008). In seawater, MeHg complexes with chloride ions to form chlorinated methylmercury (MeHg-Cl), which is more resistant to photodemethylation than unmodified MeHg (Braune et al., in press). Because methylation rates are also lower in freshwater, MeHg concentrations are much lower in freshwater than seawater (Kirk et al. 2008), with cold, shallow, oligotrophic lakes exhibiting lower concentrations of MeHg than any other Arctic aquatic environment (Chételat et al., in press).
Aqueous Hg2+ that becomes trapped in snowpack, sea ice, or bound to sediment is not available to bacteria for methylation (Emmerton et al. 2014). When it eventually melts or is released, it is either converted to MeHg or photoreduced and evaporated. Very little aquatic mercury naturally exists as stored Hg2+ (Emmerton et al. 2014), though some Hg2+ can be stored for long periods of time in permafrost (Cole et al. 2013).
MERCURY CONCENTRATIONS IN ARCTIC BIOTA
Fish and other Arctic wildlife are exposed to mercury primarily through their diet, as the concentrations of mercury in the air and water are too small to appreciably affect biota via absorbance (Braune et al., in press). Phytoplankton absorb mercury species as they take nutrients from the water, and then higher-trophic-level organisms consume phytoplankton, acquiring their Hg. MeHg bioaccumulates as it ascends the trophic chain, so that top predators have higher concentrations of MeHg by weight compared to lower-trophic level organisms. The ratio of MeHg concentrations between two trophic levels is referred to as the biomagnification factor (BMF). Lowest observable effects levels (LOAELs) of MeHg concentrations, at which adverse physiological effects of mercury poisoning appear, range from 0.5 to 1.0 µg MeHg g-1 wet tissue weight in fish to 75 to 100 µg MeHg g-1 in mammals (Scheuhammer et al. 2014). The U.S. EPA and Environment Canada counsel against consuming fish with mercury concentrations above 0.5 µg g-1 wet tissue weight (U.S. EPA 2014).
Recent studies suggest that zooplankton, the lowest-trophic-level consumers of phytoplankton, have the highest BMF along the Arctic trophic chain. Zooplankton do not have higher MeHg concentrations than their phytoplankton prey, but dietary intake of MeHg only accounts for 30% of the MeHg body burden in zooplankton, suggesting that zooplankton somehow mediate a major step in biomagnification. Zooplankton are believed to harbor iron-reducing bacteria in their gut, allowing them to methylate Hg species acquired from the water or in their diet (Pucko et al. 2014). This internal methylation poses a threat to higher-trophic-level organisms, which, by consuming zooplankton, increase their load of neurotoxic MeHg relative to their load of less-toxic Hg2+. Analysis of Hg loading at different trophic levels indeed shows that, while MeHg comprises only 10-40% of the Hg load in zooplankton, consumers of zooplankton have an Hg load containing 80-100% MeHg (Pucko et al. 2014).
In the vast majority of fish species, mercury concentrations remain well below LOAELs (Scheuhammer et al. 2014), though between 70 and 90% of that mercury is MeHg (Braune et al, in press). For reasons that are not yet fully understood, freshwater fish have higher Hg concentrations than marine fish, despite the lower concentrations of mercury in freshwater (Scheuhammer et al. 2014). Some human food species, however, including golden shiners (Notemigonuscryso leucas) and certain populations of walleye (Sander vitreus) and burbot (Lota lota), have mercury concentrations in excess of 0.5 µg g-1 (Dietz et al. 2013, Scheuhammer et al. 2014). Affected individuals from these species have exhibited physiological effects of mercury poisoning, including disrupted endocrine function, hyperactivity, and altered shoaling behavior (Dietz et al. 2013). Differences between species-specific mercury concentrations are attributed to differences in feeding habit: fish that feed higher on the trophic chain, including walleye and burbot, have higher mercury concentrations than fish lower on the trophic chain, such as Pacific herring (Clupea pallasii) or rainbow smelt (Osmerus mordax) (Braune et al., in press). A similar trend is observed within populations, where older, larger fish have higher mercury concentrations because they have been exposed to more mercury over their lifetime (Tran et al. 2014).
Higher-trophic-level predators exhibit higher mercury concentrations, as expected given the ability of MeHg to bioaccumulate as it rises up the trophic chain. Piscivorous seabirds, such as thick-billed murres (Uria lomvia) and northern fulmars (Fulmarus glacialis), have mercury concentrations between 3 to 7 µg g-1, noticeably higher than in their prey, although no adverse physiological effects have yet been observed (Braune et al. 2014). However, 100% of the mercury transferred from adult female birds to their eggs is MeHg, suggesting that juvenile birds may be particularly susceptible to increasing mercury concentrations in their parents (Scheuhammer et al. 2014). Seabirds have higher concentrations of mercury than terrestrial birds, due to the higher concentrations of marine mercury, though seabirds are also hypothesized to be more tolerant of mercury toxicity than terrestrial birds (Scheuhammer et al. 2014).
Marine mammals and other top predators have the highest mercury concentrations, ranging from 1 µg g-1 to almost 100 µg g-1. The long-lived Greenland shark (Somniosus microcephalus) has more mercury than any other fish species (Scheuhammer et al. 2014). Mercury concentrations in whales are species-specific, depending on feeding habit; piscivorous whales, such as the pilot whale (Globicephala sp.) and beluga whale (Delphinapterus leucas), have mercury concentrations in excess of LOAELs, whereas planktonic- or benthic-feeding whales such as the grey whale (Eschrichtius robustus) have mercury concentrations far below LOAELs (Dietz et al. 2013). Among free-ranging polar bears (Ursus maritimus) from the Southern Beaufort Sea, juveniles have mildly impaired endocrine function attributable to mercury poisoning (Knott et al. 2012). In a study of coastal-dwelling grizzly bears (Ursus arctos ssp.), 70% of sampled grizzlies had MeHg concentrations in excess of LOAELs, and analysis of mercury in their fur indicates that the mercury concentration in their bodies increased dramatically within the past decade of their life (Noël et al. 2014).
Although mercury levels among subsistence Arctic fish species have not yet reached a point where mercury poisoning would be epidemic in Arctic human consumers, mild effects of mercury toxicity have been observed in sensitive human populations. Mercury concentrations measured in Inuit newborns are much higher than those measured elsewhere in North America, and Inuit preschoolers from northern Québec have slower pattern-reversal visual evoked potentials, indicating impaired or retarded neurological development (Saint-Amour et al. 2006). In another Arctic community, prenatal exposure to MeHg via the mother’s diet led to measurable neurological and psychological deficits by seven years of age, including impaired memory, language skill, and reaction time. Consistent, low-dose exposure to MeHg during gestation thus produces the same neurological effects as a single high-dose exposure (Hansen and Gilman 2005). The negative effects of even low-dose MeHg exposure warrants efforts to reduce mercury inputs to the Arctic ecosystem in order to safeguard human health.
Among all species, distinct geographical trends in mercury distribution have been observed. Mercury generally increases along an east-to-west, and a south-to-north, gradient (Braune et al. 2014; Pucko et al. 2014; Tran et al., in press). Mercury concentrations are consequently highest along the Greenland coast and in the Beaufort Sea, and lowest in Hudson Bay (Braune et al., in press). These gradients remain consistent even on a larger scale: mercury concentrations in the High Arctic exceed concentrations at sub-Arctic and mid-latitude sites, due to riverine and atmospheric transport of mercury northward from lower latitudes. The slower kinetics of the reduction and demethylation reactions at the colder temperatures of the High Arctic exacerbate this trend. (Cole et al. 2013). Some of the east-to-west gradient is explained by the increased exposure of the western Arctic to atmospheric mercury being transported from Asia (Wilson et al. 2010). Variations in the population size and species composition of microbial communities responsible for mercury methylation have also been suggested as a cause of this gradient (Braune et al., in press).
Despite the recognition that Arctic mercury originates from sources across the globe, no international attempts to regulate mercury emissions were undertaken until 2013. In October 2013, more than 140 nations signed the Minamata Convention on Mercury, named after a 1950s incident in Minamata, Japan, where over 900 individuals were killed from mercury poisoning associated with mercury effluent from a nearby factory (Sharma 2014). This international convention was drafted in recognition of mercury’s capacity for rapid diffusion even from a single country (UNEP 2013). According to the standards of this convention, signatory nations are required to reduce mercury emissions from anthropogenic sources by phasing out mercury amalgam in dental fillings (Mackey et al. 2014), prohibiting development of and phasing out or restructuring existing mercury-intensive mining operations (UNEP 2013), implementing mercury-capturing coal-burning techniques within 10 years (UNEP 2013), and eliminating the use of the mercury-based preservative thimerosal in medical products (Kessler 2013).
Current studies suggest that the Minamata Convention will not significantly affect global mercury concentrations. Signatory nations are not obligated to adhere to the convention until their national government has ratified it, a process which, depending on the nation, may not be complete until 2016 or 2017 (Sharma 2014). As of September 2014, only seven countries had ratified the treaty, only one of which—the United States—contributes appreciably to global mercury pollution (Kessler 2013, Sharma 2014). The convention also has no quantitative, measurable target for mercury reduction, nor are its provisions legally binding (Mackey et al. 2014). Even if a zero mercury emissions policy was implemented worldwide in 2015, models indicate that oceanic mercury would only decline to 67% of current levels by 2100 (Kessler 2013). Models that assume all regulations of the Minamata Convention will be adhered to indicate that the convention will prevent any significant increases in mercury but will not facilitate a decline from current concentrations (Selin 2013).
However, similar reduction efforts directed at organochlorine compounds, a major ingredient in pesticides in the late 1900s, have had noticeable effects on Arctic organochlorine contamination. Organochlorine concentration in polar bears from East Greenland declined by an average of 4.4% annually from 1983 to 2010 (Dietz et al. 2013), and in just one decade, organochlorine concentration in the Greenland Inuit population dropped to 41-56% of original levels (Bjerregaard et al. 2013). The effectiveness of organochlorine reduction efforts, which were implemented without a governing international treaty, suggests that the standards of the Minamata Convention may have some effect and should, if nothing else, be followed in accordance with the precautionary principle.
Climate change poses a unique challenge to the future of Arctic mercury concentrations, complicating the models of mercury emission reductions developed in response to the Minamata Convention. Increasing global temperatures will melt permafrost, releasing a significant amount of currently stored Hg2+ into the environment (Emmerton et al. 2014). At warmer temperatures the methylation reaction will also be faster, suggesting that both MeHg and overall Hg concentrations will increase due to climate change (Sunderland and Selin 2013). However, melting sea ice will expose more of the ocean surface to UV sunlight, which catalyzes the demethylation and subsequent evasion of mercury (Braune et al., in press). The relationship between these two opposing forces is not yet fully understood: a thirty-year study of Arctic charr found that, despite a 1.6 to 2.9°C increase in mean annual temperature in certain Arctic locations, resident populations of Arctic charr exhibited no associated increase in mercury concentrations, suggesting that, for at least small temperature increases, the increased mercury release balances the increased mercury evasion (Velden et al., in press).
The Canadian Arctic ecosystem is particularly vulnerable to the effects of mercury concentrations and exposure, making it an appropriate indicator ecosystem for mercury exposure in other aquatic ecosystems (Sunderland and Selin 2013). Given that seawater mercury concentrations are expected to increase by 50% by 2050 assuming no change in anthropogenic emissions (Sunderland and Selin 2013), continued monitoring of mercury concentrations in fish, birds, and other wildlife is necessary to determine the effectiveness of the Minamata Convention mercury-reduction practices and to ensure mercury in species consumed by humans remains at non-toxic concentrations. A better understanding of the mercury cycle in the Arctic aquatic environment may also reveal mechanisms by which humans could remove mercury from the environment before it enters the trophic web, including stockpiling mercury-laden river sediments and isolating the mercury for re-use (Emmerton et al. 2014).
Thanks for reading. Check out The Path is Not Always Clear →
Bjerregaard, P., H. S. Pederson, N. O. Nielsen, E. Dewailly. 2013. Population surveys in Greenland 1993-2009: Temporal trend of PCBs and pesticides in the general Inuit population by age and urbanization. Science of the Total Environment 454:283-288.
Braune, B., J. Chételat, M. Amyot, T. Brown, M. Claydon, M. Evans, A. Fisk, A. Gaden, C. Girard, A. Hare, J. Kirk, I. Lehnherr, R. Letcher, L. Loseto, R. Macdonald, E. Mann, B. McMeans, D. Muir, N. O’Driscoll, A. Poulain, K. Reimer, and G. Stern. 2014. Mercury in the marine environment of the Canadian Arctic: Review of recent findings. Science of the Total Environment, in press.
Braune, B., A. J. Gaston, H. G. Gilchrist, M. L. Mallory, and J. F. Provencher. 2014. A geographical comparison of mercury in seabirds in the eastern Canadian Arctic. Environment International 66:92-96.
Carravieri, A., Y. Cherel, P. Blévin, M. Brault-Favrou, O. Chastel, P. Bustamante. 2014. Mercury exposure in a large subantarctic avian community. Environmental Pollution 190: 51-57.
Chahid, A., M. Hilali, A. Benlhachimi, T. Bouzid. 2014. Contents of cadmium, mercury, and lead in fish from the Atlantic Ocean determined by atomic absorption spectrometry. Food Chemistry 147: 357-360.
Chaulk, A., G. A. Stern, D. Armstrong, D. G. Barber, and F. Wang. 2011. Mercury distribution and transport across the ocean-sea-ice-atmosphere interface in the Arctic Ocean. Environmental Science and Technology 45:1866-1872.
Chételat, J., M. Amyot, P. Arp, J. M. Blais, D. Depew, C. A. Emmerton, M. Evans, M. Gamberg, N. Gantner, C. Girard, J. Graydon, J. Kirk, D. Lean, I. Lehnherr, D. Muir, M. Nasr, A. J. Poulain, M. Power, P. Roach, G. Stern, H. Swanson, and S. van der Velden. 2014. Mercury in freshwater ecosystems of the Canadian Arctic: Recent advances on its cycling and fate. Science of the Total Environment, in press.
Cole, A. S., A. Steffen, K. A. Pfaffhuber, T. Berg, M. Pilote, L. Poissant, R. Tordon, and H. Hung. 2013. Ten-year trends of atmospheric mercury in the high Arctic compared to Canadian sub-Arctic and mid-latitude sites. Atmospheric Chemistry and Physics 13:1535-1545.
Dietz, R., F. F. Riget, C. Sonne, E. W. Born, T. Bechshoft, M. A. McKinney, and R. J. Letcher. 2013. Three decades (1983-2010) of contaminant trend in East Green polar bears (Ursus maritimus). Part 1: Legacy organochlorine contaminations. Environment International 59:485-493.
Dietz, R., P. M. Outridge, and K. A Hobson. 2009. Anthropogenic contributions to mercury levels in present-day Arctic animals—A review. Science of the Total Environment 407:6120-6131.
Emmerton, C. A., J. A. Graydon, J. A. L. Gareis, V. L. St. Louis, L. F. W. Lesack, J. K. A. Banack, F. Hicks, and J. Nafziger. 2013. Mercury export to the Arctic Ocean from the Mackenzie River, Canada. Environmental Science and Technology 47:7644-7654.
Ferriss, B., and T. E. Essington. 2014. Does trophic structure dictate mercury concentrations in top predators? A comparative analysis of pelagic food webs in the Pacific Ocean. Ecological Modelling 278: 18-28.
Hansen, J. C., and A. P. Gilman. 2005. Exposure of Arctic populations to methylmercury from consumption of marine food: An updated risk-benefit assessment. International Journal of Circumpolar Health 64(2): 121-136.
Kessler, R. The Minamata Convention on Mercury: A first step toward protecting future generations. Environmental Health Perspectives 121(10):A304-A309.
Kirk, J. L, V. L. St. Louis, H. Hintelmann, I. Lehnherr, B. Else, and L. Poissant. 2008. Methylated mercury species in marine waters of the Canadian High and Sub-arctic. Environmental Science and Technology 42:8367-8373.
Knott, K. K., D. Boyd, G. M. Ylitalo, and T. M. O’Hara. 2012. Lactational transfer of mercury and polychlorinated biphenyls in polar bears. Chemosphere 88(4):395-402.
Mackey, T. K., J. T. Contreras, and B. A. Liang. 2014. The Minamata Convention on Mercury: Attempting to address the global controversy of dental amalgam use and mercury waste disposal. Science of the Total Environment 472:125-129.
Noël, M., J. Spence, K. A. Harris, C. T. Robbins, J. K. Fortin, P. S. Ross, and J. R. Christensen. 2014. Grizzly bear hair reveals toxic exposure to mercury through salmon consumption. Environmental Science and Technology 48:7560-7567.
Pirrone, N., S. Cinnirella, X. Feng, R. B. Finkelman, H. R. Fredli, J. Leaner, R. Mason, A. B. Mukherjee, G. B. Stracher, D. G. Streets, and K. Telmer. 2010. Global mercury emission to the atmosphere from anthropogenic and natural sources. Atmospheric Chemistry and Physics 10:5951-5964.
Poulain, A. J., E. Garcia, M. Amyot, P. G. C. Campbell, F. Raofie, and P. A. Ariya. 2007. Biological and chemical redox transformations of mercury in fresh and salt waters of the High Arctic during spring and summer. Environmental Science and Technology 41:1883-1888.
Pućko, M., A. Burt, W. Walkusz, F. Wang, R. W. Macdonald, S. Rysgaard, D. G. Barber, J.-É. Tremblay, and G. A. Stern. 2014. Transformation of mercury at the bottom of the Arctic food web: An overlooked puzzle in the mercury exposure narrative. Environmental Science and Technology 48:7280-7288.
Scheuhammer, A., B. Braune, H. M. Chan, H. Frouin, A. Krey, R. Letcher, L. Loseto, M. Noël, S. Ostertag, P. Ross, and M. Wayland. 2014. Recent progress on our understanding of the biological effects of mercury in fish and wildlife in the Canadian Arctic. Science of the Total Environment, in press.
Sharma, A. 2014. Legally binding Minamata Convention on Mercury: Politics and science behind. Current Science 196(8):1063-1065.
Sunderland, E. M, and N. E. Selin. 2013. Future trends in environmental mercury concentrations: Implications for prevention strategies. Environmental Health 12(2):1-5.
Tran, L., Reist, J.D., and M. Power. 2014. Totally mercury concentrations in anadramous Northern Dolly Varden from the northwestern Canadian Arctic: A historical baseline study. Science of the Total Environment, in press.
United Nations Environment Program (UNEP). 2013. Minamata Convention on Mercury: Text and Annexes.
United States Environmental Protection Agency (U.S. EPA). 2014. Fish Consumption Advice. Available: http://www.epa.gov/mercury/advisories.htm (December 2014)
Velden, S., J. B. Dempson, M. Power. 2013. Comparing mercury concentrations across a thirty-year time span in anadromous and non-anadromous Arctic charr from Labrador, Canada. Science of the Total Environment, in press.
Wilson, S., J. Munthe, K. Sundseth, K. Kindbom, P. Maxson, J. Pacyna, and F. Steenhuisen. 2010. Updating historical global inventories of anthropogenic mercury emissions to air. AMAP Technical Report No. 3, Arctic Monitoring and Assessment Programme.