April 17, 2019 Administrator Andrew Wheeler US Environmental Protection Agency 1200 Pennsylvania Avenue, NW Washington, D.C. 20460 Docket No. EPA-HQ-OAR-2018-0794 Subject: Comments on “National Emission Standard for Hazardous Air Pollutants: Coal- and Oil-Fired Electric Utility Steam Generating Units—Reconsideration of Supplemental Finding and Residual Risk and Technology Review” As academic researchers studying mercury fate, transport, and impacts, we appreciate the opportunity to comment on the Reconsideration of Supplemental Finding and Residual Risk Technology Review. We write to encourage the Environmental Protection Agency to withdraw its Proposed Finding and Residual Risk Review, given limitations in its assessment methodologies and developments in scientific research on mercury processes and impacts since 2011. This assessment is based on our own research (Giang and Selin 2016; Perlinger et al. 2018; attached; Angot et al. 2018), as well as our review of relevant scientific literature. In the below, we focus on mercury as one important component of Hazardous Air Pollutants (HAPs). Our comments are in response to solicitations for comments C-2 and C-24. (C-2) The Regulatory Impact Analysis (RIA) from 2011 quantifies a very limited range of mercury-related benefits of regulation. Recent evidence indicates that benefits that were unquantified in the RIA may be substantial, and larger than the subset that were monetized. In Giang and Selin (2016), we estimated the benefits of the Mercury and Air Toxics Standards (MATS) policy for US populations. Annualized, our estimates out to 2050 are $3.7 billion/year, using a similar methodology to the RIA. Although methodological differences between our analysis and the RIA prevent direct comparison of numerical results, our research suggests that including a larger subset of health endpoints (IQ and heart attacks) and affected populations (consumers of self-caught freshwater fish and consumers of commercial marine and estuarine fish in the US market) may lead to mercury-related benefits estimates that are orders of magnitude larger than those reported in the RIA. Considering these additional mercury-related benefits could result in benefits that are of comparable magnitude to estimated costs (Rice et al. 2010; Grandjean and Bellanger 2017). Other recent research supports the conclusion that the subset of mercury-related benefits monetized in the RIA represent an incomplete picture of mercury impacts—both in terms of health and wellness impacts considered, and in the populations considered. There are several health benefits from reducing mercury emissions that could not be quantified in the RIA: cardiovascular effects, neurobehavioral effects aside from IQ loss, and immune endpoints (Karagas et al. 2012; Roman et al. 2011; Genchi et al. 2017). While efforts to develop dose-response relationships for use in regulatory benefit-cost analysis are underway, these potentially large health benefits should be taken into account in a holistic assessment of benefits and costs. Other benefits to individual and community health and well-being that are not easily quantifiable should also be included in a holistic assessment of benefits and costs, such as socio-cultural integrity (Ranco et al. 2011; O’Neill 2004). New evidence has also affirmed the importance of considering additional exposure pathways to self-caught freshwater fish: US Electricity Generating Units (EGU) also contribute to mercury pollution in coastal fisheries, and marine fish make up the vast majority of seafood diet for the US population as a whole (Sunderland et al. 2018, 2016). In the Proposed Finding, the EPA affirms that benefits described above that were not quantified in the RIA “are relevant to any comparison of the benefits and costs of a regulation” (p. 2678). If the EPA chooses to focus only on benefits associated with HAPs in its “appropriate and necessary”

determination—an approach that does not consider the full public health benefits of MATS—it should take into account recent research on the full scope and potential magnitude of mercury-related impacts. (C-24) The residual risk analysis methodology underestimates non-cancer risks associated with dietary exposure to mercury through fish for most-exposed populations. Recent research using state-of-the-science atmospheric and aquatic modelling suggests that fish-dependent communities may be exposed to levels exceeding EPA’s reference dose—even with policies like MATS in place. The multi-pathway exposure and risk screening assessment described in the Proposed Finding, and based on the National Emissions Standards for Hazardous Air Pollutants: Benzene Emissions from Maleic Anhydride Plants, Ethylbenzene/Styrene Plants, Benzene Storage Vessels, Benzene Equipment Leaks, and Coke By-Product Recovery Plants (Benzene NESHAP), may not be appropriate for methylmercury risk assessment. The assessment focuses on local (<50 km) impacts of single-facilities, rather than aggregate and cumulative effects of multiple facilities over regional scales. Mercury is indeed a local pollutant—the importance of US EGUs as a source of domestic deposition may in fact be underestimated in the RIA (Zhou et al. 2017; Sunderland et al. 2016). However, due to its different chemical forms, mercury also has regional impacts. As a result, regions with a high density of facilities that may be below the Tier 1 screening threshold emission rate, may still experience substantial mercury inputs. Further, the complex dynamics of mercury in aquatic and terrestrial ecosystems mean that some landscapes are more sensitive to mercury inputs than others (Perlinger et al. 2018). These dynamics are not captured in the exposure and risk screening assessment used by the EPA. Assessment methods that better capture the complex biogeochemistry of mercury indicate that highly-exposed populations (such as subsistence fishers and Tribal Nations where harvesting fish is an important dietary, socio-cultural, and rights-based practice) may be exposed to methylmercury at levels exceeding EPA’s reference dose, even with policies like MATS in place. In Perlinger et al. (2018), with colleagues, we model changes in atmospheric deposition and fish mercury concentrations in the Great Lakes resulting from policies at multiple jurisdictional scales (including MATS). We explore the environmental justice implications of these policy changes for a tribe in Michigan’s Upper Peninsula (UP) with a high fish consumption rate, the Keweenaw Bay Indian Community (KBIC). We find that even with regional, domestic, and global policy, KBIC members may be exposed to levels above the EPA reference dose at their desired level of fish consumption (454 g/day).1 The UP landscape is highly sensitive to Hg deposition, with nearly 80% of lakes estimated to be impaired. Sensitivity to mercury is caused primarily by the region's abundant wetlands. This finding suggests that landscape factors should be considered in the risk assessment process. Assessment methodologies such as the one described in the Proposed Finding also do not take into account legacy emissions (i.e., recycling of previously deposited mercury), and therefore underestimate health risks (Angot et al. 2018). Because it is persistent in the environment, mercury once emitted can circulate for decades to centuries. In Angot et al. (2018), with colleagues, using an integrated modelling approach, we quantify the impact of growing legacy reservoirs of mercury and find that due to this legacy penalty, aggressive maximum feasible reductions in emissions are needed for fish mercury concentrations to approach target levels by 2050 for high fish-consuming communities such as the Aroostook Band of Micmacs.

1 Research suggests that there is no threshold level of methylmercury exposure below which neurodevelopmental effects do not occur—consequently, the EPA reference dose is a non-conservative benchmark for health risk.

In closing, we urge that the EPA consider up-to-date information in both its appropriate and necessary finding, and its assessment of residual risk. Respectfully, Amanda Giang Assistant Professor of Environmental Modelling and Policy Institute for Resources, Environment and Sustainability Department of Mechanical Engineering University of British Columbia Noelle Selin Associate Professor Institute for Data, Systems and Society and Department of Earth, Atmospheric and Planetary Sciences Massachusetts Institute of Technology With signatories: Hélène Angot Research Associate Institute of Arctic and Alpine Research University of Colorado, Boulder Hugh Gorman Professor of Environmental History and Policy Department of Social Sciences Michigan Technological University Noel Urban Professor of Environmental Engineering Department of Civil and Environmental Engineering Michigan Technological University Valoree S. Gagnon Director, University-Indigenous Community Partnerships Great Lakes Research Center Michigan Technological University Judith Perlinger Professor Civil & Environmental Engineering Department Michigan Technological University [Attachments: Giang & Selin 2016; Perlinger et al. 2018; Angot et al. 2018]

Works Cited Angot, Hélène, Nicholas Hoffman, Amanda Giang, Colin P. Thackray, Ashley N. Hendricks, Noel R.

Urban, and Noelle E. Selin. 2018. “Global and Local Impacts of Delayed Mercury Mitigation Efforts.” Environmental Science & Technology 52 (22): 12968–77. https://doi.org/10.1021/acs.est.8b04542.

Genchi, Giuseppe, Maria Stefania Sinicropi, Alessia Carocci, Graziantonio Lauria, and Alessia Catalano. 2017. “Mercury Exposure and Heart Diseases.” International Journal of Environmental Research and Public Health 14 (1): 1–13. https://doi.org/10.3390/ijerph14010074.

Giang, Amanda, and Noelle E. Selin. 2016. “Benefits of Mercury Controls for the United States.” Proceedings of the National Academy of Sciences 116 (2): 286–91. https://doi.org/10.1073/pnas.1514395113.

Grandjean, Philippe, and Martine Bellanger. 2017. “Calculation of the Disease Burden Associated with Environmental Chemical Exposures: Application of Toxicological Information in Health Economic Estimation.” Environmental Health: A Global Access Science Source 16 (1): 1–14. https://doi.org/10.1186/s12940-017-0340-3.

Karagas, Margaret R, Anna L Choi, Emily Oken, Milena Horvat, Rita Schoeny, and Elizabeth Kamai. 2012. “Evidence on the Human Health Effects of Low-Level Methylmercury Exposure.” Environmental Health Perspectives 120 (6): 799–806.

O’Neill, Catherine A. 2004. “Mercury, Risk, and Justice.” Enironmental Law Review 34: 11070–115. Perlinger, J A, N R Urban, A Giang, N E Selin, A N Hendricks, H Zhang, A Kumar, et al. 2018.

“Responses of Deposition and Bioaccumulation in the Great Lakes Region to Policy and Other Large- Scale Drivers of Mercury Emissions.” Environmental Science: Processes & Impacts 20 (1): 195–209. https://doi.org/10.1039/C7EM00547D.

Ranco, Darren J., Catherine a. O’Neill, Jamie Donatuto, and Barbara L. Harper. 2011. “Environmental Justice, American Indians and the Cultural Dilemma: Developing Environmental Management for Tribal Health and Well-Being.” Environmental Justice 4 (4): 221–30. https://doi.org/10.1089/env.2010.0036.

Rice, Glenn E, James K Hammitt, and John S Evans. 2010. “A Probabilistic Characterization of the Health Benefits of Reducing Methyl Mercury Intake in the United States.” Environmental Science & Technology 44 (13): 5216–24. https://doi.org/10.1021/es903359u.

Roman, Henry a, Tyra L Walsh, Brent a Coull, Éric Dewailly, Eliseo Guallar, Dale Hattis, Koenraad Mariën, et al. 2011. “Evaluation of the Cardiovascular Effects of Methylmercury Exposures: Current Evidence Supports Development of a Dose-Response Function for Regulatory Benefits Analysis.” Environmental Health Perspectives 119 (5): 607–14. https://doi.org/10.1289/ehp.1003012.

Sunderland, Elsie M., Charles T. Driscoll, James K. Hammitt, Philippe Grandjean, John S. Evans, Joel D. Blum, Celia Y. Chen, et al. 2016. “Benefits of Regulating Hazardous Air Pollutants from Coal and Oil-Fired Utilities in the United States.” Environmental Science and Technology 50 (5): 2117–20. https://doi.org/10.1021/acs.est.6b00239.

Sunderland, Elsie M., Miling Li, and Kurt Bullard. 2018. “Decadal Changes in the Edible Supply of Seafood and Methylmercury Exposure in the United States.” Environmental Health Perspectives 126 (2): 029003. https://doi.org/10.1289/EHP3460.

Zhou, Hao, Chuanlong Zhou, Mary M. Lynam, J. Timothy Dvonch, James A. Barres, Philip K. Hopke, Mark Cohen, and Thomas M. Holsen. 2017. “Atmospheric Mercury Temporal Trends in the Northeastern United States from 1992 to 2014: Are Measured Concentrations Responding to Decreasing Regional Emissions?” Environmental Science and Technology Letters 4 (3): 91–97. https://doi.org/10.1021/acs.estlett.6b00452.

Benefits of mercury controls for the United StatesAmanda Gianga,1 and Noelle E. Selina,b

aInstitute for Data, Systems, and Society, Massachusetts Institute of Technology, Cambridge, MA 02139; and bDepartment of Earth, Atmospheric,and Planetary Sciences, Massachusetts Institute of Technology, Cambridge, MA 02139

Edited by Catherine L. Kling, Iowa State University, Ames, IA, and approved November 18, 2015 (received for review July 21, 2015)

Mercury pollution poses risks for both human and ecosystemhealth. As a consequence, controlling mercury pollution has becomea policy goal on both global and national scales. We developed anassessment method linking global-scale atmospheric chemical trans-port modeling to regional-scale economic modeling to consistentlyevaluate the potential benefits to the United States of global (UNMinamata Convention on Mercury) and domestic [Mercury and AirToxics Standards (MATS)] policies, framed as economic gains fromavoiding mercury-related adverse health endpoints. This methodattempts to trace the policies-to-impacts path while taking intoaccount uncertainties and knowledge gaps with policy-appropriatebounding assumptions. We project that cumulative lifetime benefitsfrom the Minamata Convention for individuals affected by 2050 are$339 billion (2005 USD), with a range from $1.4 billion to $575 billionin our sensitivity scenarios. Cumulative economy-wide benefits tothe United States, realized by 2050, are $104 billion, with a rangefrom $6 million to $171 billion. Projected Minamata benefits aremore than twice those projected from the domestic policy. Thisrelative benefit is robust to several uncertainties and variabilities,with the ratio of benefits (Minamata/MATS) ranging from≈1.4 to 3.However, we find that for those consuming locally caught freshwa-ter fish from the United States, rather than marine and estuarinefish from the global market, benefits are larger from US than globalaction, suggesting domestic policies are important for protectingthese populations. Per megagram of prevented emissions, our do-mestic policy scenario results in US benefits about an order of mag-nitude higher than from our global scenario, further highlightingthe importance of domestic action.

mercury | policy | impacts assessment | Minamata Convention | economicbenefits

Toxic contamination from human activities is a global prob-lem. Although some countries have regulated toxic sub-

stances such as heavy metals and persistent organic pollutants forseveral decades, chemical contamination has still been identifiedas a key planetary boundary at risk for exceedance in the contextof global change (1). To address this challenge, existing globalenvironmental treaties try to manage the entire life cycle ofchemical contaminants (2). The newest of these is a global treatyon mercury, the Minamata Convention. In November 2013, theUnited States became the first country to fulfill the requirementsnecessary to become a party to the convention.In the United States, analyses to support domestic environ-

mental decision-making include socioeconomic valuations ofimpacts as part of the regulatory process. However, these eval-uations can be both scientifically challenging and politicallycontentious, particularly given uncertainties and knowledge gaps(as noted in arguments in a recent case heard in the US SupremeCourt, Michigan v. Environmental Protection Agency, 2015,addressing analysis of the costs and benefits of mercury regula-tion). These challenges are especially difficult for contaminantssuch as mercury, which cross temporal and spatial scales andhave both domestic and global sources. The chain of analysisfrom policies, through emissions, to impacts involves a complexpathway, which for mercury includes industrial activities, atmo-spheric chemistry, deposition processes, bioaccumulation, andhuman exposure. Existing approaches have not fully combinedinformation and knowledge from these disparate fields, and

substantial gaps exist in scientific understanding of the processesthat mercury undergoes through long-range transport. Thus, ithas historically been difficult to quantitatively estimate pro-spective domestic benefits from global environmental treaty-making in ways that can be compared with socioeconomicanalyses designed to support domestic environmental decision-making. Here, we use an assessment approach that enablestracing this pathway, accounting for best-available scientific un-derstanding and addressing uncertainties and knowledge gapswith policy-appropriate assumptions.Mercury is a naturally occurring element, but human activities

such as mining and coal combustion have mobilized additionalamounts, enhancing the amount of mercury circulating in theatmosphere and surface oceans by a factor of three or more(3, 4). Mercury previously deposited to land and water canrevolatilize over decades to centuries. Thus, human activitieshave fundamentally altered the global biogeochemical cycle ofmercury (5). Deposited mercury in aquatic systems can be con-verted to more toxic methylmercury (MeHg), which bio-accumulates. People are then exposed to MeHg by eatingcontaminated fish. Effects of MeHg exposure include IQ deficitsin prenatally exposed children (6–8) and may include cardiovas-cular effects in adults (7, 9). Scientific uncertainty and variabilityare substantial throughout this pathway, including but not limitedto atmospheric chemistry, deposition patterns, methylation pro-cesses, bioaccumulation and food web dynamics, dietary patterns ofexposure, and dose–response relationships. Despite these uncer-tainties, scientific analyses have been conducted to support decision-making, and state-of-the-art models exist for many of these steps.Some studies have previously traced the pathway from mer-

cury emissions to human impacts. These studies are limited inhow completely they have represented physical processes, andhow they have accounted for knowledge gaps. First, many do notexplicitly consider spatial transport through the environment on aglobal scale, and so do not explicitly link emissions to exposurechanges (10–14). Timescales associated with bioaccumulationthrough ecosystems also are often not taken into account, making

Significance

Mercury is a globally transported pollutant with potent neu-rotoxic effects for both humans and wildlife. This study intro-duces an assessment method to estimate the potential humanhealth-related economic benefits of global and domestic mer-cury control policies. It finds that for the US population as awhole, global mercury controls could lead to approximatelytwice the benefits of domestic action by 2050. This result isrobust to several uncertainties and variabilities along theemissions-to-impacts path, although we find that those con-suming locally caught freshwater fish in the United Statescould benefit more from domestic action.

Author contributions: A.G. and N.E.S. designed research; A.G. performed research; A.G.and N.E.S. analyzed data; and A.G. and N.E.S. wrote the paper.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.

Freely available online through the PNAS open access option.1To whom correspondence should be addressed. Email: agiang@mit.edu.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1514395113/-/DCSupplemental.

286–291 | PNAS | January 12, 2016 | vol. 113 | no. 2 www.pnas.org/cgi/doi/10.1073/pnas.1514395113

it difficult to evaluate how the timing of emissions changes affectsbenefits (15). Few studies have explicitly included more uncertain,but potentially important, health endpoints such as cardiovasculareffects in their estimates (12, 16). For instance, the US Environ-mental Protection Agency (15) focused on only IQ-related MeHgeffects in their analysis of the Mercury and Air Toxics Standards(MATS) in the United States. Finally, methods used for previousstudies were not designed to highlight the relative importance ofuncertainties throughout the policies-to-impacts path.We explicitly incorporate uncertainty and sensitivity analysis

for key steps along the policies-to-impacts pathway to assess therelative importance of policy-relevant uncertainties. We combinebest available models to trace projected global mercury policy sce-narios to their US impacts. We use atmospheric modeling to projectthe amount of mercury depositing to the US and global seafoodsource regions with and without global policy. We incorporate as-sessment of timescales associated with bioaccumulation throughecosystems. We then link atmospheric mercury models to economicvaluation models, generating a representation of mercury impactsthat takes into account environmental and human response time-scales. We use this assessment approach to present what is, to ourknowledge, a first assessment of potential US benefits, defined ineconomic terms, from the Minamata Convention. We explicitlycompare benefits of global and US policies, using consistentmethodology, and analyze the relative impacts of these policies onthe US population. We first present results from a base case analysisof mercury policy to 2050, using our integrated model. We thenpresent our sensitivity analyses, assessing the influence of uncer-tainties on our base case results.

Results and DiscussionTracking the Policies-to-Impacts Pathway: Base Case. Globally, ouremissions projections under the Minamata Convention willresult in 2050 in emissions of 1,870 Mg·y−1, which is roughlyequivalent to the present-day level, but 2,270 Mg·y−1 less thanour no policy (NP) scenario (17). The largest sources of an-thropogenic mercury emission are stationary coal combustion,artisanal and small-scale gold mining, and metals production(18). Under NP, emissions are projected to more than double,largely as a result of growth in coal use in Asia (19); thus, themain differences in policy and NP projections depend on as-sumptions about emission controls for coal (20). Air qualityabatement technologies such as flue gas desulfurization cancapture mercury as a cobenefit. For global emission projectionsunder the Minamata Convention, which requires the applicationof best available technologies, taking into account technical andeconomic feasibility, we assume the application of flue gas de-sulfurization or similar technology outside of the United States(17, 19). In the United States, our policy scenario is based onMATS (currently under legal challenge), which was designed tocontrol Hg emissions from power generation, with full imple-mentation by 2016 (15). In the United States, emissions in 2005were ∼90 Mg·y−1 (15). Under our MATS projection, we extend

the US Environmental Protection Agency projected trend from2016 to 2020 (15, 21) linearly, resulting in 2050 US emissions of46 Mg·y−1. Our NP case for the United States includes no furtherimprovements in emissions control technology or policy, andthus results in an approximate doubling of 2005 emissions by2050 (19). Benefits of the Minamata Convention to the UnitedStates are calculated as the difference between the globalMinamata and NP scenarios, holding US emissions constant at theMATS scenario. Benefits of MATS to the US are calculated as thedifference between the US NP and MATS scenarios, holdingemissions in the rest of the world constant at the NP scenario.Under our Minamata case, mercury deposition to the United

States and to the global oceans are 19% and 57% less than underNP in 2050, respectively. Fig. 1 maps these deposition differencesover the contiguous United States. We model the atmospherictransport and deposition of mercury using the global, 3D land-ocean-atmosphere mercury model GEOS-Chem v.9-02, at 4° ×5° resolution globally and 0.5° × 0.667° resolution over theUnited States (22–26). We use net total deposition as a measureof mercury ecosystem enrichment (27). SI Appendix, Chemicaltransport modeling gives additional details on the modeling ap-proaches. For our MATS case, deposition to the United States is20% less than under NP, and deposition to the global oceans is6% less. Although the modeled avoided deposition over theentire United States is similar under MATS and Minamata, thedistribution of these differences varies, as shown in Fig. 1.Avoided deposition under MATS is more highly concentrated inthe Northeast, where there are significant coal-fired emissionsources. In contrast, US deposition benefits under the MinamataConvention follow precipitation patterns, as policy avoids in-creases in the global background mercury concentration.Because mercury is persistent in the environment, anthropo-

genic emissions also enrich reservoirs of mercury in the subsur-face ocean and soils. Mercury from these pools can enhancereemissions, contributing further to deposition. Our GEOS-Chemsimulations take into account the effect of anthropogenic emis-sions changes on concentrations of mercury in surface reservoirsonly, and consequently underestimate the total deposition benefitsattributable to policy. To roughly estimate the extent of this un-derestimation, we use a seven-box, biogeochemical model de-veloped by Amos et al. (28, 29), which captures the deep oceanand soil reservoirs, but not the spatial distribution of impacts (SIAppendix, Chemical transport modeling). We find that globally,deposition reductions under policy are ∼30% larger when takinginto account enrichment of these subsurface pools.Recent research suggests that fish concentrations in ocean (30–32)

and freshwater (33–36) fish will likely respond proportionally tochanges in atmospheric inputs over years to decades, although themagnitude and timing of a full response may be variable, dependingon the region (see refs. 32 and 37–39 for examples). For our basecase scenario, we assume that fish MeHg in both freshwater andmarine ecosystems responds after 10 y to proportionally reflectchanges in atmospheric inputs (we test the response to this assumption

Fig. 1. Projected net deposition benefits (Δμg/m2∙y) of MATS and the Minamata Convention over NP over the contiguous United States, at 0.5° × 0.667°resolution. Global results, at 4° × 5° resolution, are shown in SI Appendix, Fig. S4.

Giang and Selin PNAS | January 12, 2016 | vol. 113 | no. 2 | 287

SUST

AINABILITY

SCIENCE

in our sensitivity analysis) (30, 37). We specify base year bloodMeHg, as a biomarker for MeHg exposure, by region, based onthe National Health and Nutrition Examination Survey (40). Wethen scale blood concentrations based on the change in intake offish MeHg (change in deposition plus time lag), taking into ac-count consumption of domestic freshwater and imported fishspecies from global fisheries, using data from US seafood marketstudies (41) and data compiled by the US Environmental Pro-tection Agency on noncommercial anglers (15, 42). Because ofdata limitations, we consider noncommercial mercury intake fromlocal, freshwater fish only. We treat noncommercial marine an-glers as average US consumers of marine and estuarine fish. Thismay slightly underestimate the benefits of MATS in our work;however, further data are necessary to quantify the MeHg intakeof noncommercial anglers in different US coastal regions (seeSI Appendix, Changes in human exposure for detailed methods).Calculated average US mercury intake in 2050, assuming a 10-y

time lag between deposition changes and fish response, as wellas constant fish intake patterns, is 91% less under our Minamatascenario than under NP (SI Appendix, Fig. S5). Our MATS scenarioreduces intake by 32% compared with the NP case. Although thedeposition decreases over the United States are roughly equivalentbetween the MATS and Minamata scenarios, changes to modeledmercury intake are larger under the latter. More than 90% of theUS commercial fish market, and the majority of US mercury intake,comes from marine and estuarine sources, particularly from Pacificand Atlantic Ocean basins (41, 43). These regions are heavilyinfluenced by emissions from non-US sources, including East andSouth Asia. In addition, even locally caught freshwater fish areaffected by the long-range transport of mercury emissions. Re-gional differences in the geographic source of dietary fish (SIAppendix, Changes in human exposure) and deposition lead tovariations in intake change patterns across scenarios, as shownin SI Appendix, Fig. S4. The majority of modeled MeHg intakein the North Central region (SI Appendix, Fig. S5) is from self-caught, local freshwater fish, leading to a diminished intakebenefit from the Minamata scenario relative to the MATSscenario. The opposite pattern holds for New York. Thesedifferences in intake lead to corresponding differences in IQdeficits and cardiovascular outcomes (see SI Appendix, IQeffects; Cardiovascular impacts; and Health impacts for healthimpacts methods and results, respectively).Annual US economic benefits to 2050 (applying a 3% discount

rate) from avoided health impacts under domestic and globalmercury policies under our base case assumptions, relative toNP, are presented in Fig. 2. We use two economic valuationapproaches: the first, a cost-of-illness and value of statistical life

(VSL) approach, estimates projected lifetime (LT) benefits ofavoided exposure for those born by 2050 and is consistent withUS regulatory practice; the second, a human capital approach,estimates economy-wide (EW) benefits realized by 2050 fromavoided labor productivity and wage losses. Given differences inmethodology, results from these two approaches are not directlycomparable (see SI Appendix, Economic modeling of healthimpacts for more details). To estimate LT benefits of avoidedhealth effects, we apply estimates of projected lost wages andmedical costs for IQ deficits and nonfatal acute myocardialinfarctions (heart attacks), and VSL for premature fatalitiesresulting from myocardial infarctions (see ref. 12 and exampleslisted in ref. 44 of studies that use this approach), for each year’sprojected birth cohort (IQ) and affected adult population (heartattacks). The second method uses the US Regional Energy Policymodel, a computable general equilibrium model of the US econ-omy (45). Consistent with previous work valuing economic effectsof air pollution through computable general equilibrium modeling(46), we take into account the effects of IQ deficits and fatal andnonfatal heart attacks on the labor force, and its cumulative effectover time. Base case cumulative EW benefits of the MinamataConvention to the United States by 2050 are $104 billion (2005USD) (Fig. 2, Top, blue line), and cumulative LT benefits for thoseborn by 2050 are $339 billion (Fig. 2, Bottom, green line). EWbenefits from our MATS scenario (Fig. 2, Top, red line) are $43billion by 2050, and LT benefits are $147 billion (Fig. 2, Bottom,purple line). Both EW and LT benefits are dominated (>90% forLT and >99% for EW) by avoided cardiovascular effects, consis-tent with previous studies, including these health endpoints (12,16). Relative to US domestic action, estimated cumulative benefitsfrom the Minamata Convention are more than twice as large.Considered per unit of avoided emissions, however, the pro-

jected benefits of MATS to the United States are larger than thoseof the Minamata Convention: $324 million/Mg compared with$46 million/Mg for EW benefits by 2050, and $1.1 billion/Mgcompared with $150 million/Mg for LT benefits for those born by2050. Given its global scope, the Minamata Convention is likely toprevent more emissions than MATS. However, as mercury pollu-tion has effects on both local and global scales, avoided emissionswithin the United States, on a per unit basis, lead to larger benefits.

Policies-to-Impacts Sensitivity Analysis. We assess uncertainty andvariability along the policies-to-impacts pathway by identifyingkey drivers of uncertainty in our base case integrated model, andcalculating how changes in assumptions affect our quantificationof US benefits from the Minamata Convention, MATS, andrelative benefits. Key assumptions addressed here include the ef-fect of atmospheric chemistry, ecosystem time lags, dietary choices,dose–response parameters linking MeHg exposure and health ef-fects, economic costs, and discount rates. We run the integratedmodel for realistic and policy-relevant low and high bounds forthese assumptions. Fig. 2 shows the range of calculated benefitsfrom these sensitivity scenarios, described further here. The un-certain range spanned by these cases is illustrated by the lines inFig. 2; however, the bounds delineated by these lines for theMinamata (blue/green) and MATS (red/purple) scenarios are notindependent. Some sensitivity scenarios lead to the same di-rectional change in benefits over the base case for both the do-mestic and global scenario, such that the magnitude of cumulativebenefits for the Minamata scenario remain larger than for MATS.This result is illustrated in Fig. 3, which shows the range in ratio ofbenefits between Minamata and MATS, under different sensitivityscenarios. Details of the low and high cases addressed are pre-sented in SI Appendix, Table S7 and Sensitivity analysis.Our low and high cases for atmospheric chemistry bound un-

certainty about the form of mercury emissions and atmosphericredox reactions. Although policies address total mercury emis-sions, emissions of mercury occur as different chemical specieswith different atmospheric lifetimes. Mercury emitted in itselemental form, Hg(0), has an atmospheric lifetime of 6 mo to ayear, enabling it to transport globally before its oxidation and

Fig. 2. Trajectories of welfare benefits under global and domestic policyuntil 2050, discounted at 3%. (Top) Modeled EW benefits realized in a givenyear. (Bottom) Projected LT benefits for that year’s affected population. Basecases are indicated with markers. Unmarked lines show the range of trajec-tories from sensitivity cases.

288 | www.pnas.org/cgi/doi/10.1073/pnas.1514395113 Giang and Selin

subsequent deposition. Mercury emitted in its oxidized form, Hg(II),in the gas phase, or Hg(P) in the particle phase, is more soluble andcan deposit closer to its source. In addition, the speciation of present-day mercury emissions is uncertain. Reduction reactions may convertHg(II) to Hg(0), lengthening its lifetime; this process may occur inthe atmosphere in the aqueous phase (47), or in power plant plumes(48, 49). However, the mechanism of potential reduction is unknown.To bound this uncertainty, we assume for our low case that 90% ofglobal Hg reductions over NP occur as Hg(II) or Hg(P), and for thehigh case, that 90% of reductions occur as Hg(0). This results in arange of cumulative EW benefits for Minamata between $102 billion(low) and $123 billion (high) in 2005 USD, and a range of LTbenefits of $338 billion to $405 billion. That the low case results inonly a small difference from the base case reflects the emphasis oncontrol technologies that capture oxidized mercury in the base caseassumptions (19). The relative benefits of Minamata versus ourMATS case vary to a factor of 2.9 from the base case. If policyprevents primarily Hg(0) emissions, or there is a high rate of in-plumereduction, there is greater long-range benefit to the United Statesand global oceans from avoided emissions occurring elsewhere.If fish MeHg responds rapidly and quantitatively to changes

in deposition, cumulative EW and LT benefits to 2050 fromMinamata are projected to be $171 billion and $575 billion (2005USD), whereas a slower response reduces projected EW and LTbenefits to $18 billion and $60 billion. Although reductions inmercury deposition, all else equal, will eventually result indecreased environmental and fish concentrations, benefits withina given time horizon, which in this case is 2050, will depend onhow long ecosystems take to respond. Estimated economic

benefits are therefore highly sensitive to the temporal scope ofanalysis. For instance, EW benefits from IQ effects are primarilyaccrued when those in birth cohorts with reduced exposure are ofworking age (see SI Appendix, Economic modeling of healthimpacts), and consequently are not fully captured by our 2050 timehorizon. Population growth and discounting assumptions (we use a3% discount rate; see SI Appendix, Economic valuation for others)also influence our cumulative benefit assessment. Timing effects arefurther discussed in SI Appendix, Economic valuation. Our lowerbound incorporates an instantaneous response, which is the as-sumption commonly used in regulatory analyses (15, 42), and thatmay be roughly consistent with the behavior of certain classes offreshwater bodies (37). Our upper bound is 50 y, consistent with thehigh range of estimated response times for surface open oceanwaters (30), where MeHg production and biomagnification arehypothesized to occur (31), and midrange estimates for watershed-fed coastal ecosystems and some lake systems, which may be theslowest to respond to changes in atmospheric deposition (32, 36).Population dietary choice between local freshwater and global

market fish alters our Minamata base case cumulative EW ben-efits from $17 billion (2005 USD) to $127 billion, and cumulativeLT benefits from $56 billion to $418 billion. Our base case as-sumes that population dietary choices between local fish andglobal market fish remain constant over time. For low and highbounds, respectively, we assume that people’s diets are 100% fromeither local freshwater or global sources. Where US seafoodconsumers eat a larger fraction of market marine and estuarinefish, benefits from Minamata are higher. Under the 100% localfreshwater diet assumption, benefits from MATS exceed those ofMinamata (Minamata/MATS ratio of 0.4 in Fig. 3).With different assumptions about pharmacokinetics and dose–

response functions between mercury intake and human healtheffects, our results for the Minamata scenario vary from $6million to $160 billion (2005 USD) in EW benefits, and from$1.4 billion to $498 billion in LT benefits. Although convincingevidence is present to associate MeHg with adverse human ef-fects at low to medium doses, particularly for IQ deficits (7, 50),there may be variability in the magnitude of this effect; for in-stance, because of genetic variability (51). As a result, we use95% confidence interval bounds for high and low cases for bio-marker and dose–response parameters (SI Appendix, Table S3).Associations between mercury exposure and cardiovascular im-pacts are less certain than IQ effects (9). Previous studies haveexpressed this uncertainty, using an expected value approachtaking into account both the plausibility of a relationship be-tween MeHg and cardiovascular impacts and uncertainties in theparameters of the relationship (12). Our lower bound does notinclude cardiovascular impacts, whereas our base case and upperbound do, with the 97.5 percentile estimate of the relationshipbetween hair mercury and heart attack risk used in the high case(SI Appendix, Sensitivity Analysis) (52). A more detailed review ofthe epidemiological evidence contributing to these parameteri-zations is given in SI Appendix, IQ effects and Cardiovascularimpacts. Although using different exposure–response functionsleads to the largest absolute range in cumulative benefits amongthe sensitivity cases considered (Fig. 3), the relative benefitsbetween Minamata and MATS do not change as substantially.High and low assumptions for the economic valuation of

mercury-related health effects lead to a range of $58 billion to$121 billion (2005 USD) in EW benefits from the Minamatascenario by 2050, and a range of $87 billion to $518 billion in LTbenefits. Our sensitivity scenarios for EW benefits address onlymorbidity, and not mortality, effects: medical costs associatedwith heart attacks, and the relationship between IQ deficits andlost earnings. We use the 95% confidence interval for the IQ toincome relationship and the range of estimates for medical costsfrom the literature as bounding cases (SI Appendix, Table S7).For LT valuations, we use central and range estimates for VSL andLT lost income from regulatory literature (15, 53). The valuationuncertainties considered have the smallest effect on the ratio ofbenefits between global and domestic scenarios (Fig. 3).

Fig. 3. (Top) Range in cumulative benefits of the Minamata scenario to2050. Note the different scales for LT and EW benefits. (Bottom) Range inratio of cumulative benefits to 2050 (Minamata Benefits/MATS Benefits).Blue and green lines show base case results for LT and EW benefits, re-spectively. Bars indicate the sensitivity of cumulative benefits to high andlow case assumptions for uncertain parameters.

Giang and Selin PNAS | January 12, 2016 | vol. 113 | no. 2 | 289

SUST

AINABILITY

SCIENCE

Implications for Policy Evaluation. We developed and applied anassessment method to examine the complex pathways frompolicies to environmental effects for global toxic pollution frommercury that accounts for uncertainties and knowledge gaps in astructured way. We showed, using this method, that by 2050, theMinamata Convention could have approximately twice the benefitof our scenario simulating domestic actions ($104 billion comparedwith $43 billion in cumulative EW benefits, and $339 billion com-pared with $147 billion in cumulative LT benefits). The relativebenefit is robust to several uncertainties assessed along the policies-to-impacts pathway, including atmospheric chemical processes,ecosystem time lags, and exposure–response relationships; how-ever, we find that domestic action has a larger benefit when dietaryfish is sourced from local freshwater bodies. Per megagram ofavoided emissions, the benefits to the United States of domesticaction are nearly an order of magnitude larger than global action,highlighting that although mercury is a global pollutant, local pol-icies contribute strongly to local benefits. As shown in SI Appendix,Fig. S4, avoided emissions associated with the Minamata Conven-tion outside of the United States may lead to large benefits in Asiaand Southern Europe. Abatement costs will also vary by region.Although we have conducted what is, to our knowledge, the first

global-scale attempt to link future emissions trajectories to domesticimpacts, our ability to incorporate detailed models of the entirepathway is limited by existing scientific knowledge. In additionto these knowledge gaps, there are also variabilities in mercury’sbehavior across ecosystems and regions, as well as in humanresponses (physical and social). Our approach uses boundingassumptions along the policies-to-impacts pathway as a proxy toassess the relative influence of various uncertainties, from arange of disciplines. In a number of previous analyses, range inthe benefits of mercury reduction has been specified by the rangein exposure–response functions (12, 13). Although our analysisunderlines the importance of these uncertainties, particularlythose related to cardiovascular effects, it also suggests that pre-vious approaches miss other potentially large contributors touncertainty in economic effects (particularly within a given timehorizon), such as marine and freshwater ecosystem dynamics anddietary intake variabilities.Although, all else being equal, mercury emissions reductions

will ultimately result in exposure reductions, our analysis in-dicates that uncertainties in ecosystem dynamics affecting thetimescale of these reductions will strongly influence benefitswithin a given time horizon. Many of the processes affecting theconversion of inorganic mercury to MeHg and subsequentuptake in biota are poorly understood, particularly in marineecosystems (54, 55). In addition, there is variability among eco-system types, both freshwater (37) and marine and estuarine(32), in how quickly these systems and biota within them respondto changes in deposition. As described previously, our analysisfocuses on changes to mercury in surface reservoirs, and ac-counting for these effects could increase benefits estimates by∼30%. Future research should more fully address the timescalesof reemissions from subsurface reservoirs, both land and ocean,and their effects on benefits estimates. Better understanding ofmercury cycling, methylation and bioaccumulation processes,their variability, and the potential effects of global changes toclimate, land use, and other environmental contaminants will becritical for improving policy evaluation (56), particularly forbetter understanding the distribution of benefits between currentand future generations.Our analysis also reveals the importance of social factors in

estimating the absolute and relative benefits of different policies.Dietary choices, including fish selection and consumption rate,can have a potentially larger influence on the ratio of benefitsfrom global compared with domestic action than substantialscientific uncertainties about mercury′s environmental behavior.This sensitivity result suggests that domestic actions may beparticularly important for reducing exposure for communitiesthat consume mostly fish sourced from the contiguous UnitedStates, such as certain Indigenous peoples and immigrant groups,

subsistence fishers, and recreational anglers. In addition, ithighlights the policy need for analysis and data collection on theevolving patterns in fisheries production and fish consumption(43). It has been noted that dietary guidance on fish selectionand consumption frequency could be part of an adaptationstrategy to minimize mercury exposure (57), and our resultspoint toward their potentially large effect as a policy lever.However, dietary advice is highly complex. Fish consumption,and specific fish selection, can have substantial benefits, bothnutritional (58, 59) and sociocultural (60). Balancing the risksand benefits of fish consumption therefore requires carefulconsideration of contextual factors. Even with such adaptiveapproaches, there is continued need to mitigate future emissions.Although uncertainties related to chemical speciation of

emissions reductions led to the smallest range in cumulativebenefits for the Minamata scenario, interactions between theseuncertainties and variabilities in dietary fish source could affectthe relative benefits of global versus domestic action. At thistime, our ability to constrain these speciation uncertainties ispartially limited by measurement challenges (61). Improvedmeasurement techniques could provide insight into distributionalaspects of control policies.Differences in valuation methods for health endpoints could

lead to substantial variation in benefits estimates. Our two val-uation approaches highlight some of these potential variations:Our EW approach emphasizes compounding economy-widegains over time, but considers only effects to the economy (notindividuals) realized within the 2050 time horizon; in contrast,our LT approach more closely resembles regulatory studies,taking into account projected lifetime and nonmarket effects toindividuals (e.g., pain and suffering). As highlighted previously,economic benefits estimates are very sensitive to choices oftemporal scope of analysis and discounting. Estimates are alsosensitive to the endpoints considered: In addition to the healtheffects considered here, there may be other human and wildlifehealth endpoints not included in this study that, although notwell characterized at this time (7), may also have economic ef-fects. No less important, there may be dimensions of individualand community health and well-being that are not quantifiablewithin this economic framework, which should be considered in aholistic assessment of policy benefits (62).Our assessment of US benefits from global and domestic policy

is designed to be illustrative, drawing attention to uncertainties inestimating economic benefits and methods to take these uncer-tainties into account. As a consequence, our estimates should notbe taken as a comprehensive projection of impacts. However, asscientific knowledge evolves, many uncertainties can be addressedusing similar methodology. Policies-to-impacts analyses similar tothe one presented here can be valuable for synthesizing availableinformation, identifying its limitations, and when combined withsensitivity analysis, suggesting areas where scientific data collec-tion to narrow uncertainty would lead to uncertainty reduction ofimportance to policy-making.

Materials and MethodsBrief explanations of methods have been included throughout Results andDiscussion. In the SI Appendix, Supplementary methods, we provide a detaileddescription of methodology and data sources for emissions projections,chemical transport modeling, translating changes in deposition to changes inhuman exposure, IQ and cardiovascular impacts modeling, economic modelingof health impacts, and sensitivity analysis. Institutional review and informedconsent were not necessary for this modeling study, as all human health andecosystem input data were drawn from published sources.

ACKNOWLEDGMENTS. This work was supported by the National ScienceFoundation (Awards 1053648 and 131755), the J. H. and E. V. Wade Fund(Massachusetts Institute of Technology), and fellowships from the NaturalScience and Engineering Research Council of Canada and the MIT SociotechnicalSystems Research Center Stokes Fund (to A.G.). This work used a modeling tool(US Regional Energy Policy) developed by the MIT Joint Program on the Scienceand Policy of Global Change, which is supported by a number of federal agenciesand a consortium of 40 industrial and foundation sponsors. A complete list ofsponsors is available at globalchange.mit.edu.

290 | www.pnas.org/cgi/doi/10.1073/pnas.1514395113 Giang and Selin

1. Rockström J, et al. (2009) A safe operating space for humanity. Nature 461(7263):472–475.

2. Selin H (2014) Global Environmental Law and Treaty-Making on Hazardous Sub-stances: The Minamata Convention and Mercury Abatement. Glob Environ Politics14(1):1–19.

3. Lamborg CH, et al. (2014) A global ocean inventory of anthropogenic mercury basedon water column measurements. Nature 512(7512):65–68.

4. Amos HM, et al. (2015) Observational and modeling constraints on global anthro-pogenic enrichment of mercury. Environ Sci Technol 49(7):4036–4047.

5. Driscoll CT, Mason RP, Chan HM, Jacob DJ, Pirrone N (2013) Mercury as a globalpollutant: Sources, pathways, and effects. Environ Sci Technol 47(10):4967–4983.

6. Grandjean P, Satoh H, Murata K, Eto K (2010) Adverse effects of methylmercury:Environmental health research implications. Environ Health Perspect 118(8):1137–1145.

7. Karagas MR, et al. (2012) Evidence on the human health effects of low-level meth-ylmercury exposure. Environ Health Perspect 120(6):799–806.

8. Sheehan MC, et al. (2014) Global methylmercury exposure from seafood consumptionand risk of developmental neurotoxicity: A systematic review. Bull World HealthOrgan 92(4):254–269F.

9. Roman HA, et al. (2011) Evaluation of the cardiovascular effects of methylmercuryexposures: Current evidence supports development of a dose-response function forregulatory benefits analysis. Environ Health Perspect 119(5):607–614.

10. Trasande L, Landrigan PJ, Schechter C (2005) Public health and economic conse-quences of methyl mercury toxicity to the developing brain. Environ Health Perspect113(5):590–596.

11. Spadaro JV, Rabl A (2008) Global health impacts and costs due to mercury emissions.Risk Anal 28(3):603–613.

12. Rice GE, Hammitt JK, Evans JS (2010) A probabilistic characterization of the healthbenefits of reducing methyl mercury intake in the United States. Environ Sci Technol44(13):5216–5224.

13. Bellanger M, et al.; DEMO/COPHES (2013) Economic benefits of methylmercury exposurecontrol in Europe: Monetary value of neurotoxicity prevention. Environ Health 12:3.

14. Griffiths C, McGartland A, Miller M (2007) A comparison of the monetized impact ofIQ decrements from mercury emissions. Environ Health Perspect 115(6):841–847.

15. US Environmental Protection Agency (2011) Regulatory Impact Analysis for the FinalMercury and Air Toxics Standards (US Environmental Protection Agency, Office of AirQuality and Planning Standards, Research Triangle Park, NC).

16. Rice GE, Hammitt JK (2005) Economic Valuation of Human Health Benefits of Con-trolling Mercury Emissions from U.S. Coal-Fired Power Plants. (Northeast States forCoordinated Air Use Management, Boston, MA).

17. Selin NE (2014) Global change and mercury cycling: Challenges for implementing aglobal mercury treaty. Environ Toxicol Chem 33(6):1202–1210.

18. United Nations Environment Programme. (2013) Global Mercury Assessment 2013:Sources, Emissions, Releases and Environmental Transport, (UNEP Chemicals Branch,Geneva, Switzerland).

19. Streets DG, Zhang Q, Wu Y (2009) Projections of global mercury emissions in 2050.Environ Sci Technol 43(8):2983–2988.

20. Giang A, Stokes LC, Streets DG, Corbitt ES, Selin NE (2015) Impacts of the minamataconvention on mercury emissions and global deposition from coal-fired power gen-eration in Asia. Environ Sci Technol 49(9):5326–5335.

21. US Environmental Protection Agency. (2011) IPM Analysis of the Final Mercury and AirToxics Standards (US Environmental Protection Agency, Clean Air Markets Division,Washington, DC).

22. Selin NE, et al. (2008) Global 3-D land-ocean-atmosphere model for mercury: Present-day versus preindustrial cycles and anthropogenic enrichment factors for deposition.Global Biogeochem Cycles 22(2):1–13.

23. Soerensen AL, et al. (2010) An improved global model for air-sea exchange of mer-cury: High concentrations over the North Atlantic. Environ Sci Technol 44(22):8574–8580.

24. Holmes CD, et al. (2010) Global atmospheric model for mercury including oxidation bybromine atoms. Atmos Chem Phys 10(24):12037–12057.

25. Amos HM, et al. (2012) Gas-particle partitioning of atmospheric Hg(II) and its effecton global mercury deposition. Atmos Chem Phys 12(1):591–603.

26. Zhang Y, et al. (2012) Nested-grid simulation of mercury over North America. AtmosChem Phys 12:6095–6111.

27. Corbitt ES, Jacob DJ, Holmes CD, Streets DG, Sunderland EM (2011) Global source-receptor relationships for mercury deposition under present-day and 2050 emissionsscenarios. Environ Sci Technol 45(24):10477–10484.

28. Amos HM, Jacob DJ, Streets DG, Sunderland EM (2013) Legacy impacts of all-timeanthropogenic emissions on the global mercury cycle. Global Biogeochem Cycles27(2):410–421.

29. Amos HM, et al. (2014) Global biogeochemical implications of mercury dischargesfrom rivers and sediment burial. Environ Sci Technol 48(16):9514–9522.

30. Sunderland EM, Mason RP (2007) Human impacts on open ocean mercury concen-trations. Global Biogeochem Cycles 21(4):GB4022.

31. Mason RP, et al. (2012) Mercury biogeochemical cycling in the ocean and policy im-plications. Environ Res 119:101–117.

32. Chen C, et al. (2012) Sources to Seafood: Mercury Pollution in the MarineEnvironment (Toxic Metals Superfund Research Program, Dartmouth College, Hano-ver, NH).

33. Harris RC, et al. (2007) Whole-ecosystem study shows rapid fish-mercury response tochanges in mercury deposition. Proc Natl Acad Sci USA 104(42):16586–16591.

34. Evers DC, et al. (2011) Mercury in the Great Lakes region: Bioaccumulation, spatio-temporal patterns, ecological risks, and policy. Ecotoxicology 20(7):1487–1499.

35. Drevnick PE, et al. (2012) Spatial and temporal patterns of mercury accumulation inlacustrine sediments across the Laurentian Great Lakes region. Environ Pollut 161:252–260.

36. Vijayaraghavan K, Levin L, Parker L, Yarwood G, Streets D (2014) Response of fishtissue mercury in a freshwater lake to local, regional, and global changes in mercuryemissions. Environ Toxicol Chem 33(6):1238–1247.

37. Knightes CD, Sunderland EM, Craig Barber M, Johnston JM, Ambrose RBJ (2009)Application of ecosystem-scale fate and bioaccumulation models to predict fishmercury response times to changes in atmospheric deposition. Environ Toxicol Chem28(4):881–893.

38. Selin NE, Sunderland EM, Knightes CD, Mason RP (2010) Sources of mercury exposurefor U.S. seafood consumers: Implications for policy. Environ Health Perspect 118(1):137–143.

39. Drenner RW, et al. (2013) Effects of mercury deposition and coniferous forests on themercury contamination of fish in the South Central United States. Environ Sci Technol47(3):1274–1279.

40. Mahaffey KR, Clickner RP, Jeffries RA (2009) Adult women’s blood mercury concen-trations vary regionally in the United States: Association with patterns of fish con-sumption (NHANES 1999-2004). Environ Health Perspect 117(1):47–53.

41. Sunderland EM (2007) Mercury exposure from domestic and imported estuarine andmarine fish in the U.S. seafood market. Environ Health Perspect 115(2):235–242.

42. US Environmental Protection Agency(2005) Regulatory Impact Analysis of the FinalClean Air Mercury Rule, (US Environmental Protection Agency, Office of Air QualityPlanning and Standards, Research Triangle Park, NC).

43. Karimi R, Fitzgerald TP, Fisher NS (2012) A quantitative synthesis of mercury incommercial seafood and implications for exposure in the United States. EnvironHealth Perspect 120(11):1512–1519.

44. Bell ML, Morgenstern RD, Harrington W (2011) Quantifying the human health ben-efits of air pollution policies: Review of recent studies and new directions in ac-countability research. Environ Sci Policy 14(4):357–368.

45. Rausch S, Metcalf GE, Reilly JM (2010) Distributional Implications of Alternative U.S.Greenhouse Gas Control Measures, (National Bureau of Economic Research WorkingPaper 16053, Cambridge, MA).

46. Saari RK, Selin NE, Rausch S, Thompson TM (2015) A self-consistent method to assessair quality co-benefits from U.S. climate policies. J Air Waste Manag Assoc 65(1):74–89.

47. Subir M, Ariya PA, Dastoor AP (2012) A review of the sources of uncertainties in at-mospheric mercury modeling II. Mercury surface and heterogeneous chemistry – Amissing link. Atmos Environ 46:1–10.

48. Landis MS, Ryan JV, ter Schure AFH, Laudal D (2014) Behavior of mercury emissionsfrom a commercial coal-fired power plant: The relationship between stack speciationand near-field plume measurements. Environ Sci Technol 48(22):13540–13548.

49. Tong Y, et al. (2014) Comparison of heterogeneous photolytic reduction of Hg(II) inthe coal fly ashes and synthetic aerosols. Atmos Res 138:324–329.

50. NRC (2000) Toxicological Effects of Methylmercury. (National Research Council, NationalAcademy Press, Washington, DC).

51. Basu N, Goodrich JM, Head J (2014) Ecogenetics of mercury: From genetic polymor-phisms and epigenetics to risk assessment and decision-making. Environ Toxicol Chem33(6):1248–1258.

52. Virtanen JK, et al. (2005) Mercury, fish oils, and risk of acute coronary events andcardiovascular disease, coronary heart disease, and all-cause mortality in men ineastern Finland. Arterioscler Thromb Vasc Biol 25(1):228–233.

53. RTI International. (2015) Environmental Benefits Mapping and Analysis Program -Community Edition: User’s Manual Appendices, (Research Triangle Park, NC).

54. Lamborg C, et al. (2014) Mercury in the Anthropocene Ocean. Oceanography (WashDC) 27(1):76–87.

55. Hsu-Kim H, Kucharzyk KH, Zhang T, Deshusses MA (2013) Mechanisms regulatingmercury bioavailability for methylating microorganisms in the aquatic environment:A critical review. Environ Sci Technol 47(6):2441–2456.

56. Sundseth K, Pacyna JM, Banel A, Pacyna EG, Rautio A (2015) Climate change impactson environmental and human exposure to mercury in the arctic. Int J Environ ResPublic Health 12(4):3579–3599.

57. Xue J, et al. (2015) Modeling tribal exposures to methyl mercury from fish con-sumption. Sci Total Environ 533:102–109.

58. Oken E, et al. (2012) Which fish should I eat? Perspectives influencing fish con-sumption choices. Environ Health Perspect 120(6):790–798.

59. Mozaffarian D, Rimm EB (2006) Fish intake, contaminants, and human health: Eval-uating the risks and the benefits. JAMA 296(15):1885–1899.

60. O’Neill CA (2004) Mercury, Risk, and Justice. Environ Law Rev 34(12):11070–11115.61. Jaffe DA, et al. (2014) Progress on understanding atmospheric mercury hampered by

uncertain measurements. Environ Sci Technol 48(13):7204–7206.62. Ranco DJ, O’Neill CA, Donatuto J, Harper BL (2011) Environmental Justice, American

Indians and the Cultural Dilemma: Developing Environmental Management for TribalHealth and Well-being. Environ Justice 4(4):221–230.

Giang and Selin PNAS | January 12, 2016 | vol. 113 | no. 2 | 291

SUST

AINABILITY

SCIENCE

EnvironmentalScienceProcesses & Impacts

PAPER

Publ

ishe

d on

23

Janu

ary

2018

. Dow

nloa

ded

on 4

/17/

2019

6:0

0:29

PM

.

View Article OnlineView Journal | View Issue

Responses of de

aCivil and Environmental Engineering

University, Houghton, MI 49931, USA. E-mabInstitute for Data, Systems, and Society, D

Planetary Sciences, Massachusetts Institute

USAcGeological and Mining Engineering a

Technological University, Houghton, MI 499dSocial Sciences Department, Michigan Te

49931, USA

Cite this: Environ. Sci.: ProcessesImpacts, 2018, 20, 195

Received 13th November 2017Accepted 2nd January 2018

DOI: 10.1039/c7em00547d

rsc.li/espi

This journal is © The Royal Society of C

position and bioaccumulation inthe Great Lakes region to policy and other large-scale drivers of mercury emissions†

J. A. Perlinger, *a N. R. Urban,a A. Giang, ‡b N. E. Selin, b A. N. Hendricks,a

H. Zhang,§c A. Kumar,c S. Wu,c V. S. Gagnon, d H. S. Gormand and E. S. Norman e

Mercury (Hg) emissions pose a global problem that requires global cooperation for a solution. However,

neither emissions nor regulations are uniform world-wide, and hence the impacts of regulations are also

likely to vary regionally. We report here an approach to model the effectiveness of regulations at

different scales (local, regional, global) in reducing Hg deposition and fish Hg concentrations in the

Laurentian Great Lakes (GL) region. The potential effects of global change on deposition are also

modeled. We focus on one of the most vulnerable communities within the region, an Indigenous tribe in

Michigan's Upper Peninsula (UP) with a high fish consumption rate. For the GL region, elements of global

change (climate, biomass burning, land use) are projected to have modest impacts (<5% change from

the year 2000) on Hg deposition. For this region, our estimate of the effects of elimination of

anthropogenic emissions is a 70% decrease in deposition, while our minimal regulation scenario

increases emissions by 35%. Existing policies have the potential to reduce deposition by 20% with most

of the reduction attributable to U.S. policies. Local policies within the Great Lakes region show little

effect, and global policy as embedded in the Minamata Convention is projected to decrease deposition

by approximately 2.8%. Even within the GL region, effects of policy are not uniform; areas close to

emission sources (Illinois, Indiana, Ohio, Pennsylvania) experience larger decreases in deposition than

other areas including Michigan's UP. The UP landscape is highly sensitive to Hg deposition, with nearly

80% of lakes estimated to be impaired. Sensitivity to mercury is caused primarily by the region's

abundant wetlands. None of the modeled policy scenarios are projected to reduce fish Hg

concentrations to the target that would be safe for the local tribe. Regions like Michigan's UP that are

highly sensitive to mercury deposition and that will see little reduction in deposition due to regulations

require more aggressive policies to reduce emissions to achieve recovery. We highlight scientific

uncertainties that continue to limit our ability to accurately predict fish Hg changes over time.

Environmental signicance

Here, we project the responses within the Great Lakes region of rates of atmospheric deposition and shmercury concentrations to changes in policy at multiplejurisdictional scales and to global change. We demonstrate that even within this region, responses to policy and global change drivers vary spatially with thelargest reductions in deposition occurring in closer proximity to major Hg emission sources. Projected responses to atmospheric deposition are greater for U.S.policies than for either local or global Hg control policies. We focus further on the environmental justice implications of policy changes for sh mercurycontamination facing one of the most vulnerable communities within the region, a tribe in Michigan's Upper Peninsula with a high sh consumption rate. TheUpper Peninsula is shown to be a highly sensitive landscape that readily converts atmospherically deposited Hg to bioaccumulated methylmercury, but one thatwill respond little to any of the policies evaluated. Thoughtful and more aggressive policy changes would be required to alleviate the existing environmentalinjustice.

Department, Michigan Technological

il: jperl@mtu.edu

epartment of Earth, Atmospheric and

of Technology, Cambridge, MA 02139,

nd Science Department, Michigan

31, USA

chnological University, Houghton, MI

eNative Environmental Science Department, Northwest Indian College, Bellingham,

WA 98226, USA

† Electronic supplementary information (ESI) available. See DOI:10.1039/c7em00547d

‡ Currently at the Institute for Resources, Environment and Sustainability andDepartment of Mechanical Engineering, University of British Columbia,Vancouver, BC V6T 1Z4

§ Currently at the Center for Global and Regional Environmental Research,University of Iowa, Iowa City, IA, 52242, USA.

hemistry 2018 Environ. Sci.: Processes Impacts, 2018, 20, 195–209 | 195

Environmental Science: Processes & Impacts Paper

Publ

ishe

d on

23

Janu

ary

2018

. Dow

nloa

ded

on 4

/17/

2019

6:0

0:29

PM

. View Article Online

1. Introduction

Mercury (Hg) is a toxic pollutant that, when emitted to the envi-ronment, undergoes volatilization, long-range transport anddeposition in its oxidized form, transformation to its more toxicform, methylmercury (MeHg), and bioaccumulation in ecosys-tems. It is at this point that Hg poses threats to human healththrough consumption of sh containing high levels of MeHg. Inthe Laurentian Great Lakes (GL) region, sh tissue concentrationsare unsafe for GL residents as evidenced by statewide shconsumption advisories.1–3 Fishing communities are burdenedwith the majority of negative impacts,4–8 and Native Americanshing rights and cultures, in particular, have been severelyimpacted by Hg contamination;4,7,9–12 thus Hg contamination is anenvironmental and social justice issue.Most stakeholders involvedin the effort to reduce Hg contamination recognize that action isneeded at the global as well as the regional level. Toward this end,nations have signed and ratied the Minamata Convention onmercury, which sets expectations associated with controls onemissions of Hg.13 A challenge now is to estimate when it will bepossible to safely consume sh in places such as the GL region asnations implement the Minamata Convention. In partnershipwith the Keweenaw Bay Indian Community (KBIC), an Indigenoustribe in Michigan's Upper Peninsula (UP), we formulated14 andconducted an integrated assessment to address the question, “Canwe, by 2050, meet the goal of safe sh consumption?”

Because of the atmospheric residence times of different Hgspecies, the answer to this question is location dependent. Muchof the Hg deposition to forests is via dry deposition of Hg0 whichbecause of its long atmospheric residence time (2.7–12months)15,16 can be due to either regional or global sources. Incontrast, deposition (wet and dry) of the much shorter lived (0.5–2 days)15 Hg(II) species is affected primarily by local and regionalemissions. Understanding the importance of Hg deposition fromlocal and regional sources in the U.S. (the largest fraction ofwhich comprise coal-red power plants), in particular within theGL region, continues to evolve. In the 1980s some thought thatatmospheric residence times of Hg were long (12–18 months),17

and that atmospheric Hg deposition was due to global-scaleprocesses. In 1992 it was shown that local and regional anthro-pogenic sources of Hg in the GL region could determine Hgconcentrations in the litter and mineral soil horizons;18 thoseconcentrations paralleled changes in wet sulfate deposition andhuman activity. In 2004, using a combination of a global chem-ical transport model (CTM) and a nested regional CTM, Seigneuret al.19 showed that Hg deposition in the contiguous U.S. was, onaverage, comprised of�30% emissions fromNorth America.19 Byconstraining the estimates of wet deposition by a global transportmodel (GEOS-Chem) with measurements, Selin and Jacob20

estimated that North American anthropogenic emissionscontributed 50% of total deposition in the Midwest and EasternU.S. Cohen et al.21 used trajectory analysis (HYSPLIT) to discerndifferences in emission sources of Hg to the ve GL. They foundthat Lake Erie, which is downwind of signicant local/regionalemission sources, was the most impacted by local and regionalemissions (58% of the base case total deposition), while Lake

196 | Environ. Sci.: Processes Impacts, 2018, 20, 195–209

Superior, with the fewest upwind local/regional sources, was theleast impacted by these sources (27%). Recently, Zhou et al.showed that decreased atmospheric Hg concentrationsmeasured during the past decade in the northeastern U.S. areconsistent with decreased Hg emissions from regional pointsources, and that increasing global emissions have not over-whelmed those decreases.22 Potential effects of local, regionaland global regulation of Hg emissions on deposition across theGL region are investigated in this paper.

The concept of landscape sensitivity to Hg deposition is wellestablished in the literature.23–25 It is clear that landscapes withabundant forests and wetlands, as well as low nutrients andalkalinity in runoff are predisposed to having high MeHgconcentrations in lakes and sh.24,26 Such regions require lowerrates of atmospheric Hg deposition than other regions to avoidhighMeHg concentrations in sh. Studies also suggest that someof these regions may respond more slowly to decreases in Hgdeposition because the Hg accumulated in organic-rich soils maycontinue to be mobilized to lakes by the high and potentiallyincreasing27–29 concentrations of dissolved organic matter inrunoff typical of these regions. Multiple studies have shown thatsome lakes can recover quickly from Hg contamination30–32 orreduced Hg or sulfate deposition.33–37 However, modelingstudies,38–40 large surveys36 and synthesis studies41 revealedconsiderable variability among individual lakes that is mediatedby biogeochemical conditions in a lake and/or its watershed aswell as by the mechanism for Hg input. In northern Wisconsin,a seepage lake with little catchment showed a rapid decline inwater and sh Hg over the same time that a drainage lake withconsiderable wetland in the catchment showed a decline in totalHg but little change in MeHg.42 Within Michigan, concentrationsof Hg in multiple lakes and rivers increased from 1998 to 2009while atmospheric deposition of Hg and sulfate decreased.43 InEngland, the leaching of Hg from catchment soils continues toprolong lake recovery.44 The combination of variability amonglakes and our level of understanding of key watershed featuresresults in a wide range (a few years to centuries) in model-predicted times for lake recovery.45–47 In this paper, we high-light spatial variability in the landscape that results in differentresponses of lakes to Hg deposition.

The objectives of this paper are to present an approach toestimate effects of policy and global changes on Hg concen-trations in sh, to apply this approach to the GL region, toexamine spatial differences in response to global and policychange, and to evaluate implications of the predicted lakeresponses particularly for the inhabitants of Michigan's UpperPeninsula. We restrict our regional scale comparisons tochanges in atmospheric deposition and landscape sensitivity,two of the primary determinants of bioaccumulation in lakes.We evaluate lake responses and their implications on a localscale, focusing on Michigan's UP.

2. Experimental2.1 Description of study area

This study contrasted forcings (global change, policies) andresponses on scales ranging from global to regional (hundreds

This journal is © The Royal Society of Chemistry 2018

Paper Environmental Science: Processes & Impacts

Publ

ishe

d on

23

Janu

ary

2018

. Dow

nloa

ded

on 4

/17/

2019

6:0

0:29

PM

. View Article Online

to a few thousand kilometers) to local (tens to a few hundredkilometers). We present here only the regional (Great Lakesregion) and local (UP, Adirondack area) scale responses(atmospheric deposition of Hg, sh Hg concentrations).We dened the Great Lakes region as bounded by 40� and50� N latitude and by 73� to 95� W longitude. The UP (44 � 103

km2) is dened spatially by the land border betweenWisconsin and Michigan to the west, Lake Superior to thenorth, and Lake Michigan–Huron to the south and east. The11.3 � 103 km2 Adirondack area dened for this project wasthe rectangle bounded south to north by 43.7 to 44.3� N andeast to west by 73.9 to 75.3� W. Both local areas are largelycovered with mixed deciduous-coniferous forests (UP – 77%,Adirondacks – 74%; National Land Cover Database101), andboth have a high density of lakes (UP – 3.8% of area; Adir-ondacks – 5.1% of area). The UP has more wetlands (29% vs.11%; National Wetland Inventory), while the Adirondackshave higher elevation, steeper topography, and higher rates ofsulfate deposition.

2.2 Modeling approach to estimate atmospheric depositionand sh Hg responses to global change and emissions policies

In general terms, our approach to estimating lake responses toforcings was to construct multiple scenarios embodyingdifferent regulatory approaches or global change conditions, toestimate Hg emissions for these scenarios, to use GEOS-Chemto predict the spatial distribution of Hg deposition rateswithin the Great Lakes region, and then to utilize the predictedHg deposition rates in a model of watershed and lake Hgcycling. Scenarios used the approximate year 2000 as the“present day” situation with which to compare alternate emis-sion scenarios. To obtain a signicant difference from thepresent but avoid projecting too far into an uncertain future, weselected the year 2050 as the common future time at whichscenarios would be compared. Because Hg emissions anddistributions are inuenced by other drivers of global change(climate change, land-use/cover, biomass burning), we createdscenarios to estimate the changes in Hg deposition caused bythese forcings. For all scenarios, we evaluated the predictedspatial changes in Hg deposition within the Great Lakes region,and for selected policy scenarios we contrasted atmosphericdeposition in two local areas to highlight spatial variabilitywithin the region.

2.2.a Target MeHg concentration in sh. To contextualizethe current and predicted mercury concentrations in sh,project social scientists engaged our community partner, theKBIC, as described in detail by Gagnon et al.14 The high shconsumption rate of this tribal community renders it one ofthe most vulnerable subpopulations within the Great Lakesregion. The community partners dened ‘desired’ walleyeconsumption as two 8-oz. (227 gram) meals per day, or a totalof 0.454 kg sh per day. This represents a rate typical of thespring walleye harvest, when regional shing activities arehighest.14 This sh consumption rate was entered into theequation to compute a target MeHg concentration in sh asfollows.

This journal is © The Royal Society of Chemistry 2018

Fish MeHg Conc:

�mg

kg wet wt: fish

�

¼ RfD

�mg

kg� d

�� body wt:ðkgÞ

� 1

fish consumption rate�kg wet wt: fish

d

�

¼�0:1

mg

kg� d

�� 70 kg� 1

0:454kg wet wt: fish

d

¼ 15mg

kg wet wt: fish¼ 15 ppb ¼ 0:015 ppm

where, for illustration purposes, we use a body weight of 70 kgto be consistent with the sh consumption rate, and, asa benchmark, the U.S. Environmental Protection Agency (EPA)

reference dose48 RfD of 0:1mg

kg � d. We acknowledge that recent

studies have identied mercury impacts at levels that are lowerthan the studies used to set the EPA's guideline,49 and thatadvocacy groups have called for lowering this limit.50 The targetconcentration provided a benchmark against which to judge theeffects of policy scenarios, but clearly the target varies with thetarget population, is not uniform across the region and willevolve as knowledge of health impacts improves.

2.2.b Formulation of global change scenarios. We useda scenario approach to consider the potential impact of fourglobal change drivers on future atmospheric deposition andsh tissue concentrations of Hg in the GL region: policy, climatechange, land-use and land-cover change, and biomass burning.Although there are interactions between these four drivers, weevaluated the impact of each separately, holding the otherfactors constant, to better isolate inuences on Hg biogeo-chemical cycling. These scenarios draw from previous publica-tions, so they are described only briey and summarized inTable 1.

2.2.c Policy scenarios. To better understand the impact ofanthropogenic forcing on future atmospheric deposition andsh tissue concentrations, we considered present day emis-sions, and three future emissions scenarios capturing a range ofpolicy ambition: aspirational, policy-in-action, and minimal-regulation. We implemented the anthropogenic emissionscenarios dened below in spatially resolved emissions inven-tories that are described fully by Corbitt et al.58 and Giang andSelin.51 Our present day, turn-of-the-21st-century emissions arebased on year 2005 for the U.S. and Canada,51,59 and 2006emissions elsewhere in the world.58,60

Aspirational. Our aspirational scenario assumed cessation ofanthropogenic emissions of Hg in 2050. Preventing theseemissions would require transitions to Hg-free alternatives (e.g.,renewable energy sources such as wind and solar, LED lightbulbs), and advanced capture and control technologies, some ofwhichmay not currently exist. In this scenario and all others, weassume that legacy emissions remain at their year 2000 value, tofacilitate comparison of the direct anthropogenic component.

Policy-in-action. This scenario assumed that by 2050, recentpolicy proposals targeting Hg emissions are fully implemented.

Environ. Sci.: Processes Impacts, 2018, 20, 195–209 | 197

Table 1 Summary of global change scenarios

Driver Scenario name Description

Policy51 2000s Anthropogenic emissions for 2005 over the U.S. and Canada and for 2006elsewhere in the world are used as representative present day emissions at the turnof the 21st century.

Future: aspirational Assumed cessation of anthropogenic emissions of Hg in 2050.Future: policy-in-action Recent policy proposals targeting Hg emissions (in the GL, in the U.S., and

globally) are assumed to be fully implemented by 2050.Future: minimal regulation Limited global progress on Hg and other sustainability goals are assumed by 2050.

Climate change52,53 2000s Meteorology simulated with the NASA GISS GCM. An average of simulated 1998–2002 meteorology is used to represent present, turn of the 21st century climate.

Future NASA GISS GCMused with greenhouse gas trends following the IPCC A1B scenario.An average of simulated 2048–2052 meteorology, is used to represent mid-centuryclimate.

Land-use and land-coverchange54,55

2000s Land use simulated by the IMAGE model. Land cover (1998–2002) simulated usingthe LPJ dynamic global vegetation model driven with the GISS GCM meteorologyand CO2 concentrations.

Future Future (2050s) represents the period 2048–2052. Changes in land use follow theIPCC A1B scenario. Land cover distributions generated from the LPJ model withGISS GCM meteorology and CO2 concentrations following the IPCC A1B scenario.

Biomass burning56,57 2000s Biomass burning emissions estimated from a re emissions model. Averageemissions for the period 1998–2002 representing present day are used.

Future Refers to the 2050s. Fire emissions model driven with 2050s re frequencies, landuse-land cover distributions, meteorology and burned area following the IPCC A1Bscenario. Average emissions for the period 2048–2052 are used.

Environmental Science: Processes & Impacts Paper

Publ

ishe

d on

23

Janu

ary

2018

. Dow

nloa

ded

on 4

/17/

2019

6:0

0:29

PM

. View Article Online

This scenario was developed at three nested geographicdomains: local (Lake Superior Basin), regional (U.S.), andglobal. The global and regional scenarios are described fully inGiang and Selin.51 Globally, we assumed the recently adoptedUnited Nations Minamata Convention on mercury is fully inforce in the global community.13 Moreover, we assumed someglobal cooperation on broad sustainability issues like climatechange, following the IPCC SRES B1 scenario, as described byStreets et al.60 Current projections based on assessments oftechnology and energy trends suggest that with mitigationefforts consistent with the Minamata Convention text andtransitions away from coal combustion, emissions in 2050could be stabilized roughly at current-day levels.60–62 Additionalbenets from more ambitious climate action could result infurther reductions, though we do not consider those here.63

Regionally, within the U.S., we assumed that regulations tar-geting emissions under the Clean Air Act are in place andenforced through 2050, including the Mercury and Air ToxicsStandards (MATS).51 Locally, in the Lake Superior Basin, weassumed that zero discharge goals have been achieved by2050.64

Minimal-regulation. In our minimal-regulation scenario, wemodeled limited global progress on Hg and other sustainabilitygoals by 2050. We assumed limited advances in the efficacy ofair pollution control technologies, limited adoption of existingcontrol technologies globally, and continued global growth incoal combustion consistent with the IPCC SRES A1B scenario,as described by Streets et al.60 Under these minimal-regulationconditions, Hg emissions in 2050 are close to double those ofpresent day levels.60 Additional details for this scenario areprovided by Giang and Selin.51

198 | Environ. Sci.: Processes Impacts, 2018, 20, 195–209

2.2.d GEOS-Chem simulation of atmospheric deposition oftotal Hg. The GEOS-Chem coupled atmosphere-land-surfaceocean Hg model, version v9-02, was used to simulate theimpact of changes in emissions (anthropogenic, biomassburning), land use/land cover change, and climate on atmo-spheric Hg deposition to the GL.65–68 Additional details onmodel and simulation design are given by Zhang et al.,55 Giangand Selin,51 and Kumar et al.57

For anthropogenic emissions scenarios under all policyoptions, our GEOS-Chem simulations are driven with assimi-lated meteorology from the NASA Goddard Earth ObservingSystem (GEOS-5). Using this meteorology allowed us to simulateHg chemistry and transport at a global resolution of 4� � 5�,and at a ner 1/2� � 2/3� resolution over North America througha one-way nested grid model.59 For each of the four anthropo-genic emissions scenarios for 2050 dened above, we simulatedmeteorological years 2007–2011, using the rst two years asinitialization. By holding meteorology constant, we isolated theeffect of emissions changes alone. Results presented are inter-annual averages for meteorological years 2009–2011.

To examine the impacts from climate change and land use/land cover change in the coming decades, we carried out modelsimulations under various scenarios for climate, land use andland cover. For example, we compared simulation results withthe present (year 2000) and future (mid-century: 2050) climateto derive the potential impact of climate change on Hg depo-sition to the GL region. More details about the scenarios can befound in Zhang et al.55 All GEOS-Chem simulations were drivenby meteorological elds from the NASA Goddard Institute forSpace Studies (GISS) General Circulation Model (GCM 3).69 Thecoupling between GISS GCM3 and GEOS-Chem model is

This journal is © The Royal Society of Chemistry 2018

Fig. 1 Schematic of mercury processes and pathways incorporated inthe lake and watershed mercury model. The watershed is semi-distributed spatially to account for categories of land use (wetland,forest, urban). Results presented here assume runoff coefficients foreach land use are fractions of annual deposition. In addition to thecycling of mercury, the model incorporates seasonal variations intemperature, lake mixing, DOC inputs, and phytoplankton abundance.

Table 2 Summary of characteristics of the modeled lake. Data frommixed sources as summarized by Hendricks86

Parameter Value or range

Surface area 9.7 � 106 m2

Volume 1.4 � 108 m3

Mean depth 15 mWatershed area 1.9 � 108 m2

Wetland area 2.7 � 107 m2

pH 7.55DOC 7.4 mg L�1

Lake temperature 1.2–18.1 �CWater residence time 1 yearSediment Hg 200–400 mg g�1

Lake Hg 0.6–0.8 ng L�1

Paper Environmental Science: Processes & Impacts

Publ

ishe

d on

23

Janu

ary

2018

. Dow

nloa

ded

on 4

/17/

2019

6:0

0:29

PM

. View Article Online

described by Wu et al.52 We simulated meteorological years1998–2002 for the present day and 2048–2052 for the futurescenario at 4� � 5� horizontal resolution globally. We analyzethe results based on ve-year averages.

We also accounted for the potential impacts of global change(including changes in population, climate, land use and landcover) on Hg emissions from wildres and consequently on Hgdeposition. Changes in land use/land cover can affect vegeta-tion type/availability while climate is an important determinantof conditions favorable for re occurrence and natural ignitions(lightning). In addition, human interactions with naturalvegetation (e.g. use of re for vegetation clearing) are a source ofanthropogenic ignitions for wildres while re suppression isgreater in areas of high population density. We took advantageof a recently developed re emission model for Hg.57 Wecalculated the Hg emissions from wildres for the present day(2000s) and future (2050s), respectively, and implemented theseemissions in the GEOS-Chem model to examine the perturba-tions to atmospheric Hg burden and deposition.

The GEOS-Chem model as used here represents impactsfrom near-term changes in emissions, but does not capturealterations in longer-term legacy biogeochemical cyclingchanges in response to emissions changes, that were heldconstant in all of our scenarios as noted above. The potentialimplications of these changes that can impact legacy emissionsare discussed further below.

2.2.e Lake mass balance modeling. The mass balancemodel for Hg of Hendricks et al.70 was used to predict aqueousconcentrations based on the characteristics of a lake (depth,water retention time, trophic state) and its watershed (area, landcover and wetland extent), and the local rates of Hg atmosphericdeposition (Fig. 1). Hg species included elemental, divalent, andMeHg in the lake compartments epilimnion, hypolimnion, andsediments; this resulted in a system of nine ordinary differentialequations (ODE) that was solved using a numerical ordinarydifferential equation solver package for R.71 The model wasbased loosely on EPA's spreadsheet model SERAFM,72 but it wasadjusted for non-steady state conditions to enable prediction oflake responses to changes in loadings. Chemical and physicalprocesses of Hg in the lake that were modeled included redoxreactions of divalent and elemental Hg, methylation, deme-thylation, photodemethylation, thermocline dispersion, parti-tioning (phases included DOC, abiotic solids, biotic solids, andsediments), diffusion in sediments, settling, burial, and resus-pension. Seasonal changes in lake temperature, stratication,and ice cover were simulated. Redox reactions, methylation,and demethylation were corrected for these changes intemperature of the lake. Photodemethylation was corrected forthe amount of light attenuation received in the water column.The rate of thermocline dispersion changed seasonally to allowfor mixing and stratication in this dimictic lake. Atmosphericdeposition during ice cover “accumulated” on top of the iceuntil spring melt when it entered the lake. The model does notinclude sh population dynamics; rather, it uses bio-accumulation factors for mixed feeders and piscivorous shtaken from Knightes.73 Hence, the model ignores the 3–7 yearsrequired for sh populations to come to steady state when

This journal is © The Royal Society of Chemistry 2018

loadings change.74,75 In the results presented in this paper,uncertainties reect only the distribution of bioconcentrationfactors, rather than the total error calculated via Bayesiantechniques by Hendricks.70 This model was validated for a lakein Michigan's UP for which Hg concentrations had beenmeasured in the water, sediments and sh. Characteristics ofthis lake are summarized in Table 2.

The model requires inputs of atmospheric concentrations ofHg(0), particulate bound mercury (PBM), MeHg, and reactivegaseous Hg (RGM) to calculate deposition to the lake as well aswet and dry deposition uxes of all species to the watershed. Incontrast to many previous studies,76–78 we did not assume thatdry deposition would be the same to lakes as to their water-sheds. The theoretical basis for our assumption is well

Environ. Sci.: Processes Impacts, 2018, 20, 195–209 | 199

Environmental Science: Processes & Impacts Paper

Publ

ishe

d on

23

Janu

ary

2018

. Dow

nloa

ded

on 4

/17/

2019

6:0

0:29

PM

. View Article Online

established. The roughness length for lakes is smaller than fortree canopies resulting in a higher friction velocity.79,80 Alter-natively, the difference may be understood in terms of theresistance components to deposition; for tree canopies thesetypically include surface (stomatal, cuticular), aerodynamic,and quasi-laminar boundary resistances81 while water hasdifferent surface components. Additionally, in small lakes air–water gas exchange is facilitated by convective mixing such thattransfer velocities are faster than those estimated based onboundary layer or surface renewal models.82 To estimate drydeposition to lakes and watersheds, we used the Hg species- andlandform-specic deposition velocities in GEOS-Chem.67 Gasexchange with the lake was estimated using deposition veloci-ties for RGM and MeHg, and with a boundary layer-basedtransfer velocity for Hg0.70 Concentrations of inorganic Hgspecies were taken from the output of GEOS-Chem; MeHgconcentrations in wet deposition were taken as a ratio toconcentrations of total Hg as reported in the literature.83 Ourpredicted wet deposition rates compared favorably with nearbymeasurements by the Mercury Deposition Network, and ourestimates of dry deposition to the watershed were consistentwith nearby measured Hg uxes in litterfall.84,85

The model also requires inputs from the watershed to thelake of Hg species and dissolved organic carbon (DOC). Inputsof DOC are modeled as a sine function to reect seasonalvariability; the mean and amplitude are specic to the lakebeing modeled. It has been reported in the literature that ratesof total Hg runoff from catchments to lakes range from 1–35%of the rate of atmospheric Hg deposition.76,87,88 Multiplemodels of watershed Hg runoff of varying complexity areavailable,89,90 but inter-model comparison has shown largediscrepancies in predictions.91 It is also clear that there may beconsiderable delays between deposition to a catchment andrunoff to streams and lakes.41,92 The approach of setting the Hgrunoff from catchments as proportional to atmosphericdeposition72 assumes that only recently deposited Hg issusceptible to leaching; this approach yields a fast lakeresponse to changes in Hg deposition. We use a semi-distributed spatial model that multiplies runoff coefficientsfor areas of land use categories by annual Hg deposition ratesto estimate the maximum possible lake response to changes inatmospheric deposition. This approach allows us to say if it isnot possible to reach safe sh concentrations by 2050.Specically, for divalent Hg we use a runoff coefficient of 10%for “sensitive upland landscapes”, and a value of 5% for lesssensitive upland regions. Coefficients of 20% are used forrunoff of divalent Hg from wetlands in all areas, and a coeffi-cient of 4.9 is used for MeHg runoff from wetlands.73 Urbanand agricultural areas were negligible in the modeled water-sheds. Multiple factors (hydrology, owpaths, soil character-istics) regulate runoff of Hg from upland soils;89,92–94 because itis widely observed that Hg runoff is proportional to runoff orconcentrations of OC (both particulate and dissolved),87,89 andbecause runoff of DOC from forest soils is reported to be highin the UP95 the runoff coefficient for Hg from upland soils wasset at 10% of atmospheric deposition rates based on the ratiosreported by Babiarz et al.96

200 | Environ. Sci.: Processes Impacts, 2018, 20, 195–209

Responses of sh Hg concentrations to scenarios of policychange were evaluated, but not to scenarios of change ofglobal climate, land use, and biomass burning. Modeling ofatmospheric deposition was performed only for the presentand the year 2050. To determine the maximum lake changelikely under the chosen scenarios, we ran the lake model forthe 2050 conditions. Because the Hg residence time in themodeled lake (excluding sh) is short (�2 months),70 the lakeresponds quickly to changes in inputs. While the sedimentstake longer to reach steady state (20–60 years), low organicmatter content of the study lake results in low in-lake meth-ylation and a fast response in this lake to changes in inputs. Asmentioned, we do not include the time for Hg to stabilize inthe food web, and to estimate the maximum possible lakeresponse we assumed a rapid response of the watershed toalterations in deposition.

2.2.f Virtual experiments: effect of landscape sensitivity.Fish accumulation of Hg is a response to the interplay of rates ofatmospheric deposition with watershed (i.e., landscape sensi-tivity) and lake characteristics. To highlight the effect of land-scape sensitivity on lake response to deposition, we performedtwo virtual experiments. First, we compared predicted sh Hgconcentrations for a UP lake under two regimes of atmosphericdeposition (policy-in-action deposition for the UP and Adir-ondacks). Second, we also changed the watershed conditions tothose typical of the Adirondacks area. To clarify the reasons fordifferent lake responses in the two regions, we presenta summary of lake and watershed characteristics and sh Hgconcentrations in the two locations. The primary data sourcefor sh Hg in the UP was Michigan's Fish Contaminant Moni-toring Program (http://www.michigan.gov/deq/0,4561,7-135-3313_3681_3686_3728-32393–,00.html); only concentrations inwalleye are summarized here. To eliminate sh size as a causeof variability among lakes, regressions of walleye Hg vs. walleyelength were performed for each lake, and concentrations ata common length of 42 cm (the median size of all walleyesampled) were calculated and are reported here. Data sourcesfor water quality included Michigan Surface Water InformationManagement System (http://www.mcgi.state.mi.us/miswims/),EPA National Lake Assessment (https://www.epa.gov/national-aquatic-resource-surveys/data-national-aquatic-resource-surveys),and a variety of reports and published papers.97–100 Althoughmany data sources are available for the Adirondacks, weused only data presented by Yu et al.24 Data on the abun-dance of lakes and wetland areas in lake catchments for bothregions were gathered from the National Wetland Inventory(https://www.fws.gov/wetlands/Data/State-Downloads), andforest cover was taken from the National Land CoverDatabase.101

3. Results3.1 Atmospheric deposition of Hg to the GL region,Michigan's UP, and New York's Adirondacks region in thepresent vs. 2050

Changes in total (wet + dry) Hg deposition estimated fromGEOS-Chem simulations from the present to the future (2050)

This journal is © The Royal Society of Chemistry 2018

Tab

le3

Chan

gein

totalH

gatmosp

hericdepositionfrom

thepresentto

2050forthegivenregionan

dscenario,a

nd%

contributionofagivenscaleregionto

thepolic

y-in-actionscenario

Region

Scen

ario

Great

Lake

sMichigan

Upp

erPe

ninsu

laNew

YorkAdironda

cks

Increase

(+)or

decrease

(�)

intotalHg

depo

sition

(%)

%contribution

topo

licy-in

-action

scen

ario

Increase

(+)or

decrease

(�)in

totalHg

depo

sition

(%)

%contributionto

policy-in

-action

scen

ario

Increase

(+)or

decrease

(�)

intotalHgde

position

(%)

%contribution

topo

licy-in

-actionscen

ario

2050

aspiration

al�7

0�6

5�7

820

50po

licy-in

-action

�20

�15

�29

World

1425

0.1

U.S.

8570

97La

keSu

perior

15

2.9

2050

minim

al-regulation

+35

+34

2050

clim

atech

ange

+3.8

+5.2

2050

landus

ech

ange

�0.2

+0.7

2050

biom

assbu

rning

+1.9

+2.3

Paper Environmental Science: Processes & Impacts

Publ

ishe

d on

23

Janu

ary

2018

. Dow

nloa

ded

on 4

/17/

2019

6:0

0:29

PM

. View Article Online

are summarized in Table 3, along with % contributions to thepolicy-in-action scenario from the three policy scales (global,U.S., or Lake Superior). Total deposition is estimated todecrease by 70% due to the aspirational policy change scenarioin the GL region (Fig. S1†), while it is estimated to decrease by65% in Michigan's Upper Peninsula (UP) and 78% in the NewYork Adirondack region. These large estimated decreases arereasonable given that for the aspirational scenario, all anthro-pogenic emissions in the 2050 simulation were turned off.Because legacy emissions are held constant, this scenario doesnot represent pre-industrial emissions.

The combined policy-in-action scenario leads to 20%, 15%,and 29% decreases in atmospheric deposition in 2050 in theGL, Michigan's UP, and New York Adirondack regions, respec-tively (Table 3). Fig. 2 presents the effects of individual scales ofpolicies (Lake Superior, U.S., and global) on the combinedpolicy-in-action deposition benet for the GL region. Approxi-mately 85% of the total decrease is due to U.S. Clean Air Actregulations including MATS, while the local (Lake SuperiorBasin) and global (Minamata Convention) policies each accountfor 1% and 14% of the total decrease, respectively. Fig. 2Bdemonstrates the effect of reducing U.S. emissions on GLregional deposition. The two times greater reduction in depo-sition in the Adirondacks compared to the UP for this scenariois therefore largely related to differences in the inuence of U.S.policies.

Compared to the GL region policy-in-action scenariocontributions of 14% from global policies, 85% from U.S., and1% from Lake Superior, the UP receives 25% of total depositiondecrease from global sources, 70% from the U.S., and 5% fromLake Superior (Table 3). In the Adirondacks region, approxi-mately 97% of the total decrease in deposition for this scenariois due to Clean Air Act regulations in the U.S., while local (LakeSuperior Basin) and global (Minamata Convention) eachaccount for 2.9% and 0.1% of the total decrease, respectively.These values clearly show the large differences in contributionsthat proximity to up-wind sources make. The global contribu-tion (as a percentage) of emissions reductions to decreasedatmospheric deposition is over two orders of magnitude greaterfor the UP than for the Adirondacks (Table 3). Sources withinthe U.S. (mostly coal-red power plants) provide a greater frac-tion of the deposition to the Adirondacks as compared to theUP.

The minimal-regulation scenario leads to increases in totaldeposition of 35% and 34% in the GL (Fig. S2†) and Michigan'sUP region, respectively (Table 3). The similarity in these esti-mates is likely coincidental. Based on the policy-in-actionscenario, differences in deposition benet are largely relatedto U.S.-scale changes in emissions, and the largely U.S. emis-sions that comprise the minimal-regulation total depositionestimates happen to lead to similar increases in deposition.

The climate change, land use/land cover change, andbiomass burning scenarios lead to smaller total Hg depositionchanges in 2050 as compared to the policy scenarios (Table 3).In the GL region, these estimates are 3.8% (Fig. S3†), �0%(Fig. S4†), and 1.9% (Fig. S5†), respectively, whereas for Michi-gan's UP, the changes are 5.2%, �0%, and 2.3%, respectively.

This journal is © The Royal Society of Chemistry 2018 Environ. Sci.: Processes Impacts, 2018, 20, 195–209 | 201

Fig. 2 Lake Superior (A), U.S. (B), global (C), and combined (D) policy total Hg deposition benefits from present day. The panels on the leftrepresent the policy benefits in Dmg per m2 per year, while those on the right represent them as % change from the present.

Environmental Science: Processes & Impacts Paper

Publ

ishe

d on

23

Janu

ary

2018

. Dow

nloa

ded

on 4

/17/

2019

6:0

0:29

PM

. View Article Online

The climate change increase is primarily due to an increase inHg(0) dry deposition and total wet deposition uxes. Thenegligibly small land use/land cover changes are a result of bothincreases and decreases in land cover in 2050 that inuence thesurfaces to which Hg deposits (Fig. S4†). The increase in

202 | Environ. Sci.: Processes Impacts, 2018, 20, 195–209

deposition from biomass burning is a result of higher biomassburning emissions in the boreal parts of North America andWestern U.S. These increases are, in turn, a result of greater reactivity caused by more vegetation availability and a warmerclimate in 2050.

This journal is © The Royal Society of Chemistry 2018

Fig. 4 Spatial distribution of walleye Hg concentrations in Michigan'sUP. Depicted are all 74 lakes for which data were located. Concen-trations in each lake were calculated for walleye of 42 cm length.

Paper Environmental Science: Processes & Impacts

Publ

ishe

d on

23

Janu

ary

2018

. Dow

nloa

ded

on 4

/17/

2019

6:0

0:29

PM

. View Article Online

3.2 Modeled responses of sh MeHg concentrations tochanges in deposition and watershed characteristics

The lake Hg cycling model was validated with measuredmercury concentrations in Torch Lake.70 Measured Hgconcentrations in Torch Lake water and sediments were usedin calibrating the model, and the more numerous measure-ments of sh Hg concentrations were used as validation. Hgconcentrations in brown bullhead, walleye, northern pike andsmallmouth bass have been measured approximately every sixyears since 1988.102 Model predictions for both mixed feedersand piscivorous sh overlapped with the range of measuredconcentrations in these sh groups, although predictionsunderestimated mean sh concentrations. The model doesnot explicitly account for sh size; rather, it predicts concen-trations for sh with a distribution of sizes about an invariantmean.

In the UP lake, modeled sh Hg content decreases inproportion to changes in atmospheric deposition. The aspira-tional scenario reduced sh Hg content by 65%, and the policy-in-action scenario led to an 11% decrease (Fig. 3). Theseresponses are lake-specic and result from the combination ofatmospheric deposition, in-lake processes, and catchmentcharacteristics. If policy-in-action rates of atmospheric deposi-tion from the Adirondacks area are applied, the model predictssh concentrations 34% higher than those with UP depositionrates for the same scenario. If, in addition to Adirondacksdeposition rates, watershed characteristics are changed tovalues representative of a landscape less sensitive than the UP(i.e., reduced wetland area by 38%, reduced runoff from uplandsoils from 10% of deposition to 5%), then sh Hg concentra-tions are predicted to be only 70% of the original prediction forUP conditions. In other words, the effects of varying landscapecharacteristics within the GL region are larger than the effects ofthe variable rates of atmospheric deposition found in thisregion.

Fig. 3 Predicted Hg concentrations in piscivorous fish in the UP lakeunder three policy scenarios and two virtual experiments. The aspi-rational scenario decreases fish Hg by 65% while the policy-in-actionscenario leads to an 11% decrease. The first virtual experimentincreases deposition over the policy-in-action scenario by 32% andfish Hg increases by 35%. The second virtual experiment maintains thehigh deposition rate but considers a less sensitive watershed (fewerwetlands, less runoff from forests); fish Hg is only 70% of the policy-in-action scenario with UP conditions.

This journal is © The Royal Society of Chemistry 2018

3.3 Spatial distribution of Hg in Michigan's UP lakes andwalleye

If the 74 lakes sampled for walleye Hg are representative of the�15 000 UP lakes,103 the survey results (Fig. 4) suggest that 77%of UP lakes are impaired with respect to EPA's sh Hg criterion(0.3 ppm wet wt). The percentage of impaired lakes in the UP isconsiderably above the national average of 49% (ref. 104) andabove the 23% reported as impaired in the Adirondacks.24

Possibly as a result of historical mining activities,105 concen-trations of Hg in walleye are higher in western than in easternUP lakes (t-test, p¼ 0.037), although fewer lakes in the east weresampled than in the west. Multiple linear regression and prin-cipal component analyses indicated that sh Hg concentrationsare best predicted by watershed area and trophic state for large(>3 km2) lakes, and by pH, wetland area in catchment, lake area,maximum depth and total phosphorus for small lakes.106

4. Discussion4.1 Regional nature of deposition responses to policychanges

Mapping of the projected changes in deposition due to envi-ronmental changes (climate, land use/land cover, biomassburning) and policy clearly shows that even within the GLregion there is spatial variability (Fig. 2 and Table 3). Areas inclose proximity to upwind sources show a larger response tocontrols of those sources than do areas far from sources. In alllocations, the majority of the simulated decrease in depositionis due to decreased Hg(II) deposition rather than Hg(0) deposi-tion; for the GL region as a whole, 75% of the decrease is due toHg(II) and 25% due to Hg(0). However, in the Adirondacks, 80%of the decrease is due to Hg(II) while in the UP 70% is due toHg(II). This projected outcome results from the short atmo-spheric residence time of Hg(II) emissions (0.5–2 days15) anddoes not reect regional differences in availability of oxidantsfor Hg(0).16 The projected large decrease in Hg(II) deposition isat odds with the absence of a trend in wet deposition of Hg butan observed decrease in litterfall Hg uxes in the Adirondacks37

that are postulated to result from recent declines in Hg(0)

Environ. Sci.: Processes Impacts, 2018, 20, 195–209 | 203

Environmental Science: Processes & Impacts Paper

Publ

ishe

d on

23

Janu

ary

2018

. Dow

nloa

ded

on 4

/17/

2019

6:0

0:29

PM

. View Article Online

deposition. According to our scenario modeling, those areascurrently receiving the highest rates of atmospheric depositionshow the largest percentage decreases in deposition due toenactment of existing policies in the U.S. In contrast, the UP isnot downwind of major Hg emission sources. U.S. policy-in-action changes in emissions still represent the majority ofchanges in deposition in the UP (70%), but the rest of the worldcontributes 1.8 times more to UP deposition than to the GLregion, and 250 times more than to the Adirondacks.

One implication of these regional differences in trajectoriesof Hg deposition is that it will be easiest to detect systemresponses to policy in those regions projected to show thelargest responses. This study suggests that it will be difficult todetect policy-related changes in lake monitoring data in Mich-igan's UP over the next 50 years, unless additional policy stepsare taken beyond those included in the scenarios of this project.Similarly, because the predicted changes in deposition due toclimate- and land-use changes are small relative to those asso-ciated with policy changes, the likelihood is small that thosechanges can be detected in the GL region unless anthropogenicemissions stabilize. Within the GL region, changes will beeasiest to detect in northern Illinois, Indiana, Ohio and Penn-sylvania (Fig. 2).

4.2 Local nature of lake sensitivity

Surveys23,24,107,108 have shown that lakes in certain geographicareas are more susceptible to having high MeHg concentrationsin sh. This sensitivity is manifested in the geographic distri-bution of state-wide sh consumption advisories for Hg in theU.S.;109 each U.S. state bordering the GL or Canada has a state-wide advisory although many of these states have rates ofatmospheric deposition lower than states to the south.20 Factorsresponsible for this sensitivity have been shown to includeprevalence of forests and wetlands in the catchment, low alka-linity, low pH, high sulfate loading, low nutrients and longwater residence times.23,107,110 Importantly, the susceptibility oflakes to having high sh MeHg content is determined more byfactors leading to production of MeHg than by the total supplyof Hg to lakes,111 although factors affecting accumulation ofMeHg in biota (e.g., sh growth rate, length of food web, OCcontent of sediments) are also important.110,112

This study suggests that Michigan's UP is a sensitive land-scape that readily converts atmospherically deposited Hg intoMeHg that is bioaccumulated in sh. Among 74 lakes for whichwalleye Hg concentrations are available, 77% had concentra-tions above the U.S. EPA criterion value of 0.3 ppm, and themedian concentration was 0.43 ppm.109 Comparisons amongregions are difficult because of different sh species present,different sampling strategies, and different analytical protocols.Yu et al.24 measured Hg concentrations in multiple sh species(not including walleye but including large- and small-mouthbass, sh of comparable trophic level) from 44 Adirondacklakes, a region repeatedly called a biological hotspot forHg.23,24,113 They found that in only 10 of 44 lakes were sh foundwith concentrations above 0.3 ppm, and in less than 10% oflakes were sh found with concentrations above 0.42 ppm, the

204 | Environ. Sci.: Processes Impacts, 2018, 20, 195–209

median value for UP walleye. This comparison supports theassertion that Michigan's UP is a highly sensitive landscapewith respect to promoting Hg bioaccumulation in sh.

The virtual experiments, performed to quantify the magni-tude of the landscape effect, suggest that landscape sensitivitydifferences within the GL region may cause larger differences insh Hg than do the variations in rates of atmospheric Hgdeposition. The virtual experiments kept conditions within thelake (Hg process rates, pH, alkalinity, sulfate, etc.) the same, butadded sequentially an increase in atmospheric deposition(32%) followed by a decrease in landscape sensitivity (reducedwetland area, reduced Hg runoff from forests). As expected, anincrease in atmospheric deposition increased sh Hg concen-trations (35%); the similarity in increases in deposition ratesand sh Hg concentrations reects the lake and watershedconditions used. Importantly, when the landscape sensitivity ischanged, the sh Hg drops more than enough to counter theincreased atmospheric deposition. The magnitude of the dropagain reects the particular conditions that were modeled(watershed : lake area ratio, wetland areas, runoff coefficientsused), but the conditions were selected to be representative ofareas within the GL region.

The factor rendering UP lakes vulnerable to high bio-accumulation of MeHg is primarily the abundance of wetlands.In the UP, 29% of the land area is occupied by wetlands; among85 lakes, the median percentage of the catchment occupied bywetlands was 18%. Wetlands are important sites of Hg meth-ylation; the MeHg may then be transported to lakes.114,115

Wetlands also export DOC to lakes; the DOC can bind with andtransport inorganic Hg to lakes.114 UP lakes are characterized byrelatively high DOC; among 61 UP lakes, the median DOC was7.7 mg L�1. High DOC in lakes results in reduced light pene-tration that can reduce algal production, and microbial degra-dation of DOC can consume oxygen in lake hypolimnia therebyrendering them suitable for Hg methylation. It is instructive tocompare characteristics promoting Hg bioaccumulationbetween the UP and the Adirondacks (Table 4); data for 44Adirondack lakes were taken from Yu et al.24 The sampled lakesin both areas are similar in having largely forested catchments,and being of comparable size. The Adirondack lakes have lowpH as expected in low alkalinity waters at high elevations thatreceive signicant acid deposition. The UP lakes have highDOC, consistent with the high percentage of wetlands in thecatchments, and higher chlorophyll. Yu et al. concluded thatlow pH and low alkalinity were the most prominent factorscontributing to Hg bioaccumulation in the Adirondack lakes.Based on comparison with the Adirondack lakes, these factorsdo not appear to be the controlling factors in the sampled UPlakes; rather, the UP lakes are distinguished by having higherDOC and higher wetland areas.

The UP lakes will not “recover” to the point that Indigenouspeoples can safely consume sh in the quantities desired by2050. Even low rates of atmospheric Hg deposition will lead tohigh bioaccumulation in sh. The modeled lake does notapproach the target concentration of <0.015 ppm under anyscenario. The aspirational scenario reduced Hg depositionapproximately three-fold, nearly the magnitude of the increase

This journal is © The Royal Society of Chemistry 2018

Table 4 Comparison of features affecting Hg bioaccumulation in UPand Adirondack lakesa

Parameter UP lakes Adirondack lakesb

Non-wetland forest area (%)101 77% 74%Wetland in catchment (%) 18% (12%, 85) 11%DOC (mg L�1) 7.7 (4.9, 61) 3.9 (1.4, 44)PH 7.5 (0.7, 99) 6.6 (0.6, 44)Alkalinity (meq L�1) 854 (820, 146) 86 (82, 44)Chlorophyll a (mg m�3) 3.7 (4.8, 105) 2.0 (3.0, 44)THg (ng L�1) 0.8 (2.5, 61) 1.5 (0.9, 44)Lake area (ha) 116 (795, 117) 132 (481, 44)

a Reported values are the median followed by the standard deviationand number of lakes in parentheses. Forest area is region-wide. b Datafrom Yu et al.24

Paper Environmental Science: Processes & Impacts

Publ

ishe

d on

23

Janu

ary

2018

. Dow

nloa

ded

on 4

/17/

2019

6:0

0:29

PM

. View Article Online

commonly observed in lake sediments since pre-industrialtimes.116,117 However, our modeling approach used for theaspirational scenario would not have brought Hg depositiondown to pre-industrial levels, because of the continuing inu-ence of legacy emissions. The insensitivity of the UP to thepolicy-in-action scenario (15% decrease in deposition) indicatesthat this region will require more aggressive policies to achievesignicant reductions in deposition. The combination of thesensitive landscape and the insensitivity of Michigan's UP tocurrent policies are likely to considerably prolong the timerequired to see substantial decreases in sh Hg.

4.3 Implications for Indigenous communities

These research results are particularly important for Indigenouscommunities that rely on harvesting and consuming sh asa means of maintaining cultural identity and socio-culturalwell-being; the sh harvest has been sustaining the originalGL people for nearly two millennia.118 The sensitivity of the UPlandscape to increased Hg deposition, methylation, and MeHgbioaccumulation is further confounded by the sensitivity of thepopulation. The presence of other airborne bioaccumulativetoxics in sh further compounds the problem. Native Americantribes have some of the highest documented sh consumptionrates in the U.S., with GL Indigenous populations currentlyconsuming MeHg at rates that are well above human healthcriteria.118,119 For the KBIC in particular, more than seventy-ve-percent of tribal members presently report sh as a primarysource of subsistence.118 Subsistence harvesting is a politicalright. As the oldest and largest federally-recognized Indian tribein Michigan, the KBIC retains treaty-protected homelands andharvesting rights across ten-million acres of the Lake Superiorwatershed, including the majority of the western UP118,120,121 (theKBIC is one of thirty-six Anishinaabe tribes that retain harvesttreaty rights throughout the GL region).

Harvesting sh is not only considered a treaty right, but also aninherent right. That is, harvesting from the region's waters isintegral to maintaining Anishinaabe identity rooted in a tradi-tional way-of-life. Contamination of the sh threatens this way-of-life and also compromises food sovereignty – an issue that iscommon for Indigenous communities throughout the world, but

This journal is © The Royal Society of Chemistry 2018

particularly salient for shing communities.122 Fish consumptionadvisories for Hg are widely disseminated in the GL region, butadvisories are minimally effective for populations that are mostdependent on sh.4–8 The socio-cultural reliance and dependenceon sh oen leads to a lack of adherence to advisory recom-mendations where Indigenous shing communities are put ina very difficult position. To forgo harvesting and sh consumptionpractices suppresses important cultural norms and preventsimportant knowledge transmission to future generations, yetconsuming sh places individuals at increased risk of physicalharm. As a result, many Anishinaabe people feel that they cannot,or will not, adhere to advisory recommendations – it is in essencea ‘false choice’.

Consumption advisories have been used to address theproblem of sh contamination for almost ve decades in theU.S., however, it is important to know that these advisories wereonly meant to be short-term means to manage sh contamina-tion.119 Initially, within the GL there was an assumption that Hgsources were regional and that regional policy could reduce andprevent Hg contamination to a point that advisories would nolonger be necessary. We now know that atmospheric sources ofHg to the UP contain a signicant fraction (25%) of global source,and, based on the modeling results presented here, policyoriented toward the Lake Superior Basin will only be minimallyeffective in achieving safe sh for vulnerable populations in theUP. Although it is important to consider implications forvulnerable populations in landscape sensitivity assessments, wemust also emphasize the socio-cultural and political factors thatsignicantly increase the magnitude of Hg contaminationconsequences and Hg reduction policy required at differingscales to address them. The ‘xity to place’ (i.e., inability torelocate) that many Indigenous peoples have to their homelands,both because of long-term connections and reserved treaty rights,requires a Hg policy approach that considers the spatial distri-bution of vulnerable populations worldwide.123,124

Multiple limitations remain in our ability to model lakeresponses to changes in Hg emissions. The scenarios used heredo not include potential longer-term changes in legacy emis-sions of Hg; hence our predictions overestimate the real-worldrate of response to changes in anthropogenic emissions, andomit the impact of legacy emissions occurring between 2000and 2050.125,126 Legacy emissions of Hg to surface waters alsowere not included in our modeling effort; work elsewhereindicates that such emissions prolong the time to recovery bydecades.45–47 Application of more complicated watershedmodels91 will be required to predict the timeframe of watershedresponses to changes in deposition. The lake Hgmodel can onlycrudely predict responses to changes in lake pH; such changesare likely in response to changes in atmospheric deposition ofacids and sulfate and to changes in DOC inputs to lakes.127,128

The model does not yet include sulfate deposition as a driver ofHg methylation in wetlands or in lakes.35,42,129 All of thesecurrent shortcomings, however, do not negate our predictionthat recovery will not occur in response to the policy scenariosconsidered here; they do prevent us from predicting the time-frame for recovery.

Environ. Sci.: Processes Impacts, 2018, 20, 195–209 | 205

Environmental Science: Processes & Impacts Paper

Publ

ishe

d on

23

Janu

ary

2018

. Dow

nloa

ded

on 4

/17/

2019

6:0

0:29

PM

. View Article Online

5. Conclusions

The approach for assessing the effects of policy on safety ofsh consumption developed in this project provided a meansto use state-of-the-art science tools (rigorous modeling usingcarefully craed scenario analysis) to answer a community-relevant question. While uncertainties in atmospheric, catch-ment, and lake models preclude an explicit prediction ofrecovery times for lakes, the conclusion that recovery will notfollow quickly aer implementation of the policies modeled inthis study is robust. Our application of this approach to the GLregion points out that recovery from Hg contamination isspatially variable. Regions such as Michigan's UP that are bothhighly sensitive to inputs from atmospheric deposition andnot close to the emission sources of that deposition willrequire careful policy development to recover from currentlevels of contamination.

Author contribution

J. Perlinger contributed to the writing of the manuscript and itsreview and revisions. N. Urban conducted lake modeling,contributed to the writing of the manuscript, and its review andrevision. A. Giang conducted GEOS-Chem modeling of policyscenarios and contributed to the Experimental section. N. Selincontributed to GEOS-Chem modeling of policy scenarios, andreviewed and revised the manuscript. A. Hendricks conductedlake modeling. H. Zhang conducted GEOS-Chem modeling ofclimate and land use/land cover scenarios and contributed tothe Experimental section. A. Kumar conducted GEOS-Chemmodeling of the biomass burning scenario and contributed tothe Experimental section. S. Wu contributed to GEOS-Chemmodeling related to the climate, land use, land cover, andbiomass burning scenarios and reviewed the Experimentalsection. V. Gagnon conducted university-community researchand engagement, and contributed to writing the manuscript,and its review and revision. H. Gorman conducted socialscience research and contributed to writing the manuscript.E. Norman conducted social science research and contributedto writing the manuscript.

Conflicts of interest

There are no conicts to declare.

Acknowledgements

The authors acknowledge the valuable contributions of theanonymous reviewers to this paper. The authors also acknowl-edge the Keweenaw Bay Indian Community, with particulargratitude to the Tribal Council, the Natural Resources Depart-ment, and the Ojibwa Community College for sharing knowl-edge and expertise with our research team throughout theproject. This research was funded by the U.S. National ScienceFoundation through Grant #ICER-1313755.

206 | Environ. Sci.: Processes Impacts, 2018, 20, 195–209

References

1 U. S. EPA, Great Lakes Fish Monitoring and SurveillanceProgram: Background, Trends, and General Conclusions,accessed September 9, 2017, https://www.epa.gov/great-lakes-monitoring/great-lakes-sh-monitoring-and-surveillance.

2 U. S. EPA, National Listing of Fish Advisories, Fish andShellsh Advisories and Safe Eating Guidelines, accessedSeptember 9, 2017, https://www.epa.gov/choose-sh-and-shellsh-wisely/sh-and-shellsh-advisories-and-safe-eating-guidelines.

3 GLIFWC, Guidance for Safe Consumption of Walleye fromInland Lakes within the Ceded Territories of Wisconsin,Michigan, and Minnesota, http://www.glifwc.org/Mercury/mercury.html, accessed September 9, 2017.

4 J. Cassady, Med. Anthropol. Q., 2010, 24, 451–471.5 J. Donatuto and B. Harper, Risk Anal., 2008, 28, 1497–1506.6 J. Donatuto, T. Sattereld and R. Gregory, Health Risk Soc.,2011, 13, 103–127.

7 C. A. O'Neill, J. Environ. Law Litigation, 2007, 22, 131–151.8 D. J. Ranco and D. Suagee, Antipode, 2007, 39, 691–707.9 V. S. Gagnon, Water History, 2016, 8, 365–384.10 E. Hoover, Ecol. Process, 2013, 2, 4.11 E. S. Norman, Area, 2013, 45, 179–187.12 D. J. Ranco, C. A. O'Neill, J. Donatuto and B. Harper,

Environ. Justice, 2011, 4, 221–230.13 UNEP, Minamata Convention on Mercury, http://

www.mercuryconvention.org/Portals/11/documents/Booklets/Minamata%20Convention%20on%20Mercury_booklet_English.pdf, accessed November 13, 2017.

14 V. S. Gagnon, H. S. Gorman and E. S. Norman, Gateways:International Journal of Community Research andEngagement, 2017, 10, 164–184.

15 C. T. Driscoll, Y. J. Han, C. Y. Chen, D. C. Evers,K. F. Lambert, T. M. Holsen, N. C. Kamman andR. K. Munson, Bioscience, 2007, 57, 17–28.

16 H. M. Horowitz, D. J. Jacob, Y. X. Zhang, T. S. Dibble,F. Slemr, H. M. Amos, J. A. Schmidt, E. S. Corbitt,E. A. Marais and E. M. Sunderland, Atmos. Chem. Phys.,2017, 17, 6353–6371.

17 W. Seiler, C. Eberling and F. Slemr, Pure Appl. Geophys.,1980, 118, 964–974.

18 E. A. Nater and D. F. Grigal, Nature, 1992, 358, 139–141.19 C. Seigneur, K. Vijayaraghavan, K. Lohman,

P. Karamchandani and C. Scott, Environ. Sci. Technol.,2004, 38, 555–569.

20 N. E. Selin and D. J. Jacob, Atmos. Environ., 2008, 42, 5193–5204.

21 M. D. Cohen, R. R. Draxler, R. S. Artz, P. Blanchard,M. S. Gustin, Y. J. Han, T. M. Holsen, D. Jaffe, P. Kelley,H. Lei, C. P. Loughner, W. T. Luke, S. N. Lyman, D. Niemi,J. M. Pacyna, M. Pilote, L. Poissant, D. Ratte, X. Ren,F. Steenhuisen, A. Steffen, R. Tordon and S. J. Wilson,Elementa, 2016, DOI: 10.12952/journal.elementa.000118.

This journal is © The Royal Society of Chemistry 2018

Paper Environmental Science: Processes & Impacts

Publ

ishe

d on

23

Janu

ary

2018

. Dow

nloa

ded

on 4

/17/

2019

6:0

0:29

PM

. View Article Online

22 H. Zhou, C. L. Zhou, M. M. Lynam, J. T. Dvonch, J. A. Barres,P. K. Hopke, M. Cohen and T. M. Holsen, Environ. Sci.Technol. Lett., 2017, 4, 91–97.

23 D. C. Evers, Y. J. Han, C. T. Driscoll, N. C. Kamman,M. W. Goodale, K. F. Lambert, T. M. Holsen, C. Y. Chen,T. A. Clair and T. Butler, BioScience, 2007, 57, 29–43.

24 X. Yu, C. T. Driscoll, M. R. Montesdeoca, D. Evers,M. Duron, K. Williams, N. Schoch and N. Kamman,Ecotoxicology, 2011, 20, 1543–1554.

25 D. C. Depew, N. M. Burgess and L. M. Campbell, Environ.Pollut., 2013, 176, 234–243.

26 K. R. Rolus, J. G. Wiener, R. J. Haro, M. B. Sandheinrich,S. W. Bailey and B. R. Seitz, Sci. Total Environ., 2015, 514,192–201.

27 C. D. Evans, P. J. Chapman, J. M. Clark, D. T. Monteith andM. S. Cresser, Global Change Biol., 2006, 12, 2044–2053.

28 M. Erlandsson, N. Cory, S. Kohler and K. Bishop, J. Geophys.Res.: Biogeosci., 2010, 115, G03004.

29 N. R. Urban, E. S. Verry, S. J. Eisenreich, D. F. Grigal andS. Sebestyen, Element cycling in upland/peatlandwatersheds, Peatland Biogeochemistry and WatershedHydrology at the Marcell Experimental Forest, ed. R. K.Kolka, S. Sebestyen, E. S. Verry and K. Brooks, CRC Press,Boca Raton, FL, 2011, pp. 213–241.

30 L. Lindestrom, Ambio, 2001, 30, 538–544.31 M. J. Parsons, D. T. Long and S. S. Yohn, Appl. Geochem.,

2010, 25, 1676–1687.32 R. C. Harris, J. W. M. Rudd, M. Amyot, C. L. Babiarz,

K. G. Beaty, P. J. Blancheld, R. A. Bodaly, B. A. Branreun,C. C. Gilmour, J. A. Graydon, A. Heyes, H. Hintelmann,J. P. Hurley, C. A. Kelly, D. P. Krabbenho, S. E. Lindberg,R. P. Mason, M. J. Paterson, C. L. Podemski, A. Robinson,K. A. Sandilands, G. R. Southworth, V. L. S. Louis andM. T. Tate, Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 16586–16591.

33 C. J. Watras, K. A. Morrison, R. J. M. Hudson, T. M. Frostand T. K. Kratz, Environ. Sci. Technol., 2000, 34, 4051–4057.

34 T. R. Hrabik and C. J. Watras, Sci. Total Environ., 2002, 297,229–237.

35 P. E. Drevnick, D. E. Caneld, P. R. Gorski,A. L. C. Shinneman, D. R. Engstrom, D. C. G. Muir,G. R. Smith, P. J. Garrison, L. B. Cleckner, J. P. Hurley,R. B. Noble, R. R. Otter and J. T. Oris, Environ. Sci.Technol., 2007, 41, 7266–7272.

36 N. Gandhi, R. W. K. Tang, S. P. Bhavsar andG. B. Arhonditsis, Environ. Sci. Technol., 2014, 48, 5404–5414.

37 J. R. Gerson and C. T. Driscoll, Environ. Sci. Technol., 2016,50, 10943–10950.

38 C. D. Knightes, E. M. Sunderland, M. Craig Barber,J. M. Johnston and R. B. Ambrose, Environ. Toxicol.Chem., 2009, 28, 881–893.

39 C. D. Knightes and R. B. Ambrose, Environ. Toxicol. Chem.,2007, 26, 807–815.

40 N. E. Selin, E. M. Sunderland, C. D. Knightes andR. A. Mason, Environ. Health Perspect., 2010, 118, 137–143.

This journal is © The Royal Society of Chemistry 2018

41 J. Munthe, R. A. Bodaly, B. A. Branreun, C. T. Driscoll,C. C. Gilmour, R. Harris, M. Horvat, M. Lucotte andO. Malm, Ambio, 2007, 36, 33–44.

42 C. J. Watras and K. A. Morrison, Can. J. Fish. Aquat. Sci.,2008, 65, 100–116.

43 MDEQ, Michigan's Water Chemistry Monitoring Program: AReport of Statewide Spatial Patterns 2005-2009 and FixedStation Status and Trends 1998-2008, W. R. Division ReportMI/DEQ/WRD-13/005, Michigan Department ofEnvironmental Quality, Lansing, MI, 2013.

44 H. D. Yang, S. Turner and N. L. Rose, Environ. Pollut., 2016,219, 1092–1101.

45 M. Meili, K. Bishop, L. Bringmark, K. Johansson, J. Munthe,H. Sverdrup and W. de Vries, Sci. Total Environ., 2003, 304,83–106.

46 L. Hakanson, Ecol. Model., 1996, 93, 251–262.47 K. Vijayaraghavan, L. Levin, L. Parker, G. Yarwood and

D. Streets, Environ. Toxicol. Chem., 2014, 33, 1238–1247.48 National listing of Fish Advisories, Fish and Shellsh

Advisories and safe eating guidelines, viewed November2016, https://www.epa.gov/choose-sh-and-shellsh-wisely/sh-and-shellsh-advisories-and-safe-eating-guidelines,accessed November 13 2017.

49 E. Oken, R. O. Wright, K. P. Kleinman, D. Bellinger,C. J. Amarasiriwardena, H. Hu, J. W. Rich-Edwards andM. W. Gillman, Environ. Health Perspect., 2005, 113, 1376–1380.

50 E. Groth, An Overview of Epidemiological Evidence on theEffects of Methylmercury on Brain Development, and ARationale for a Lower Denition of Tolerable Exposure ZeroMercury Working Group, Brussels, Belgium, 2012.

51 A. Giang and N. E. Selin, Proc. Natl. Acad. Sci. U. S. A., 2016,113, 286–291.

52 S. Wu, L. J. Mickley, D. J. Jacob, J. A. Logan, R. M. Yantoscaand D. Rind, J. Geophys. Res., 2007, 112, D05302.

53 X. M. Zhang, K. R. Rygwelski, M. D. Rowe, R. Rossmann andR. G. Kreis, J. Great Lakes Res., 2016, 42, 62–69.

54 S. Wu, L. J. Mickley, J. O. Kaplan and D. J. Jacob, Atmos.Chem. Phys., 2012, 12, 1597–1609.

55 H. Zhang, C. D. Holmes and S. Wu, Atmos. Environ., 2016,141, 230–244.

56 Y. X. Huang, S. L. Wu and J. O. Kaplan, Atmos. Environ.,2015, 121, 86–92.

57 A. Kumar, S. Wu, Y. Huang, H. Liao and J. Kaplan, Atmos.Environ., 2017, 173, 6–15.

58 E. S. Corbitt, D. J. Jacob, C. D. Holmes, D. G. Streets andE. M. Sunderland, Environ. Sci. Technol., 2011, 45, 10477–10484.

59 Y. Zhang, L. Jaegle, A. van Donkelaar, R. V. Martin,C. D. Holmes, H. M. Amos, Q. Wang, R. Talbot, R. Artz,S. Brooks, W. Luke, T. M. Holsen, D. Felton, E. K. Miller,K. D. Perry, D. Schmeltz, A. Steffen, R. Tordon, P. Weiss-Penzias and R. Zsolway, Atmos. Chem. Phys., 2012, 12,6095–6111.

60 D. G. Streets, Q. Zhang and Y. Wu, Environ. Sci. Technol.,2009, 43, 2983–2988.

Environ. Sci.: Processes Impacts, 2018, 20, 195–209 | 207

Environmental Science: Processes & Impacts Paper

Publ

ishe

d on

23

Janu

ary

2018

. Dow

nloa

ded

on 4

/17/

2019

6:0

0:29

PM

. View Article Online

61 P. Rafaj, I. Bertok, J. Cofala and W. Schopp, Atmos. Environ.,2013, 79, 472–479.

62 N. E. Selin, Environ. Toxicol. Chem., 2014, 33, 1202–1210.63 J. M. Pacyna, O. Travnikov, F. De Simone, I. M. Hedgecock,

K. Sundseth, E. G. Pacyna, F. Steenhuisen, N. Pirrone,J. Munthe and K. Kindbom, Atmos. Chem. Phys., 2016, 16,12495–12511.

64 Lake Superior Binational Program, Lake Superior ZeroDischarge Demonstration Program and Critical ChemicalReduction Milestones, 2017, https://www.epa.gov/sites/production/les/2015-11/documents/lake-superior-zero-discharge-demonstration-program-2012-8pp.pdf, accessedNovember 13, 2017.

65 H. M. Amos, D. J. Jacob, C. D. Holmes, J. A. Fisher, Q. Wang,R. M. Yantosca, E. S. Corbitt, E. Galarneau, A. P. Rutter,M. S. Gustin, A. Steffen, J. J. Schauer, J. A. Graydon,V. L. St Louis, R. W. Talbot, E. S. Edgerton, Y. Zhang andE. M. Sunderland, Atmos. Chem. Phys., 2012, 12, 591–603.

66 C. D. Holmes, D. J. Jacob, E. S. Corbitt, J. Mao, X. Yang,R. Talbot and F. Slemr, Atmos. Chem. Phys., 2010, 10,12037–12057.

67 N. E. Selin, D. J. Jacob, R. M. Yantosca, S. Strode, L. Jaegleand E. M. Sunderland, Global Biogeochem. Cycles, 2008,22, GB 2011.

68 A. L. Soerensen, E. M. Sunderland, C. D. Holmes,D. J. Jacob, R. M. Yantosca, H. Skov, J. H. Christensen,S. A. Strode and R. P. Mason, Environ. Sci. Technol., 2010,44, 8574–8580.

69 D. Rind, J. Lerner, J. Jonas and C. McLinden, J. Geophys.Res.: Atmos., 2007, 112, D09315.

70 A. N. Hendricks and N. R. Urban, Ecol. Model., in prep.71 K. Soetaert, T. Petzoldt and R. W. Setzer, J. Stat. Soware,

2010, 33, 1–25.72 C. D. Knightes and R. B. Ambrose, Development of an

Ecological Risk Assessment Methodology for AssessingWildlife Exposure Risk Associated with Mercury-Contaminated Sediments in Lake and River Systems, ReportEPA/600/R-06/073, U.S. Environmental Protection AgencyOffice of Research and Development, Washington, D.C.,2006.

73 C. D. Knightes, Environ. Model. Soware, 2008, 23, 495–510.74 L. Hakanson, Water, Air, Soil Pollut., 2000, 124, 301–317.75 M. C. Barber, Environ. Toxicol. Chem., 2008, 27, 755–777.76 E. B. Swain, D. R. Engstrom, M. E. Brigham, T. A. Henning

and P. L. Brezonik, Science, 1992, 257, 784–787.77 P. Lorey and C. T. Driscoll, Environ. Sci. Technol., 1999, 33,

718–722.78 P. E. Drevnick, D. R. Engstrom, C. T. Driscoll, E. B. Swain,

S. J. Balogh, N. C. Kamman, D. T. Long, D. G. C. Muir,M. J. Parsons, K. R. Rolus and R. Rossmann, Environ.Pollut., 2012, 161, 252–260.

79 J. Kwan and P. A. Taylor, Bound.-Layer Meteorol., 1994, 68,339–356.

80 D. E. Anderson, R. G. Striegl, D. I. Stannard,C. M. Michmerhuizen, T. A. McConnaughey andJ. W. LaBaugh, Limnol. Oceanogr., 1999, 44, 988–1001.

208 | Environ. Sci.: Processes Impacts, 2018, 20, 195–209

81 T. R. Khan and J. A. Perlinger, Geosci. Model Dev., 2017, 10,3861–3888.

82 J. S. Read, D. P. Hamilton, A. R. Desai, K. C. Rose,S. MacIntyre, J. D. Lenters, R. L. Smyth, P. C. Hanson,J. J. Cole, P. A. Staehr, J. A. Rusak, D. C. Pierson,J. D. Brookes, A. Laas and C. H. Wu, Geophys. Res. Lett.,2012, 39, DOI: 10.1029/2012GL051886.

83 B. D. Hall, H. Manolopoulos, J. P. Hurley, J. J. Schauer,V. L. St Louis, D. Kenski, J. Graydon, C. L. Babiarz,L. B. Cleckner and G. J. Keeler, Atmos. Environ., 2005, 39,7557–7569.

84 M. R. Risch, J. F. Dewild, D. A. Gay, L. Zhang, E. W. Boyerand D. P. Krabbenho, Environ. Pollut., 2017, 228, 8–18.

85 M. R. Risch, D. A. Gay, K. K. Fowler, G. J. Keeler,S. M. Backus, P. Blanchard, J. A. Barres and J. T. Dvonch,Environ. Pollut., 2012, 161, 261–271.

86 A. Hendricks, A Model to Predict Concentrations andUncertainty for Mercury Species in Lakes, M.S. thesis,Michigan Technological University, 2018.

87 D. F. Grigal, Environ. Rev., 2002, 10, 1–39.88 H. Hsu-Kim, C. S. Eckley, D. Acha, X. Feng, C. Gilmour,

S. Jonsson and C. P. J. Mitchell, Ambio, in review.89 M. N. Futter, A. E. Poste, D. Buttereld, P. J. Dillon,

P. G. Whitehead, A. P. Dastoor and D. R. S. Lean, Sci.Total Environ., 2012, 424, 219–231.

90 H. E. Golden, C. D. Knightes, P. A. Conrads, G. M. Davis,T. D. Feaster, C. A. Journey, S. T. Benedict, M. E. Brighamand P. M. Bradley, J. Geophys. Res.: Biogeosci., 2012, 117,G01006.

91 H. E. Golden, C. D. Knightes, P. A. Conrads, T. D. Feaster,G. M. Davis, S. T. Benedict and P. M. Bradley, Environ.Toxicol. Chem., 2013, 32, 2165–2174.

92 R. B. Mills, A. M. Paterson, J. M. Blais, D. R. S. Lean,J. P. Smol and G. Mierle, Can. J. Fish. Aquat. Sci., 2009, 66,187–200.

93 J. T. Bushey, C. T. Driscoll, M. J. Mitchell, P. Selvendiranand M. R. Montesdeoca, Hydrol. Processes, 2008, 22, 4813–4826.

94 R. K. Kolka, D. F. Grigal, E. A. Nater and E. S. Verry, Soil Sci.Soc. Am. J., 2001, 65, 897–905.

95 P. C. Frost, J. H. Larson, G. A. Lamberti, C. A. Johnston,K. C. Young, P. A. Maurice and S. D. Bridgham, Aquat.Sci., 2006, 68, 40–51.

96 C. Babiarz, S. Hoffmann, A. Wieben, J. Hurley, A. Andren,M. Shafer and D. Armstrong, Environ. Pollut., 2012, 161,299–310.

97 P. A. Soranno, K. S. Cheruvelil, T. Wagner, K. E. Websterand M. T. Bremigan, PLoS One, 2015, 10(8), e0135454.

98 K. S. Cheruvelil, P. A. Soranno, K. E. Webster andM. T. Bremigan, Ecol. Appl., 2013, 23, 1603–1618.

99 J. E. Breck, Compilation of Databases on Michigan LakesMDNR Fisheries Division Lansing, MI, 2004.

100 D. Knauer, G. Pelkola, S. Casey, D. Krabbenho andJ. F. Dewild, The distribution of total and methyl mercury inMichigan's Upper Peninsula Lakes, Michigan Dept.Environmental Quality, Lansing, MI, 2011.

This journal is © The Royal Society of Chemistry 2018

Paper Environmental Science: Processes & Impacts

Publ

ishe

d on

23

Janu

ary

2018

. Dow

nloa

ded

on 4

/17/

2019

6:0

0:29

PM

. View Article Online

101 National Land Cover Database 2011, https://www.mrlc.gov/nlcd2011.php, accessed November 13, 2017.

102 MDEQ, Status of Fish Contaminant Levels in the Torch LakeArea of Concern 2013, W. R. D. Michigan Department ofEnvironmental Quality Report MI/DEQ/WRD-16/003,Michigan Dept. Environmental Quality, Lansing,Michigan, 2016.

103 C. P. McDonald, J. A. Rover, E. G. Stets and R. G. Striegl,Limnol. Oceanogr., 2012, 57, 597–606.

104 L. L. Stahl, B. D. Snyder, A. R. Olsen and J. L. Pitt, Environ.Monit. Assess., 2009, 150, 3–19.

105 W. C. Kerfoot, N. R. Urban, C. P. McDonald, H. Zhang,R. Rossmann, J. A. Perlinger, T. Khan, A. Hendricks,M. Priyadarshini and M. Bolstad, Environ. Sci.: ProcessesImpacts, 2018, in press.

106 M. Priyadarshini, Factors Contributing to ElevatedConcentrations of Mercury and PCBs in Fish in theInland Lakes of Michigan's Upper Peninsula and LakeSuperior, M.S. thesis, Michigan Technological University,M.S., 2017.

107 C. A. Eagles-Smith, J. T. Ackerman, J. J. Willacker,M. T. Tate, M. A. Lutz, J. A. Fleck, A. R. Stewart,J. G. Wiener, D. C. Evers, J. M. Lepak, J. A. Davis andC. F. Pritz, Sci. Total Environ., 2016, 568, 1171–1184.

108 D. C. Evers, J. G. Wiener, C. T. Driscoll, D. A. Gay, N. Basu,B. A. Monson, K. F. Lambert, H. A. Morrison, J. T. Morgan,K. A. Williams and A. G. Soehl, Great Lakes MercuryConnections: The Extent and Effects of Mercury Pollution inthe Great Lakes Region, Biodiversity Research Institute,Gorham, Maine, 2011.

109 National Listing of Fish Advisories, https://www.epa.gov/sites/production/les/2015-06/documents/maps-and-graphics-2011.pdf, accessed November 13, 2017.

110 C. Y. Chen, C. L. Folt, R. S. Stemberger, N. C. Kamman andB. M. Mayes, Ecotoxicology, 2005, 14, 135–147.

111 C. A. Eagles-Smith, J. G. Wiener, C. S. Eckley, J. J. Willacker,D. C. Evers, M. Marvin-DiPasquale, D. Obrist, J. A. Fleck,G. R. Aiken, J. M. Lepak, A. K. Jackson, J. P. Webster,A. R. Stewart, J. A. Davis, C. N. Alpers and J. T. Ackerman,Sci. Total Environ., 2016, 568, 1213–1226.

112 C. Y. Chen, M. Dionne, B. M. Mayes, D. M. Ward, S. Sturupand B. P. Jackson, Environ. Sci. Technol., 2009, 43, 1804–1810.

113 D. C. Evers, R. P. Mason, N. C. Kamman, C. Y. Chen,A. L. Bogomolni, D. L. Taylor, C. R. Hammerschmidt,S. H. Jones, N. M. Burgess, K. Munney and K. C. Parsons,EcoHealth, 2008, 5, 426–441.

This journal is © The Royal Society of Chemistry 2018

114 P. Selvendiran, C. T. Driscoll, J. T. Bushey andM. R. Montesdeoca, Environ. Pollut., 2008, 154, 46–55.

115 P. Selvendiran, C. T. Driscoll, M. R. Montesdeoca andJ. T. Bushey, J. Geophys. Res.: Biogeosci., 2008, 113, G00C01.

116 D. R. Engstrom, E. B. Swain, T. A. Henning, M. E. Brighamand P. L. Brezonik, Adv. Chem. Ser., 1993, 237, 33–66.

117 D. R. Engstrom, W. F. Fitzgerald, C. A. Cooke,C. H. Lamborg, P. E. Drevnick, E. B. Swain, S. J. Baloghand P. H. Balcom, Environ. Sci. Technol., 2014, 48, 6533–6543.

118 V. S. Gagnon, P. Nankervis and E. Johnston, Keweenaw BayIndian Community Wildlife and Natural Resources Report,KBIC Natural Resources Department, Baraga, MI,2013, http://nrd.kbic-nsn.gov/sites/default/les/KBIC%20WILDLIFE%20and%20NR%20REPORT%2011_2013%20%28hyperlinked%20TABLE%20of%20CONTENTS%29_0.pdf, accessed November 13, 2017.

119 C. A. O'Neill, Mercury, risk & justice, Environmental LawInstitute, Environmental Law Reporter News and Analysis,2004.

120 A. McCammon-Soltis and J. S. Kekek, Fullling Ojibwetreaty promises—an overview and compendium ofrelevant cases, statutes and agreement, Great Lakes IndianFish & Wildlife Commission Division of IntergovernmentalAffairs, 2009, http://www.glifwc.org/minwaajimo/Papers/Legal%20Paper%20-%20DIA.pdf, accessed November 13,2017.

121 C. F. Wilkinson, Blood Struggle: The Rise of Modern IndianNations, NY, 2005.

122 E. S. Norman, Soc. Nat. Resour., 2017, 30, 537–553.123 E. S. Norman, Water Alternatives, 2012, 5, 138–160.124 E. S. Norman, Governing Transboundary Waters: Canada, the

United States, and Indigenous Communities, Routledge,Florence, KY, 2014.

125 H. M. Amos, D. J. Jacob, D. G. Streets and E. M. Sunderland,Global Biogeochem. Cycles, 2013, 27, 410–421.

126 M. Jiskra, J. G. Wiederhold, U. Skyllberg, R. M. Kronberg,I. Hajdas and R. Kretzschmar, Environ. Sci. Technol., 2015,49, 7188–7196.

127 M. Erlandsson, I. Buffam, J. Folster, H. Laudon,J. Temnerud, G. A. Weyhenmeyer and K. Bishop, GlobalChange Biol., 2008, 14, 1191–1198.

128 B. L. Skjelkvale, H. Borg, A. Hindar and A. Wilander, Appl.Geochem., 2007, 22, 1174–1180.

129 J. D. Jeremiason, D. R. Engstrom, J. E. Almendinger,E. B. Swain, B. A. Monson, E. A. Nater, B. M. Johnson andR. K. Kolka, Environ. Sci. Technol., 2006, 40, 3800–3806.

Environ. Sci.: Processes Impacts, 2018, 20, 195–209 | 209

Global and Local Impacts of Delayed Mercury Mitigation EffortsHelene Angot,*,† Nicholas Hoffman,‡ Amanda Giang,†,§ Colin P. Thackray,∥ Ashley N. Hendricks,⊥

Noel R. Urban,⊥ and Noelle E. Selin†,‡

†Institute for Data, Systems, and Society, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, United States‡Department of Earth, Atmospheric, and Planetary Sciences, Massachusetts Institute of Technology, Cambridge, Massachusetts02139, United States§Institute for Resources, Environment and Sustainability, University of British Columbia, Vancouver, British Columbia Canada V6T1Z4∥Harvard John A. Paulson School of Engineering and Applied Sciences, Harvard University, Cambridge, Massachusetts 02138,United States⊥Civil and Environmental Engineering Department, Michigan Technological University, Houghton, Michigan 49931, United States

*S Supporting Information

ABSTRACT: Mercury (Hg) is emitted to air by natural andanthropogenic sources, transports and deposits globally, andbioaccumulates to toxic levels in food webs. It is addressed underthe global 2017 Minamata Convention, for which periodiceffectiveness evaluation is required. Previous analyses have estimatedthe impact of different regulatory strategies for future mercurydeposition. However, analyses using atmospheric models tradition-ally hold legacy emissions (recycling of previously deposited Hg)constant, and do not account for their possible future growth. Here,using an integrated modeling approach, we investigate how delays inimplementing emissions reductions and the associated growinglegacy reservoir affect deposition fluxes to ecosystems in differentglobal regions. Assuming nearly constant yearly emissions relative to2010, each 5-year delay in peak emissions defers by additional extra ca. 4 years the return to year 2010 global deposition. On aglobal average, each 5-year delay leads to a 14% decrease in policy impacts on local-scale Hg deposition. We also investigate theresponse of fish contamination in remote lakes to delayed action. We quantify the consequences of delay for limiting the Hgburden of future generations and show that traditional analyses of policy impacts provide best-case estimates.

1. INTRODUCTION

Mercury (Hg) is an environmental toxicant dangerous tohuman health and the environment. Because of its long lifetimein the atmosphere (0.3−1 year),1,2 Hg travels regionally andglobally in its gaseous elemental form (Hg(0)). It deposits toecosystems by wet and dry processes as Hg(0) and gaseous/particulate divalent Hg (Hg(II)), and converts to highly toxicmethylmercury (MeHg) which bioaccumulates in aquaticsystems.1,3 Fish consumption is thus a main source of exposureto Hg for the general population.4−6

Regulatory actions to reduce human exposure to Hg aim toreduce anthropogenic inputs to the environment. In thatcontext, the Minamata Convention on Mercury7 entered intoforce in August 2017 and has 101 parties as of November2018. Under Article 8, parties “shall take measures to control,and where feasible, reduce emissions of Hg to the atmosphere”,and anthropogenic Hg emissions to the atmosphere areprojected to begin to decrease based on current and enhancedpolicy efforts.8,9 However, emitted Hg circulates for decades tocenturies, and anthropogenic emissions have a long-lasting

impact on the global Hg cycle.10 Legacy emissions (i.e.,recycling of previously deposited Hg) from soil and oceanicreservoirs account for about three-fifths of Hg annually emittedto the atmosphere.11 Even if anthropogenic emissions stayconstant, Hg deposition will continue to increase due to legacyemissions.11,12

Parameterizing legacy emissions in atmospheric models ischallenging due to a paucity of models which can capture boththree-dimensional atmospheric transport and oceanic/terres-trial cycling simultaneously.13,14 The majority of policyanalyses conducted using atmospheric models therefore onlyreflect changes in direct anthropogenic emissions, and do notconsider the effect of changing legacy emissions. For example,Pacyna et al.9 estimated that year 2035 global anthropogenicemissions would be reduced by 940 tons compared to 2010

Received: August 13, 2018Revised: October 17, 2018Accepted: October 30, 2018Published: October 30, 2018

Policy Analysis

pubs.acs.org/estCite This: Environ. Sci. Technol. 2018, 52, 12968−12977

© 2018 American Chemical Society 12968 DOI: 10.1021/acs.est.8b04542Environ. Sci. Technol. 2018, 52, 12968−12977

Dow

nloa

ded

via

UN

IV O

F B

RIT

ISH

CO

LU

MB

IA o

n A

pril

17, 2

019

at 1

7:00

:52

(UT

C).

Se

e ht

tps:

//pub

s.ac

s.or

g/sh

arin

ggui

delin

es f

or o

ptio

ns o

n ho

w to

legi

timat

ely

shar

e pu

blis

hed

artic

les.

and Hg deposition to ecosystems by 20−30% (except in India)if policy commitments and plans are fully implemented.However, in their analysis, while anthropogenic and naturalsources keep emitting Hg during the period 2010−2035 andthe Hg legacy reservoir grows, legacy emissions were simulatedat 2010 level.9 Another recent analysis15 of future policy,investigating projected Hg deposition in Asia by 2050, also didnot account for legacy Hg. However, using a biogeochemicalcycle model, the authors estimated that legacy Hg changescould alter the magnitude of calculated policy impacts by 30%,but they did not resolve this effect spatially.Here, using an integrated approach that combines

biogeochemical cycle modeling with global-scale chemicaltransport modeling, we investigate the impacts of delayedglobal action on global- and local-scale Hg deposition andevaluate associated changes in policy impacts, including theirspatial resolution. We test the hypothesis that a longer delay innear-term peak emissions will lead to a larger Hg legacy pooland thus a measurable influence on expected policy impacts.

We further examine consequences of delayed global mitigationefforts in different regions of the world to illustrateimplications relevant to environmental justice concerns. Forone of these regions, a Native American community in theU.S., we examine the impact on fish contamination in remotelakes where the main source of contamination is atmosphericdeposition from the global Hg pool.

2. MATERIALS AND METHODS

2.1. Conceptual Framework. The conceptual frameworkof this study is presented in Figure 1. Present-day (2010) andfuture simulations were performed using the chemicaltransport model (CTM) GEOS-Chem (see Section 2.3). Asin traditional future projections with other atmosphericmodels,9 legacy emissions were initially assumed constant to2010 level to calculate the effects of changes in directanthropogenic emissions (see Section 2.2). A global bio-geochemical cycle (GBC) model was used to calculate a globallegacy penalty as a function of the amount of emissions since

Figure 1. Conceptual framework. Due to a challenging parametrization, chemical transport models (CTMs, GEOS-Chem in this study)traditionally hold legacy emissions constant at present-day (2010) levels when making future projections. The latter thus only reflect changes indirect anthropogenic emissions. Using a fully coupled seven-reservoir global biogeochemical cycle model11,16 we account for future legacy emissionsas a result of both past and future emissions by adjusting GEOS-Chem outputs. The effective policy impacts in terms of local Hg deposition arethen dependent on policy delay. The adjusted deposition flux is then used as input to a lake Hg model to evaluate the response of fishcontamination to delayed action. Current Policy (CP), New Policy (NP) and Maximum Feasible Reduction (MFR) refer to future global emissionsscenarios developed by Pacyna et al.9

Table 1. Simulations Performed with the Chemical Transport Model (CTM) GEOS-Chem and the Seven-Reservoir GlobalBiogeochemical Cycle (GBC) Model11,16a

simulation CTM meteorological year simulated: 2010b GBC model years simulated: 2000 BCE-2100 CE

BASE 2010 AMAP/UNEP inventory17 and emissions controls18c Street et al.19 from 2000 BCE to 2008 CE. CP from 2009 onward

PRE-2010LEGACY

Primary anthropogenic emissions zeroed out Primary anthropogenic emissions completely eliminated as of 2010

FUTURE Future (CP, NP or MFR) emissions inventories9 NP or MFR implemented in YYYY = 2010, 2020, 2025, 2030, 2035, or 2050

PRE-YYYYLEGACY

Same as PRE-2010 LEGACY since 2010-YYYY emissions arenot taken into account

Primary anthropogenic emissions completely eliminated as of: YYYY = 2020,2025, 2030, 2035, or 2050

aCurrent Policy (CP), New Policy (NP), and Maximum Feasible Reduction (MFR) refer to future global emissions scenarios developed by Pacynaet al.9 A number of simulations listed here were performed to check the robustness of the method (see SI Section 1.1). bThe model was run formeteorological years 2007−2010, with the first three years used as the initialization period. We used consistent 2007−2010 meteorology for presentand future runs to isolate the effect of emissions. cIn order to evaluate present-day model outputs against observations and account for interannualvariability, this simulation was also performed for meteorological years 2009−2015 following a three-year spin-up.

Environmental Science & Technology Policy Analysis

DOI: 10.1021/acs.est.8b04542Environ. Sci. Technol. 2018, 52, 12968−12977

12969

2010 (see Section 2.4). This global legacy penalty was thenspatially distributed and added to future deposition fluxes fromthe CTM (see Section 2.4). Using this approach, adjusted Hgdeposition from the CTM and the effective policy impactsdiffer as a function of the total amount of mercury emitted,which is greater for longer policy delay. Adjusted Hgdeposition fluxes were then used as inputs to a lake Hgmodel (see Section 2.5) to investigate the influence of delayedglobal action on lacustrine fish contamination.Throughout this paper, we compare policy impacts assuming

an immediate (i.e., traditional method−no delay/legacypenalty) or delayed global action to reduce emissions. Foreach grid box of the CTM, policy impacts (PI, in Δμg/m2/yr)are calculated as the difference in total deposition betweenfuture and present-day simulations. The percent change (PC)in policy impacts due to a global action delayed to year YYYY(YYYY = 2020, 2025, 2030, 2035, or 2050) is calculatedaccording to eq 1:

PCPI PI

PI100YYYY

action delayed to YYYY traditional method

traditional method=

−×

(1)

The mean percent change in policy impacts, given by eq 2, isthe average percent change due to a near-term (2020−2035)5-year delay:

PC(PC PC ) (PC PC ) (PC PC )

32035 2030 2030 2025 2025 2020=

− + − + −

(2)

2.2. Present-Day and Future Emissions Scenarios.Simulations performed with the CTM and the GBC model arelisted in Table 1. The GEOS-Chem present-day (2010) BASEsimulation was performed using the AMAP/UNEP inven-tory,17 applying emission controls to U.S., Canadian, Euro-pean, and Chinese emissions from coal fired power plants.18

The CTM PRE-2010 LEGACY simulation was performed toquantify deposition from current (2010) legacy emissions andevaluate its spatial pattern. FUTURE simulations wereperformed using gridded emissions inventories developed byPacyna et al.9 Briefly, the Current Policy (CP) scenarioprojects that annual Hg emissions will slightly increase in 2035(ca. +75 Mg compared to 2010, i.e., +3.02 Mg yr−1).Increasing energy demand contributing to increased emissionsglobally will be offset by the implementation of additionalcontrol measures. The more stringent New Policy (NP)scenario indicates that annual emissions will significantlydecrease by 2035 (ca. −820 Mg compared to 2010, i.e., −32.7Mg yr−1). This scenario assumes that policy commitments andplans announced by countries worldwide to reduce greenhousegas emissions and phase out fossil fuel subsidies are fullyimplemented. Additionally, this scenario assumes that the useof Hg in products will be reduced by 70% by implementingArticle 4 of the Minamata Convention on Hg-added products.Finally, the Maximum Feasible Reduction (MFR) scenarioleads to a dramatic decrease of annual Hg anthropogenicemissions (ca. −1500 Mg compared to 2010, that is, −59.9 Mgyr−1). In this scenario, all countries reach the highest feasiblereduction efficiency in each emission sector.The GBC model was driven by 2000 BCE to 2008 CE

primary anthropogenic emissions from Streets et al.,19 Hgdischarges from rivers (held constant at present-day levels),16

and global geogenic emissions (90 Mg yr−1).20,21 Forconsistency with the emissions inventories used in the CTM,

we did not include additional 1850−2008 atmospheric Hgemissions from commercial products (105 Gg) proposed byHorowitz et al.22 For 2009-onward primary anthropogenicemissions, we used future emissions scenarios developed byPacyna et al.9 as described above.

2.3. Chemical Transport Modeling. The global CTMGEOS-Chem (www.geos-chem.org) was used to projectpresent-day and future total (wet+dry) gross Hg depositionfluxes to ecosystems. The model is driven by assimilatedmeteorological data from the NASA GMAO Goddard EarthObserving System.23 MERRA-2 data were used for thesimulations (https://gmao.gsfc.nasa.gov/products/). GEOS-Chem is a global-scale model that couples a 3D atmos-phere,2,24,25 a 2D surface-slab ocean,26 and a 2D terrestrialreservoir27 at a 2° × 2.5° horizontal resolution. A two-stepoxidation mechanism of Hg(0) initiated by Br was used. Thesecond-stage HgBr oxidation is mainly by the NO2 and HO2radicals using the new mechanism for atmospheric redoxchemistry developed by Horowitz et al.2 Oxidant fields fromSchmidt et al.28 have 4°× 5° horizontal resolution. Photo-reduction of aqueous-phase Hg(II)−organic complexes isdependent on the local concentration of organic aerosols, theNO2 photolysis frequency, and an adjusted coefficient(K_RED_JNO2) set to 9.828 × 10−2 m3 μg−1 here forsimulations with MERRA-2 meteorological fields and use ofthe slab ocean.29 For further details, a comprehensivedescription of the model is available elsewhere.30

2.4. Legacy Penalty. We used a previously published fullycoupled seven-reservoir GBC model11,16 (available at https://github.com/SunderlandLab/gbc-boxmodel) to scale legacyemissions. The model allows full coupling of the atmosphere,ocean (surface, subsurface, and deep ocean), and terrestrialecosystems (fast terrestrial, slow and armored soils). Hg cyclesbetween reservoirs and is ultimately removed by burial in deepmarine sediments. In order to evaluate the impact of delayedaction on global Hg deposition, incremental 5-year delays inimplementing a NP or MFR scenario were tested (see Table 1,FUTURE simulations). We assumed a CP scenario (i.e., a 3.02Mg yr−1 increase) until implementation of a NP or MFRscenario. A total of six simulations were performed with theGBC model (PRE-2010 and PRE-YYYY LEGACY in Table 1)in order to quantify the contribution to global Hg depositionof emissions during the policy delay Δt. The Δt-dependentglobal legacy penalty was defined as the difference in globaldeposition between PRE-YYYY and PRE-2010 LEGACYsimulations (where YYYY = 2020, 2025, 2030, 2035, or2050). The global legacy penalty was then spatially distributed(see below) and added to future deposition fluxes from theCTM in each grid-box (see Figure 1). Rather than assuming aglobally homogeneous distribution of legacy emissions (i.e.,global legacy penalty evenly divided among all 2° × 2.5° gridboxes of the CTM), we used the spatial distribution of thelegacy emissions contribution to Hg deposition (PRE-2010LEGACY simulation with the CTM, see Table 1). Spatialdifferences in legacy impact relate to atmospheric transport,geographic factors (e.g., land vs ocean) and reemissions. InGEOS-Chem, reemission of Hg previously deposited to landfollows the deposition patterns of current sources.31

Local consequences of delayed global action for Hgdeposition fluxes to ecosystems are discussed (see Section3.2) at four selected sites located at varying distance fromanthropogenic sources: (A) tribal areas of eastern Maine,representative of remote regions and used to illustrate

Environmental Science & Technology Policy Analysis

DOI: 10.1021/acs.est.8b04542Environ. Sci. Technol. 2018, 52, 12968−12977

12970

implications relevant to environmental justice concerns (seeSection 3.3); (B) Ahmedabad, the largest city of the Indianstate of Gujarat and the location of two coal-fired power plantsof more than 1000 MW electricity generation;32 (C) Shanghai,China’s biggest city and one of the main industrial centers,where elevated atmospheric Hg concentrations have beenreported;33−35 and (D) an area of the Southern Pacific knownfor albacore tuna fisheries.36 Sunderland et al.37 recentlyestimated that seafood harvested from the Equatorial andSouth Pacific Ocean accounts for 25% of the U.S. population-wide MeHg intake.2.5. Fish Contamination. To investigate the influence of

delayed global action on fish contamination, we used a recentlydeveloped implementation of the mechanistic model SERAFM(Spreadsheet-based Ecological Risk Assessment for the Fate ofMercury) of Hg in aquatic environments developed by the U.S.Environmental Protection Agency (EPA).38,39 This model hasbeen widely used and evaluated since its development.40−46

The implementation of SERAFM used here was developed byHendricks47 in the programming language R and modified fornonsteady state conditions to enable prediction of lakeresponses to changes in loadings. A description of this massbalance model is provided by Perlinger et al.48 Briefly, themodel is a three-reservoir box model (epilimnion, hypolimn-ion, and sediments). It enables prediction of aqueous Hgconcentrations based on the characteristics of the lake ofinterest (e.g., depth, retention time), its watershed (e.g.,surface area), and the local Hg atmospheric deposition flux.The model does not include fish population dynamics andsolves for an annually averaged MeHg concentration in thewater column that is multiplied by bioaccumulation factors(BAFs) for mixed feeders or piscivorous fish38 to give adistribution of MeHg concentrations in fish. We thereforeneglect the time required (3−7 years) for fish populations toreach steady state following a change in Hg loadings.48−50

Additionally, Hg runoff from catchments is set to beproportional to atmospheric deposition (see Supporting

Information (SI) Section 1.2.a). This approach assumes thatonly recently deposited Hg is susceptible to leaching leading toa fast response to changes in Hg deposition.48 Moreinformation regarding the parametrization used here and themodel performance can be found in SI Section 1.2 and FigureS1.Although some lake Hg contamination can be attributed to

direct inputs from local sources,51,52 we focus here on remotelakes where the main source of contamination is atmosphericdeposition from the global Hg pool (see Section 3.2). Morespecifically, we concentrate on remote tribal regions of easternMaine (see SI Figure S2) since Native Americans areparticularly vulnerable to Hg contamination due to traditionalsubsistence fishing.53 In order to investigate the response tochanges in atmospheric deposition, the model was first run for10 years (2000−2010) to reach steady-state, and thentransiently using the adjusted deposition values from theCTM described in Section 2.1. Here, we evaluated theresponse of fish contamination to delayed global actionassuming everything else (e.g., food web structure, nutrientloading) constant. While Hg biogeochemical cycling will beaffected by climate and land-use change,14,54−56 this is nottaken into account here for consistency and ease of comparisonwith other traditional policy impact studies.

3. RESULTS AND DISCUSSION

3.1. Impact of Delayed Action on Global HgDeposition. The impact of delayed action on global Hgdeposition was quantified using the fully coupled seven-reservoir GBC model11,16 (FUTURE simulations, see Table1). Figure 2a shows global anthropogenic emissions to theatmosphere from 1950 onward. Emissions rise steadily after1950 due to increased coal use and artisanal gold mining.19

Over recent years, decreasing emissions in Europe and NorthAmerica due to domestic regulation have been offset by anincrease in East Asia, leading to an overall increase.9,19 GlobalHg deposition is depicted in Figure 2b. The atmosphere

Figure 2. (a) Global primary anthropogenic emissions of Hg to the atmosphere (in Mg). A New Policy (NP) is implemented in 2010 (black), 2020(blue), 2025 (red), 2030 (green), 2035 (yellow) or 2050 (purple). The NP annual emissions target9 is reached at a rate of −32.7 Mg yr−1even incase of delayed global action. (b) Global atmospheric Hg deposition to ecosystems (in Mg). Each 5 year delay in implementing NP delays byadditional extra ca. 4 years the return of Hg deposition to its year 2010 level (chosen for illustrative purposes) due to legacy emissions.

Environmental Science & Technology Policy Analysis

DOI: 10.1021/acs.est.8b04542Environ. Sci. Technol. 2018, 52, 12968−12977

12971

responds relatively quickly (though not proportionally) todecreasing emissions. If a NP scenario is implemented in 2020,deposition begins to decrease by 2021. In this scenario, globalprimary emissions decrease by ca. 980 Mg from 2020 to 2050while deposition decreases by ca. 635 Mg. These results reflectthe balance and cycling of Hg between the various reservoirs.Return to year 2010 deposition (arbitrary threshold) isachieved in 2038, that is, 18 years after NP implementation.On the other hand, return to year 2010 deposition is achievedin 2027 if a more stringent MFR scenario is implemented in2020 (see SI Figure S3).To evaluate the impact of delayed global action, a NP

scenario was implemented for various years between 2020 and2050. Return to year 2010 deposition level is reached in 2038,2047, 2056, or 2064 if a NP scenario is implemented in 2020,2025, 2030, or 2035, respectively. On average, each near-term5 year delay in implementing a NP scenario in turns delays byadditional extra ca. 4 years a return to its year 2010 level (thislevel is not the goal of policy action but used here forillustrative purposes). Each near-term 5 year delay leads to aca. 2.2% increase of the atmospheric reservoir mass, mainly dueto the feedback from legacy emissions.Based on these results, we also compare the emissions

reduction rate needed in order to reach the same givendeposition target at the same time but assuming delayed globalaction. According to Figure 2b, return to year 2010 depositionlevel is reached in 2038 if emissions reduction is initiated in2020 at a −32.7 Mg yr−1 reduction rate (NP). To reach thesame deposition target (year 2010 level) at the same time(2038), reduction rates of −48.0, −83.0, and −230.0 Mg yr−1

are needed if emissions reduction is initiated in 2025, 2030, or2035, respectively. In other words, emissions reduction mustbe ca. 1.5, 2.5, or 7.0 times more stringent if initiated in 2025,2030, or 2035, respectively, instead of 2020 due to legacyemissions and increasingly shortened recovery periods.

3.2. Local Consequences and Percent Change inPolicy Impacts. Using the integrated modeling approachdescribed in Section 2.1, we calculated the local consequencesof delayed global action for Hg deposition fluxes to ecosystemsat the four selected sites (see Section 2.4). Following atraditional atmospheric modeling method (i.e., no delay/legacypenalty−see Figure 1), the NP scenario leads to a 15.3%,55.1%, and 12.9% decrease in Hg deposition (vs present-daylevels) in Maine, Shanghai, and the South Pacific, respectively(see Table 2). In contrast, a 25.9% increase is observed inAhmedabad, India due to projected growth of regionalanthropogenic emissions.9 The MFR scenario leads toconsistent global-scale Hg deposition reduction, with a25.8%, 38.4%, 68.3%, and 22.1% decrease in Maine,Ahmedabad, Shanghai, and the South Pacific, respectively.These results are consistent with those reported by Pacyna etal.9 despite the use of the a different CTM with a formulation,spatial resolution, and physical and chemical process para-metrizations considerably different from GLEMOS andECHMERIT.30

As expected, policy impacts are lower in remote regions, farfrom emissions sources.9,48,57 While North America contrib-utes a significant fraction of global anthropogenic emis-sions,17,58,59 Hg emissions are low in Maine60 (∼50 kg yr−1)and in the neighboring New England states (see SI Figures S4and S5) due to a lack of major emitting sources as well as theadoption in 1998 of a regional Hg action plan with aggressiveemission reduction goals.61 Based on Figure 1 in Giang andSelin,57 little impact is expected from domestic U.S. regulationsin eastern Maine in terms of avoided deposition. As inferred by2007−2016 hourly air back-trajectories computed with theHYSPLIT model62 (see SI Figure S6), Maine tribal areas aremainly influenced by air masses originating from Canada andthe Arctic (Hudson Bay), that is, the Northern Hemisphereatmospheric background, rather than U.S. emissions. Sunder-land et al.63 also showed that, in the early 2000s,

Table 2. Policy Impact on Local Hg Deposition (Δμg/m2/yr, Percent in Parentheses) Assuming an ImmediateImplementation and No Legacy Penalty (Traditional Method)a and Percent Change in Policy Impact Depending on Year ofImplementation, i.e., Length of the Delayb

policy impact in Δμg/m2/yr (% change vs present) percent change in policy impact (%)

year of implementation traditional method 2020 2025 2030 2035 2050 mean

Global AverageNP −1.4 (−13.6%) −33.6 −48.5 −62.7 −76.2 −114.1 −14.2MFR −2.5 (−23.8%) −19.1 −27.5 −35.5 −43.2 −64.6 −8.0global legacy penalty (Mg) 0.0 272 392 506 615 921 114.5

Maine (A)NP −2.9 (−15.3%) −29.5 −42.6 −55.0 −66.9 −100.0 −12.5MFR −4.9 (−25.8%) −17.5 −25.3 −32.7 −39.7 −59.5 −7.4

Ahmedabad, India (B)NP +4.2 (+25.9%) −12.7 −18.4 −23.7 −28.9 −43.2 −5.4MFR −6.3 (−38.4%) −8.6 −12.4 −16.0 −19.5 −29.2 −3.6

Shanghai, China (C)NP −6.3 (−55.1%) −2.7 −3.9 −5.0 −6.1 −9.1 −1.1MFR −7.8 (−68.3%) −2.2 −3.1 −4.1 −4.9 −7.4 −0.9

South Pacific (D)NP −1.7 (−12.9%) −38.3 −55.3 −71.4 −86.9 −130.1 −16.2MFR −3.0 (−22.1%) −22.8 −32.8 −42.4 −51.6 −77.2 −9.6

aThe policy impact is calculated as the difference in deposition between FUTURE and BASE simulations. bThe percent change in policy impact iscalculated according to eq 1. The mean percent change in policy impact is the average percent change due to a near-term (2020−2035) 5 yeardelay, calculated according to eq 2. New Policy (NP) and Maximum Feasible Reduction (MFR) refer to future global emissions scenariosdeveloped by Pacyna et al.9.

Environmental Science & Technology Policy Analysis

DOI: 10.1021/acs.est.8b04542Environ. Sci. Technol. 2018, 52, 12968−12977

12972

anthropogenic emissions in the U.S. and Canada resulted in∼30% of Hg deposition to the Gulf of Maine, with the rest(∼70%) from global anthropogenic and natural sources. Inthat context, a decrease of Hg deposition in Maine tribal areascan only be achieved through the reduction of the globalbackground Hg concentration, that is, through global action.Potential additional effects of global change (climate, biomassburning, land use) on Hg deposition have recently beeninvestigated in Michigan’s Upper Peninsula and projected tohave modest impacts compared to changes in directanthropogenic emissions.48

Figure 3 depicts the mean percent change in policy impactsdue to a near-term (2020−2035) 5 year delayed implementa-

tion of a NP scenario. On a global average, each 5 year delayleads to a ca. 14% decrease in NP impacts. The consequencesof delayed global action depend on the stringency of the policyas each 5 year delay leads to a ca. 8% decrease in MFR impacts(global average, see Table 2). Remote regions are proportion-ally more impacted by delayed global action than regions closeto emission sources and a clear gradient between the Northernand Southern Hemispheres can be observed (see Figure 3).While a 5 year delay leads to a −12.5 and −16.2% change inNP impacts in Maine and South Pacific, respectively, it inducesa −5.4 and −1.1% change in Ahmedabad and Shanghai,respectively (see Table 2 and Figure 3). This can be explainedby the relatively lower policy impact in remote regions (seeabove) and therefore proportionally higher influence of thelegacy penalty. Consequences in terms of human exposurethrough fish consumption are further discussed in the nextsection, with a specific focus on remote inland waters ofeastern Maine, USA.3.3. Local Impacts on Fish Contamination: A Tribal

Case Study. In the U.S., rates of fish consumption and type offish consumed vary widely. Whereas fish forms a smallcomponent of the diet of many Americans, some groupssuch as Native American tribes eat fish as frequently asdaily.65−67 Additionally, fishing is an important component ofcultural and religious practice for many Native Americans.68

Therefore, fish contamination poses special risks for tribalmembers and is an issue relevant to environmental justice.68

According to a study performed in Maine tribal areas,69 totalHg concentration in predatory fish exceeds the 0.3 mg kg−1

U.S. EPA threshold70 in 16 out of 20 lakes (see SI Figure S7)despite their remoteness from emissions sources (see Section3.2). As discussed by Perlinger et al.,48 the susceptibility oflakes to being contaminated depends on the total supply of Hgto lakes but more importantly on factors leading to productionand accumulation of MeHg (e.g., prevalence of forests andwetlands in the catchment, low alkalinity, pH or nutrients, andlong water residence time). In order to navigate the gapbetween safe and desired fish consumption levels forpopulations with significant exposure to Hg, it is necessaryto model changes in fish contamination over time71 and toinvestigate the response to delayed global action.The median response (over 20 lakes, see SI Section 1.2) of

lacustrine predatory fish contamination to changes inatmospheric deposition can be seen in Figure 4. All the lakes

within the study area respond with a rapid decrease in MeHgconcentration over a decade, followed by a slower declinetoward steady state. Using the SERAFM model, Knightes etal.39 modeled the response to a hypothetical 50% decline indeposition across a range of lake types and also found a similarresponse. However, the time response reported here is at theupper range of those reported in Knightes et al.39 Assuming animmediate implementation of a NP scenario (i.e., no delay/legacy penalty−traditional method), the median MeHgconcentration in predatory fish rapidly declines by ca. 11%.This suggests that even in the case of an immediate NPimplementation and of a rather fast response time to changesin deposition, the median MeHg concentration is still abovethe U.S. EPA threshold. These results are in line with thoserecently reported in Michigan’s Upper Peninsula.48 Substantialand rapid response of fish contamination to reduced emissionshave been observed near emissions sources.72−74 However, therelatively lower policy impacts in remote areas (see Section3.2) are likely to considerably prolong the time required to seesubstantial decreases in fish Hg concentrations. Under an

Figure 3. Mean percent change in policy impacts due to a near-term(2020−2035) 5-year delayed implementation of a New Policy (NP)scenario. Results are discussed at selected sites with varying impactfrom emissions sources, focusing on (A) tribal areas of eastern Maine,(B) Ahmedabad, India, (C) Shanghai, China, and (D) an area of theSouthern Pacific known for albacore tuna fisheries. This Figure wasmade using the R package autoimage.64

Figure 4. Median response of eastern Maine lacustrine predatory fishcontamination to delayed implementation of a New Policy (NP) orMaximum Feasible Reduction (MFR) scenario. Black dashed line:year 2010 MeHg concentration. Red dashed line: U.S. EPA referencedose for MeHg (0.3 mg kg−1).

Environmental Science & Technology Policy Analysis

DOI: 10.1021/acs.est.8b04542Environ. Sci. Technol. 2018, 52, 12968−12977

12973

immediate implementation of a MFR scenario, the medianMeHg concentration drops to about 0.3 mg kg−1, that is, theU.S. EPA threshold. A desired subsistence fish consumption of300−500 g per day requires a safe level target of ∼0.018 mgkg−1.48,66,71 The predatory fish Hg concentrations under theMFR scenario therefore indicate that, even under the strictestglobal Hg regulations, a traditional-subsistence diet high inpredatory fishes (e.g., brook trout, brown trout, burbot,landlocked salmon, smallmouth bass) will lead to unsafeMeHg exposure in Maine tribal areas. While flawed from thestandpoint of environmental justice, a diet that shifts towardmixed feeders (e.g., white sucker, also known as mullet or bayfish) would reduce MeHg exposure (see SI Figure S1). Thissuggestion also holds true for riverine fish in the study area.75

Although not a true substitute for a pristine environment,another alternative is the increasing consumption of lower-Hgcontaining fish from aquaculture.55 The Aroostook Band ofMicmacs located in Presque Isle (see SI Figure S2) hasrecently made this choice and created a recirculatingaquaculture brook trout fish hatchery.76

The impact of delayed global action on MeHg fishconcentrations was then evaluated. Return to year 2010MeHg level (an arbitrarily defined threshold used here forillustrative purposes) is achieved in 2021, 2027, 2033, or 2043if a NP scenario is implemented in 2020, 2025, 2030, or 2035,respectively. In other words, the longer the delay, the longer ittakes to reach the same MeHg concentration target. While theU.S. EPA threshold is reached in a decade if a MFR scenario isimplemented in 2010 (see Figure 4), this target is neverreached in case of delayed MFR implementation (see SI FigureS8).3.4. Uncertainties in Atmospheric Hg Modeling.

Uncertainties in atmospheric Hg modeling for policyevaluation, in particular for linking sources to receptors, haverecently been thoroughly discussed by Kwon and Selin.13

Major uncertainties arise from biogeochemical cycling,atmospheric chemistry, and anthropogenic emissions. Thereis for example ongoing controversy in the literature and arapidly evolving understanding of Hg pool sizes and fluxes inthe global Hg cycle.13,22,77−81 The fully coupled seven-reservoir GBC model11,16 used in this study to scale legacyemissions is based on Streets et al.’s19 all-time emissioninventory, which assumes a major atmospheric Hg impact fromlate 19th century Gold Rush mining in North America. Recentstudies, including historical documents on Hg use, oregeochemistry and a large array of ice and lake sedimentrecords, have challenged this account, as documented in acritical review by Outridge et al.77 This synthesis argues for a“low-mining emissions” scenario which translates into smallerlegacy pools in the oceans and soils than considered until now.In order to investigate the consequences of a smaller legacypool on the calculated legacy penalty, we cut historical (1850−1920 CE) mining emissions in the Streets et al.19 inventory by50%, as proposed by Engstrom et al.78 This “low-miningemissions” scenario did not lead to any significant change inthe influence of delayed action on NP policy impacts. Assuggested by Amos et al.,79 atmospheric deposition is mostsensitive to the profile of anthropogenic emissions in recentdecades, and results presented here are robust to theuncertainty in historical emissions. Our results also dependon the emissions scenario used from year 2010 untilimplementation of the NP scenario. We conducted aperturbation analysis to investigate the effect of this

uncertainty on the percent change in policy impact. Wecalculated a mean legacy penalty for each near-term 5-yeardelay of 107 Mg, 115 Mg (see Table 2), or 123 Mg assumingconstant emissions (+ 0 Mg yr−1), a CP scenario (+3.02 Mgyr−1), or a 2 × CP (+6.04 Mg yr−1) rate. We find that, on aglobal average, each 5-year delay leads to a 13%, 14%, or 15%decrease in NP policy impacts, respectively. Perturbations tothe legacy penalty therefore have little impact on our results.

3.5. Implications. Our results show that traditionalspatially resolved analyses of prospective policy impacts frommercury reductions that do not consider future changes inlegacy emissions can overestimate changes driven by policyimplementation by up to 110% by 2050 and should beconsidered as best-case estimates. Though legacy impacts havepreviously been evaluated at global scale using globalbiogeochemical cycling models, these effects are not widelyappreciated by policy-makers. Selin12 recently proposed aglobal metric to help policy-makers better understand theimplications of policy options by taking into account near-termchanges in legacy Hg. Using the integrated modeling approachdescribed here, the effect of legacy Hg changes on policyimpact can be resolved spatially. Future work could build uponthis study by (re)examining the impact of other or upcomingfuture emission scenarios beyond those developed by Pacynaet al.9 evaluated here. Future policy analyses should account forfuture legacy emissions and associated deposition to betterinform policy decision-making; the approach outlined hereprovides a straightforward methodology to estimate this effectwithout relying on advanced coupled atmosphere−oceanmodels. Alternately, an even simpler approach could scalelegacy emissions and resulting deposition globally using aglobal-scale estimate of biogeochemical model output12 ifrunning a global biogeochemical cycle is infeasible.Our results highlight the benefits of near-term aggressive Hg

mitigation efforts. Return to year 2010 global deposition isachieved 2.6 times faster under a MFR vs NP scenario (seeSection 3.1). Contrary to the NP scenario, the MFR scenarioleads to consistent global-scale Hg deposition reduction (seeSection 3.2). We also show that each near-term delay in takingglobal action to reduce emissions has a non-negligibleinfluence on expected policy impacts due to legacy emissions.Global emissions reduction must be ca. 1.5, 2.5, or 7.0 timesmore stringent if initiated in 2025, 2030, or 2035, respectively,instead of 2020 (see Section 3.1). On a global average, each 5-year delay leads to a ca. 14% decrease in NP impacts (seeSection 3.2). Finally, while the median MeHg concentration inpredatory fish in Maine lakes is still ca. 25% too high for safefish consumption in case of an immediate NP implementation,the U.S. EPA threshold is achieved under immediateimplementation of a MFR scenario (see Section 3.3).However, this level is never reached if policy is delayed. Itshould also be emphasized that under a business-as-usualscenario (CP), deposition fluxes to ecosystems will graduallyincrease. Even if moderately delayed, NP and MFR scenarioslead to reductions in Hg deposition and MeHg concentration,highlighting the positive impact of concerted global action.

■ ASSOCIATED CONTENT

*S Supporting InformationThe Supporting Information is available free of charge on theACS Publications website at DOI: 10.1021/acs.est.8b04542.

Environmental Science & Technology Policy Analysis

DOI: 10.1021/acs.est.8b04542Environ. Sci. Technol. 2018, 52, 12968−12977

12974

Results of tests performed to check the robustness of ourmethod and a detailed description of the SERAFMmodel parametrization are available. Additional Tablesand Figures are also available: Summary of somecharacteristics of the modeled lakes (Table S1), Lakegeometry values from a lake in Mighigan’s UpperPeninsula (Table S2), Ratios of lake characteristics froma lake in Mighigan’s Upper Peninsula (Table S3), LakeHg model evaluation (Figure S1), Location of Maineand Tribal communities (Figure S2), Global impact ofNP vs MFR implementation (Figure S3), Hg emissionsin Maine (Figure S4), Year 2011 U.S. state-level Hgemissions (Figure S5), Origin of air masses influencingMaine tribal areas (Figure S6), Total Hg concentrationin predatory fish fillets collected in Maine tribal areas(Figure S7), and Median response of Maine lacustrinepredatory fish contamination to delayed implementationof a MFR scenario (Figure S8) (PDF)

■ AUTHOR INFORMATION

Corresponding Author*E-mail: angot@mit.edu.

ORCIDHelene Angot: 0000-0003-4673-8249NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThis work was supported by the MIT Leading Technology &Policy Program, a core center grant P30-ES002109 from theNational Institute of Environmental Health Sciences −National Institute of Health, the MIT EAPS UndergraduateResearch Opportunities Program, the U.S. National ScienceFoundation through grant #ICER-1313755, and the NationalInstitute of Environmental Health Sciences Superfund BasicResearch Program − National Institute of Health (P42ES027707). We thank J. Pacyna and S. Cinnirella for providingfuture emissions inventories. H.A. acknowledges F. Corey, D.Macek, S. Venno, A. Ajmani, B. Longfellow, C. Johnson, N.Dalrymple, and K. M. Vandiver for rewarding discussionsduring field trips to Maine.

■ REFERENCES(1) Selin, N. E. Global Biogeochemical Cycling of Mercury: AReview. Annual Review of Environment and Resources 2009, 34 (1),43−63.(2) Horowitz, H. M.; Jacob, D. J.; Zhang, Y.; Dibble, T. S.; Slemr, F.;Amos, H. M.; Schmidt, J. A.; Corbitt, E. S.; Marais, E. A.; Sunderland,E. M. A New Mechanism for Atmospheric Mercury Redox Chemistry:Implications for the Global Mercury Budget. Atmos. Chem. Phys.2017, 17 (10), 6353−6371.(3) Gustin, M. S.; Lindberg, S. E.; Weisberg, P. J. An Update on theNatural Sources and Sinks of Atmospheric Mercury. Appl. Geochem.2008, 23, 482−493.(4) Mergler, D.; Anderson, H. A.; Chan, L. H. M.; Mahaffey, K. R.;Murray, M.; Sakamoto, M.; Stern, A. H. Methylmercury Exposure andHealth Effects in Humans: A Worldwide Concern. Ambio 2007, 36(1), 3−11.(5) Du, B.; Feng, X.; Li, P.; Yin, R.; Yu, B.; Sonke, J. E.; Guinot, B.;Anderson, C. W. N.; Maurice, L. Use of Mercury Isotopes to QuantifyMercury Exposure Sources in Inland Populations, China. Environ. Sci.Technol. 2018, 52 (9), 5407−5416.

(6) Sunderland, E. M. Mercury Exposure from Domestic andImported Estuarine and Marine Fish in the U.S. Seafood Market.Environ. Health Perspect. 2007, 115 (2), 235−242.(7) UNEP. Text of the Minamata Convention on Mercury, Availableat: Http://Mercuryconvention.Org/Portals/11/Documents/Booklets/COP1%20version/Minamata-Convention-Booklet-Eng-Full.Pdf (accessed 18 April 2018. 2017).(8) Ancora, M. P.; Zhang, L.; Wang, S.; Schreifels, J. J.; Hao, J.Meeting Minamata: Cost-Effective Compliance Options for Atmos-pheric Mercury Control in Chinese Coal-Fired Power Plants. EnergyPolicy 2016, 88 (C), 485−494.(9) Pacyna, J. M.; Travnikov, O.; De Simone, F.; Hedgecock, I. M.;Sundseth, K.; Pacyna, E. G.; Steenhuisen, F.; Pirrone, N.; Munthe, J.;Kindbom, K. Current and Future Levels of Mercury AtmosphericPollution on a Global Scale. Atmos. Chem. Phys. 2016, 16 (19),12495−12511.(10) Selin Noelle, E. Global Change and Mercury Cycling:Challenges for Implementing a Global Mercury Treaty. Environ.Toxicol. Chem. 2014, 33 (6), 1202−1210.(11) Amos, H. M.; Jacob, D. J.; Streets, D. G.; Sunderland, E. M.Legacy Impacts of All-Time Anthropogenic Emissions on the GlobalMercury Cycle. Global Biogeochem. Cycles 2013, 27 (2), 410−421.(12) Selin, N. E. A Proposed Global Metric to Aid MercuryPollution Policy. Science 2018, 360 (6389), 607−609.(13) Kwon, S. Y.; Selin, N. E. Uncertainties in Atmospheric MercuryModeling for Policy Evaluation. Curr. Pollution Rep 2016, 2 (2), 103−114.(14) Obrist, D.; Kirk, J. L.; Zhang, L.; Sunderland, E. M.; Jiskra, M.;Selin, N. E. A Review of Global Environmental Mercury Processes inResponse to Human and Natural Perturbations: Changes ofEmissions, Climate, and Land Use. Ambio 2018, 47 (2), 116−140.(15) Giang, A.; Stokes, L. C.; Streets, D. G.; Corbitt, E. S.; Selin, N.E. Impacts of the Minamata Convention on Mercury Emissions andGlobal Deposition from Coal-Fired Power Generation in Asia.Environ. Sci. Technol. 2015, 49 (9), 5326−5335.(16) Amos, H. M.; Jacob, D. J.; Kocman, D.; Horowitz, H. M.;Zhang, Y.; Dutkiewicz, S.; Horvat, M.; Corbitt, E. S.; Krabbenhoft, D.P.; Sunderland, E. M. Global Biogeochemical Implications of MercuryDischarges from Rivers and Sediment Burial. Environ. Sci. Technol.2014, 48 (16), 9514−9522.(17) AMAP/UNEP. Technical Background Report for the GlobalMercury Assessment 2013, Arctic Monitoring and AssessmentPorgramme, Oslo, Norway; UNEP Chemicals Branch, Geneva,Switzerland, 2013; p 263.(18) Zhang, Y.; Jacob, D. J.; Horowitz, H. M.; Chen, L.; Amos, H.M.; Krabbenhoft, D. P.; Slemr, F.; Louis, V. L. S.; Sunderland, E. M.Observed Decrease in Atmospheric Mercury Explained by GlobalDecline in Anthropogenic Emissions. Proc. Natl. Acad. Sci. U. S. A.2016, 113 (3), 526−531.(19) Streets, D. G.; Devane, M. K.; Lu, Z.; Bond, T. C.; Sunderland,E. M.; Jacob, D. J. All-Time Releases of Mercury to the Atmospherefrom Human Activities. Environ. Sci. Technol. 2011, 45 (24), 10485−10491.(20) Pirrone, N.; Cinnirella, S.; Feng, X.; Finkelman, R.; Friedli, H.;Leaner, J.; Mason, R.; Mukherjee, A.; Stracher, G.; Streets, D.;Telmer, K. Global Mercury Emissions to the Atmosphere fromAnthropogenic and Natural Sources. Atmos. Chem. Phys. 2010, 10,5951−5964.(21) Bagnato, E.; Aiuppa, A.; Parello, F.; Allard, P.; Shinohara, H.;Liuzzo, M.; Giudice, G. New Clues on the Contribution of Earth’sVolcanism to the Global Mercury Cycle. Bull. Volcanol. 2011, 73 (5),497−510.(22) Horowitz, H. M.; Jacob, D. J.; Amos, H. M.; Streets, D. G.;Sunderland, E. M. Historical Mercury Releases from CommercialProducts: Global Environmental Implications. Environ. Sci. Technol.2014, 48 (17), 10242−10250.(23) Bey, I.; Jacob, D. J.; Yantosca, R. M.; Logan, J. A.; Field, B. D.;Fiore, A. M.; li, Q.; Liu, H. Y.; Mickley, L. J.; Schultz, M. G. GlobalModeling of Tropospheric Chemistry with Assimilated Meteorology:

Environmental Science & Technology Policy Analysis

DOI: 10.1021/acs.est.8b04542Environ. Sci. Technol. 2018, 52, 12968−12977

12975

Model Description and Evaluation. Journal of geophysical research2001, 106 (D19), 23,073−23,095.(24) Holmes, C. D.; Jacob, D. J.; Corbitt, E. S.; Mao, J.; Yang, X.;Talbot, R.; Slemr, F. Global Atmospheric Model for MercuryIncluding Oxidation by Bromine Atoms. Atmos. Chem. Phys. 2010,10, 12037−12057.(25) Amos, H. M.; Jacob, D. J.; Holmes, C. D.; Fisher, J. A.; Wang,Q.; Yantosca, R. M.; Corbitt, E. S.; Galarneau, E.; Rutter, A. P.;Gustin, M. S.; Steffen, A.; Schauer, J. J.; Graydon, J. A.; Louis, V. L.;Talbot, R. W.; Edgerton, E. S.; Zhang, Y.; Sunderland, E. M. Gas-Particle Partitioning of Atmospheric Hg(II) and Its Effect on GlobalMercury Deposition. Atmos. Chem. Phys. 2012, 12 (1), 591−603.(26) Soerensen, A. L.; Sunderland, E. M.; Holmes, C. D.; Jacob, D.J.; Yantosca, R. M.; Skov, H.; Christensen, J. H.; Strode, S. A.; Mason,R. P. An Improved Global Model for Air-Sea Exchange of Mercury:High Concentrations over the North Atlantic. Environ. Sci. Technol.2010, 44 (22), 8574−8580.(27) Selin, N. E.; Jacob, D. J.; Yantosca, R. M.; Strode, S.; Jaegle, L.;Sunderland, E. M. Global 3-D Land-Ocean-Atmosphere Model forMercury: Present-Day versus Preindustrial Cycles and AnthropogenicEnrichment Factors for Deposition. Global Biogeochem. Cycles 2008,22 (2), GB2011.(28) Schmidt, J. A.; Jacob, D. J.; Horowitz, H. M.; Hu, L.; Sherwen,T.; Evans, M. J.; Liang, Q.; Suleiman, R. M.; Oram, D. E.; Le Breton,M.; Percival, C. J.; Wang, S.; Dix, B.; Volkamer, R. Modeling theObserved Tropospheric BrO Background: Importance of MultiphaseChemistry and Implications for Ozone, OH, and Mercury. J. Geophys.Res. Atmos. 2016, 121 (19), 2015JD024229.(29) Hg and POPs working group. GEOS-Chem mercurysimulations http://wiki.seas.harvard.edu/geos-chem/index.php/Mercury#K_RED_JNO2 (accessed May 15, 2018).(30) Travnikov, O.; Angot, H.; Artaxo, P.; Bencardino, M.; Bieser, J.;D’Amore, F.; Dastoor, A.; De Simone, F.; Dieguez, M. D. C.;Dommergue, A.; Ebinghaus, R.; Feng, X. B.; Gencarelli, C. N.;Hedgecock, I. M.; Magand, O.; Martin, L.; Matthias, V.; Mashyanov,N.; Pirrone, N.; Ramachandran, R.; Read, K. A.; Ryjkov, A.; Selin, N.E.; Sena, F.; Song, S.; Sprovieri, F.; Wip, D.; Wangberg, I.; Yang, X.Multi-Model Study of Mercury Dispersion in the Atmosphere:Atmospheric Processes and Model Evaluation. Atmos. Chem. Phys.2017, 17 (8), 5271−5295.(31) Selin, N. E.; Jacob, D. J.; Park, R. J.; Yantosca, R. M.; Strode, S.;Jaegle, L.; Jaffe, D. A. Chemical Cycling and Deposition ofAtmospheric Mercury: Global Constraints from Observations. J.Geophys. Res. 2007, 112, D02308.(32) Guttikunda, S. K.; Jawahar, P. Atmospheric Emissions andPollution from the Coal-Fired Thermal Power Plants in India. Atmos.Environ. 2014, 92, 449−460.(33) Duan, L.; Wang, X.; Wang, D.; Duan, Y.; Cheng, N.; Xiu, G.Atmospheric Mercury Speciation in Shanghai, China. Sci. TotalEnviron. 2017, 578, 460−468.(34) Han, D.; Zhang, J.; Hu, Z.; Ma, Y.; Duan, Y.; Han, Y.; Chen, X.;Zhou, Y.; Cheng, J.; Wang, W. Particulate Mercury in Ambient Air inShanghai, China: Size-Specific Distribution, Gas−Particle Partition-ing, and Association with Carbonaceous Composition. Environ. Pollut.2018, 238, 543−553.(35) Chen, X.; Balasubramanian, R.; Zhu, Q.; Behera, S. N.; Bo, D.;Huang, X.; Xie, H.; Cheng, J. Characteristics of AtmosphericParticulate Mercury in Size-Fractionated Particles during Haze Daysin Shanghai. Atmos. Environ. 2016, 131, 400−408.(36) Lehodey, P.; Senina, I.; Nicol, S.; Hampton, J. Modelling theImpact of Climate Change on South Pacific Albacore Tuna. Deep SeaRes., Part II 2015, 113, 246−259.(37) Sunderland, E. M.; Li, M.; Bullard, K. Decadal Changes in theEdible Supply of Seafood and Methylmercury Exposure in the UnitedStates. Environ. Health Perspect. 2018, 126 (1), 017006.(38) Knightes, C. D. Development and Test Application of aScreening-Level Mercury Fate Model and Tool for Evaluating WildlifeExposure Risk for Surface Waters with Mercury-Contaminated

Sediments (SERAFM). Environmental Modelling & Software 2008,23 (4), 495−510.(39) Knightes, C. D.; Sunderland, E. M.; Barber, M. C.; Johnston, J.M.; Ambrose, R. B. Application of Ecosystem-Scale Fate andBioaccumulation Models to Predict Fish Mercury Response Timesto Changes in Atmospheric Deposition. Environ. Toxicol. Chem. 2009,28 (4), 881−893.(40) Brown, S.; Saito, L.; Knightes, C.; Gustin, M. Calibration andEvaluation of a Mercury Model for a Western Stream andConstructed Wetland. Water, Air, Soil Pollut. 2007, 182 (1−4),275−290.(41) Selin, N. E.; Sunderland, E. M.; Knightes, C. D.; Mason, R. P.Sources of Mercury Exposure for US Seafood Consumers:Implications for Policy. Environ. Health Perspect. 2010, 118 (1),137−143.(42) Hong, Y.-S.; Rifkin, E.; Bouwer, E. J. Modeling MercuryDistribution in the Sarasota Bay Ecosystem Using SERAFM andStable Isotope Ratios of Nitrogen (Δ15N) in Biota. Environ. Eng. Sci.2014, 31 (3), 135−147.(43) Hope Bruce, K.; Louch, J. Pre-anthropocene Mercury Residuesin North American Freshwater Fish. Integr. Environ. Assess. Manage.2013, 10 (2), 299−308.(44) Hope, B. K. Efficacy of Translators for Establishing MercuryTotal Maximum Daily Loads. Hum. Ecol. Risk Assess. 2011, 17 (6),1263−1278.(45) Gidley, P. T.; Kreitinger, J. P.; Zakikhani, M.; Suedel, B. C.Methylmercury Screening Models for Surface Water Habitat Restoration:A Case Study in Duluth-Superior Harbor, ERDC/EL TR-17−19;ERDC-EL Vicksburg United States; ERDC-EL: Vicksburg UnitedStates, 2017.(46) Zhu, S.; Zhang, Z.; Zagar, D. Mercury Transport and FateModels in Aquatic Systems: A Review and Synthesis. Sci. TotalEnviron. 2018, 639, 538−549.(47) Hendricks, A. N. A Model to Predict Concentrations andUncertainty for Mercury Species in Lakes. Open Access Master’s Thesis,Michigan Technological University: http://digitalcommons.mtu.edu/etdr/585, 2018.(48) Perlinger, J. A.; R. Urban, N.; Giang, A.; E. Selin, N.; N.Hendricks, A.; Zhang, H.; Kumar, A.; Wu, S.; S. Gagnon, V.; S.Gorman, H.; Norman, E. Responses of Deposition and Bioaccumu-lation in the Great Lakes Region to Policy and Other Large-ScaleDrivers of Mercury Emissions. Environmental Science: Processes &Impacts 2018, 20 (1), 195−209.(49) Hakanson, L. The Derivation and Use of a Dynamic Model forMercury in Lake Fish Based on a Static (Regression) Model. Water,Air, Soil Pollut. 2000, 124 (3−4), 301−317.(50) Barber, M. C. Dietary Uptake Models Used for Modeling theBioaccumulation of Organic Contaminants in Fish. Environ. Toxicol.Chem. 2008, 27 (4), 755−777.(51) Thevenon, F.; Graham, N. D.; Chiaradia, M.; Arpagaus, P.;Wildi, W.; Pote, J. Local to Regional Scale Industrial Heavy MetalPollution Recorded in Sediments of Large Freshwater Lakes inCentral Europe (Lakes Geneva and Lucerne) over the Last Centuries.Sci. Total Environ. 2011, Complete (412−413), 239−247.(52) Glass, G. E.; Sorensen, J. A.; Schmidt, K. W.; Rapp, G. R. NewSource Identification of Mercury Contamination in the Great Lakes.Environ. Sci. Technol. 1990, 24 (7), 1059−1069.(53) Kuhnlein, H. V.; Chan, H. M. Environment and Contaminantsin Traditional Food Systems of Northern Indigenous Peoples. Annu.Rev. Nutr. 2000, 20 (1), 595−626.(54) Eagles-Smith, C. A.; Silbergeld, E. K.; Basu, N.; Bustamante, P.;Diaz-Barriga, F.; Hopkins, W. A.; Kidd, K. A.; Nyland, J. F.Modulators of Mercury Risk to Wildlife and Humans in the Contextof Rapid Global Change. Ambio 2018, 47 (2), 170−197.(55) Krabbenhoft, D. P.; Sunderland, E. M. Global Change andMercury. Science 2013, 341 (6153), 1457−1458.(56) Ahonen, S. A.; Hayden, B.; Leppanen, J. J.; Kahilainen, K. K.Climate and Productivity Affect Total Mercury Concentration and

Environmental Science & Technology Policy Analysis

DOI: 10.1021/acs.est.8b04542Environ. Sci. Technol. 2018, 52, 12968−12977

12976

Bioaccumulation Rate of Fish along a Spatial Gradient of SubarcticLakes. Sci. Total Environ. 2018, 637−638, 1586−1596.(57) Giang, A.; Selin, N. E. Benefits of Mercury Controls for theUnited States. Proc. Natl. Acad. Sci. U. S. A. 2016, 113 (2), 286−291.(58) Muntean, M.; Janssens-Maenhout, G.; Song, S.; Selin, N. E.;Olivier, J. G. J.; Guizzardi, D.; Maas, R.; Dentener, F. Trend Analysisfrom 1970 to 2008 and Model Evaluation of EDGARv4 GlobalGridded Anthropogenic Mercury Emissions. Sci. Total Environ. 2014,494−495, 337−350.(59) Muntean, M.; Janssens-Maenhout, G.; Song, S.; Giang, A.;Selin, N. E.; Zhong, H.; Zhao, Y.; Olivier, J. G. J.; Guizzardi, D.;Crippa, M.; Schaaf, E.; Dentener, F. Evaluating EDGARv4.Tox2Speciated Mercury Emissions Ex-Post Scenarios and Their Impacts onModelled Global and Regional Wet Deposition Patterns. Atmos.Environ. 2018, 184, 56−68.(60) U.S. EPA. 2011 National Emissions Inventory (NEI) Datahttps://www.epa.gov/air-emissions-inventories/2011-national-emissions-inventory-nei-data (accessed May 8, 2018).(61) Smith, C. M.; Trip, L. J. Mercury Policy and Science inNortheastern North America: The Mercury Action Plan of the NewEngland Governors and Eastern Canadian Premiers. Ecotoxicology2005, 14 (1−2), 19−35.(62) Stein, A. F.; Draxler, R. R.; Rolph, G. D.; Stunder, B. J. B.;Cohen, M. D.; Ngan, F. NOAA’s HYSPLIT Atmospheric Transportand Dispersion Modeling System. Bull. Am. Meteorol. Soc. 2015, 96(12), 2059−2077.(63) Sunderland, E. M.; Cohen, M. D.; Selin, N. E.; Chmura, G. L.Reconciling Models and Measurements to Assess Trends inAtmospheric Mercury Deposition. Environ. Pollut. 2008, 156, 526−535.(64) French, J. P. Autoimage: Multiple Heat Maps for ProjectedCoordinates. R J. 2017, 9 (1), 284−297.(65) O’Neill, C. Mercury, Risk, and Justice, SSRN Scholarly Paper ID1004769; Social Science Research Network: Rochester, NY, 2007.(66) U.S. EPA. Wabanaki Traditional Cultural Lifeways ExposureScenario https://www.epa.gov/tribal/wabanaki-traditional-cultural-lifeways-exposure-scenario (accessed May 8, 2018).(67) Xue, J.; Zartarian, V.; Mintz, B.; Weber, M.; Bailey, K.; Geller,A. Modeling Tribal Exposures to Methyl Mercury from FishConsumption. Sci. Total Environ. 2015, 533, 102−109.(68) Ranco, D. J.; O’Neill, C. A.; Donatuto, J.; Harper, B. L.Environmental Justice, American Indians and the Cultural Dilemma:Developing Environmental Management for Tribal Health and Well-Being. Environmental Justice 2011, 4 (4), 221−230.(69) U.S. EPA. Region 1 Fish Tissue Contamination QA Plan |Regional Environmental Monitoring and Assessment Program | U.S.EPA https://archive.epa.gov/emap/archive-emap/web/html/reg1_rpt.html (accessed May 8, 2018).(70) US EPA. Water Quality Criteria: Notice of Availability of WaterQuality Criterion for the Protection of Human Health: Methyl-mercury https://www.federalregister.gov/documents/2001/01/08/01-217/water-quality-criteria-notice-of-availability-of-water-quality-criterion-for-the-protection-of-human (accessed May 8, 2018).(71) Gagnon, V.; Gorman, H.; Norman, E. Power and Politics inResearch Design and Practice: Opening up Space for Social Equity inInterdisciplinary, Multi-Jurisdictional and Community-Based Re-search. Gateways: International Journal of Community Research andEngagement 2017, 10 (0), 164−184.(72) Hutcheson, M. S.; Smith, C. M.; Rose, J.; Batdorf, C.;Pancorbo, O.; West, C. R.; Strube, J.; Francis, C. Temporal andSpatial Trends in Freshwater Fish Tissue Mercury ConcentrationsAssociated with Mercury Emissions Reductions. Environ. Sci. Technol.2014, 48 (4), 2193−2202.(73) Brigham, M. E.; Sandheinrich, M. B.; Gay, D. A.; Maki, R. P.;Krabbenhoft, D. P.; Wiener, J. G. Lacustrine Responses to DecreasingWet Mercury Deposition RatesResults from a Case Study inNorthern Minnesota. Environ. Sci. Technol. 2014, 48 (11), 6115−6123.

(74) Zhou, H.; Zhou, C.; Lynam, M. M.; Dvonch, J. T.; Barres, J. A.;Hopke, P. K.; Cohen, M.; Holsen, T. M. Atmospheric MercuryTemporal Trends in the Northeastern United States from 1992 to2014: Are Measured Concentrations Responding to DecreasingRegional Emissions? Environ. Sci. Technol. Lett. 2017, 4 (3), 91−97.(75) U.S. Fish and Wildlife service. Contaminant Assessment of WhiteSuckers from Eight Rivers in the Gulf of Maine Distinct PopulationSegment for Atlantic Salmon; FY07-MEFO-1-EC; Maine Field Office,2007; p 44.(76) US Fish and Wildlife service. First Micmac Fish Harvesthttps://usfwsnortheast.wordpress.com/2017/06/28/first-micmac-fish-harvest/ (accessed May 8, 2018).(77) Outridge, P. M.; Mason, R. P.; Wang, F.; Guerrero, S.;Heimburger-Boavida, L. E. Updated Global and Oceanic MercuryBudgets for the United Nations Global Mercury Assessment 2018.Environ. Sci. Technol. 2018, 52 (20), 11466−11477.(78) Engstrom, D. R.; Fitzgerald, W. F.; Cooke, C. A.; Lamborg, C.H.; Drevnick, P. E.; Swain, E. B.; Balogh, S. J.; Balcom, P. H.Atmospheric Hg Emissions from Preindustrial Gold and SilverExtraction in the Americas: A Reevaluation from Lake-SedimentArchives. Environ. Sci. Technol. 2014, 48 (12), 6533−6543.(79) Amos, H. M.; Sonke, J. E.; Obrist, D.; Robins, N.; Hagan, N.;Horowitz, H. M.; Mason, R. P.; Witt, M.; Hedgecock, I. M.; Corbitt,E. S.; Sunderland, E. M. Observational and Modeling Constraints onGlobal Anthropogenic Enrichment of Mercury. Environ. Sci. Technol.2015, 49 (7), 4036−4047.(80) Beal, S.; Osterberg, E. C.; Zdanowicz, C.; Fisher, D. An IceCore Perspective on Mercury Pollution during the Past 600 Years.Environ. Sci. Technol. 2015, 49 (13), 7641−7647.(81) Zhang, Y.; Jaegle, L.; Thompson, L.; Streets, D. G. SixCenturies of Changing Oceanic Mercury. Global Biogeochemical Cycles2014, 28 (11), 1251−1261.

Environmental Science & Technology Policy Analysis

DOI: 10.1021/acs.est.8b04542Environ. Sci. Technol. 2018, 52, 12968−12977

12977