Projections of Global Mercury Emissions in 2050

Transcription

1 Environ. Sci. Technol. 2009, 43, Projections of Global Mercury Emissions in DAVID G. STREETS,*, QIANG ZHANG, AND YE WU Decision and Information Sciences Division, Argonne National Laboratory, Argonne, Illinois 60439, and Department of Environmental Science and Engineering, Tsinghua University, Beijing , China Received September 2, Revised manuscript received December 11, Accepted January 5, Global Hg emissions are presented for the year under a variety of assumptions about socioeconomic and technology development. We find it likely that Hg emissions will increase in the future. The range of global Hg emissions is projected to be Mg, compared to 2006 levels of 2480 Mg, reflecting a change of -4% to +96%. The main driving force for increased emissions is the expansion of coal-fired electricity generation in the developing world, particularly Asia. Our ability to arrest the growth in Hg emissions is limited by the relatively low Hg removal efficiency of the current generation of emission control technologies for coal-fired power plants (fluegas desulfurization). Large-scale deployment of advanced Hg sorbent technologies, such as Activated Carbon Injection, offers the promise of lowering the emissions range to Mg, but these technologies are not yet in commercial use. The share of elemental Hg in total emissions will decline from today s levels of 65% to 50-55% by, while the share of divalent Hg will increase. This signals a shift from long-range transport of elemental Hg to local deposition of Hg compoundssthough emissions of both species could increase under the worst case. 1. Introduction The dangerous effects of mercury (Hg) in the environment are well established, and efforts to curtail atmospheric releases of Hg have begun in many countries of the developed world. Knowing how much to reduce emissions to achieve a desired level of Hg in the environment is complicated, however, by the potential of elemental mercury (Hg0) to undergo long-range transport at hemispheric scale. The relative contributions of local sources and distant sources to Hg concentrations and deposition at particular locations are, therefore, not well-known. Further complexity is added by doubt about the relative contributions of man-made and natural sources of Hg, as well as the magnitudes of the various sinks of Hg. Indeed, the whole global Hg cycle is still under investigation. Present-day anthropogenic sources of global Hg have been studied for more than a decade. The first global emission inventories were compiled by Pacyna and Pacyna and coworkers, yielding estimates of 2140 Mg for 1990 (1) and 1910 Mg for 1995 (2, 3). Subsequent updates (4, 5) generated an * Corresponding author tel: ; fax: ; dstreets@anl.gov. Argonne National Laboratory. Tsinghua University. estimate of 2190 Mg for the year According to the most recent global inventory (4), about 65% of emissions came from stationary fuel combustion in 2000, 11% from gold production, 7% from nonferrous metals production, 6% from cement production, and the rest from a variety of smaller contributing source types. Geographically, about 54% of the emissions came from Asia, 18% from Africa, 8% from Europe, and 7% from North America. There have also been a number of studies of Hg emissions from particular regions and countries, e.g., Australia (6), China (7, 8), Europe (9), the Mediterranean (10), North America (11, 12), and South Africa (13), as well as from particular source types such as artisanal gold mining (14), petroleum products (15), and biomass burning (16-19). In addition, inferences about the magnitude of global (20, 21) and regional (22, 23) Hg emissions have been made from chemical transport modeling studies constrained by field measurements. These special studies have sometimes confirmed and sometimes disputed the values in the global inventories, but in all cases they have expanded our knowledge base about the many sources of Hg. Unification and reconciliation of current thinking about anthropogenic Hg emissions has recently been sponsored by the United Nations Environment Programme (24). There have been no published estimates to date of future global Hg emissions. We do not know whether to expect global Hg emissions to increase or decrease, nor do we know how the spatial patterns of releases might change over time. The main reason for this is the difficulty in matching up currently available projections of the future use of energy, fuels, and materials with currently available Hg emission calculation schemes. This paper remedies the situation by developing global anthropogenic Hg emission estimates for the year under four alternative IPCC scenarios of future development. 2. Methodology In previous work we studied future patterns of energy and fuel use under alternative IPCC scenarios and used them to estimate present and future emissions of the carbonaceous aerosols, black carbon (BC) and organic carbon (OC) (25). We also reported estimates of Hg emissions in the present and the future for one of the most difficult of regions to characterize, China (7, 8). In this work we marry these two schemes to project global Hg emissions in under alternative IPCC scenarios of development. The first step in the process is to develop a global Hg inventory for 1996, the base year of our BC/OC inventory (25). The BC/OC inventory consists of a detailed accounting of combustion-related energy and fuel use, disaggregated among many different sector/fuel/technology options for 17 world regions. The original list of 112 combustion options has been presented previously (25); at the time of the commencement of this Hg work, the model consisted of 144 distinct combustion options, the previous list having been augmented by the addition of some new categories like municipal waste combustion, open waste burning, liquefied petroleum gas, etc. For the purpose of calculating Hg emissions, we have added to this combustion set the eight industrial process categories that we consider the most significant for Hg release: pig iron manufacture, cement manufacture, copper smelting, lead smelting, zinc smelting, artisanal gold extraction, Hg mining, and caustic soda production. Other industrial process sources are not considered at this time. Four types of open biomass burning are included: tropical forests, extra-tropical forests, savanna/grassland, and crop /es802474j CCC: $ American Chemical Society VOL. 43, NO. 8, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY Published on Web 03/04/2009

2 TABLE 1. Emission Factors Used for Mercury Emissions in 1996 and 2006 emitting source type unit emission factor a source coal combustion power plants g/t fuel (2, 4) b industrial, residential and commercial boilers g/t fuel (2, 4) b oil combustion g/t fuel (29) biofuel combustion g/t fuel 0.02 (17) biomass burning crop residue g/t biomass (18) extra-tropical forest g/t biomass (18) savanna g/t biomass (18) tropical forest g/t biomass (18) pig iron manufacture g/t product 0.04 (2, 4) cement manufacture g/t product (30) copper smelting g/t product 5-6 (2, 4) lead smelting g/t product 3 (2, 4) zinc smelting g/t product 7-80 (2, 4, 7) a Ranges indicate values for different countries, based on level of development. b Calculated from mercury content of coals. residue burning in fields. Following precisely the same method as we used for BC and OC emissions (25), we take our forecasts of the amounts of open biomass burned (as distinct from biofuels) directly from IPCC forecasts (26-28); therefore, they are scenario-specific and consistent with the amounts of fossil fuels and biofuels burned. For biomass burning, the IPCC considers only those emissions that are directly associated with human activities. For example, forest burning is specifically associated with the need to expand agricultural land at the expense of forested land to provide additional food supplies (managed forest in IPCC terminology). The IPCC forecasts do not include the contribution from natural wildfires. We use the IMAGE managed forest projections directly and supplement them with estimates of wildfire emissions using the mature forest projections of the IPCC. Changes in mature forest area over time are assumed to be proportional to changes in wildfire emissions. The IPCC projections of grassland and crop residue burning are used directly. The total amounts of biomass burned globally decline from 6.02 Gt in 1996 to 5.71 Gt in A1B, 4.30 Gt in A2, 4.47 Gt in B1, and 4.18 Gt in B2 (see below for scenario descriptions). Using the 1996 energy and fuels data set, we then reparameterize the model to yield Hg emissions, rather than BC/OC emissions. This requires only minor modifications to the model structure, e.g., to calculate the Hg content of fuels and to add Hg removal by flue-gas desulfurization (FGD) units. Mostly it requires new data on Hg-relevant parameters such as initial release rate, ash retention efficiency, capture efficiency, etc., for each element of the [( ) 17] source matrix. In doing this we make use of data from our China inventory (7, 8), the global compilations (2-5), and U.S. EPA publications (29, 30). Table 1 summarizes our assumptions about present-day Hg emission factors. This procedure generates detailed regional Hg emission estimates in the format of our forecasting model (25). As described in the next section, our 1996 results compare well with other emission estimates, especially when recent new estimates for Australia (6) and South Africa (13) are incorporated (24). The second step of the process is to update the 1996 Hg inventory to 2006 to give a more current representation of emissions and to make the data congruent with inventories being used in the latest Hg modeling studies. We therefore replace the 1996 energy and fuels data with data for 2006 and incorporate developments in Hg capture, e.g., the penetration of FGD units between 1996 and TABLE 2. Degree of FGD Penetration in Coal-Fired Power Plants in by Scenario a A1B A2 B1 B2 Region A b Region B c Region C d a Ranges indicate values for different countries within these regions, based on level of development. Source: ref (26). b Region A: Canada, U.S., OECD Europe, Japan, and Australia. c Region B: Central America, South America, Eastern Europe, Former USSR, and Asia. d Region C: Africa. The final step is to calculate future Hg emissions. We project the 2006 inventory forward to using the same procedure reported earlier for BC/OC emissions (25). We use the IPCC forecasts developed by the IMAGE group at RIVM, the National Institute for Public Health and the Environment in The Netherlands (26). We examine four of the scenarios developed by the IPCC for its Third Assessment Report: A1B, A2, B1, and B2 (27, 28). These scenarios specify the regional energy use and fuel mix associated with alternative future pathways of human development, which can be directly used in the calculation of future emissions. The scenarios also embody different rates of technology development and technology transfer among countries. The A1B scenario is characterized by rapid energy and economic growth, low population growth, continued globalization in its western form, increasing privatization, and the rapid introduction of new and more efficient technologies. The designation B in this scenario connotes a balance of energy technologies across all source types. The A2 scenario reflects a heterogeneous world of self-reliance and preservation of local identities. Population growth is high. Technological change is slow and fragmented. The B1 scenario is characterized by high economic growth, low population growth, and continuing globalizationssimilar to A1B. However, two major differences from A1B are a change in economic structure toward a service and information economy and an emphasis on global solutions to social and environmental problems with the introduction of clean and resource-efficient technologies. Finally, the B2 scenario describes a future world in which the emphasis is on local solutions to economic, social, and environmental problems. Population growth and economic growth are moderate. There is less rapid and more diverse technological change than in A1B and B1. The rates of adoption of FGD technology by on coalfired power plants follow IMAGE assumptions by region and scenario, as shown in Table 2. The difficult parameterization to make is the rate of Hg removal that will be achieved by such systems by. For our base-case assumptions, we have followed the review by Brown et al. (31). We assume that in the A1B and A2 scenarios, no significant advance is made over reported levels in the U.S. of 40% (30). For the B1 and B2 scenarios, we assume that Hg reduction efficiency improves to 70% in the developed world (North America, OECD Europe, and Japan) and to 51.6%sthe more optimistic view of current reduction measurements at a number of U.S. power plants (31)sin the rest of the world. Because of the uncertainty in this parameter, we also examine a range of alternative values of Hg removal, including the introduction of advanced sorbent technologies. For the eight industrial process categories, we use the IMAGE industrial process SO 2 forecasts as a guide to project Hg emission growth rates ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 8, 2009

3 TABLE 3. Mercury Emissions by Sector and World Region in 1996 and 2006 (Mg/yr) emitting sector North America Central and South America Africa Europe, Russia, Middle East Asia and Oceania world 1996 residential industry power transportation biomass burning industrial processes total residential industry power transportation biomass burning industrial processes total other estimates 1995 (2)a 2000 (4)a most recent (24)a b c d e a Estimates do not include biomass burning. UNEP best estimate for Hg emissions from global biomass burning is 675 Mg/yr (24). b South Africa only. c Europe and Russia only. d China, India, and Australia only. e Includes only major world regions and countries; therefore, not globally complete. FIGURE 1. Distributions of global Hg emissions among major contributing source types in 1996 and Results and Discussion The results of the 1996 and 2006 model calculations are shown in Table 3 and Figure 1. We estimate that global mercury emissions in 1996 were 2128 Mg, coming mainly from biomass burning (614 Mg, 28% of total), power-plant fuel TABLE 4. Mercury Emissions in by Scenario and World Region (Mg/yr) Central Europe, North and South Russia, Asia and scenario America America Africa Middle East Oceania world A1B A2 B1 B combustion (321 Mg, 15%), artisanal gold extraction (300 Mg, 14%), industrial fuel combustion (266 Mg, 13%), zinc smelting (191 Mg, 9%), and residential fuel combustion (106 Mg, 5%). By 2006, global emissions had risen to 2480 Mg, an increase of 17%. The shares of contributing sources had also changed, reflecting a large increase in power sector and industrial emissions. Biomass burning had fallen to 586 Mg (25%), power-plant fuel combustion had risen sharply to 459 Mg (18%), and zinc smelting emissions had risen dramatically to 384 Mg (15%). Without good information on the time variation of emissions associated with artisanal gold extraction, we have assumed they remained constant at 300 Mg (12%). Emissions from industrial fuel combustion rose to 278 Mg (11%). Emissions from residential fuel combustion declined, because there was less use of coal for domestic heating and cooking in China and other developing countries. These results reveal the growing roles of coal combustion for VOL. 43, NO. 8, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

4 FIGURE 2. Distributions of global Hg emissions among major contributing source types in the A1B and B1 scenarios. FIGURE 3. Comparison of global Hg emissions for each of the six cases, by major contributing sector, showing sensitivity of the A1B and B1 cases to alternative values of the Hg removal efficiency of FGD in the ranges of % and %, respectively. electricity generation and of nonferrous metals production in East Asia (China). In other world regions growth was more moderate in this period (e.g., South Asia), or environmental regulations and technology renewal held Hg emissions in check (e.g., OECD Europe, U.S., and Japan). By global energy use will increase dramatically over present-day levels. Primary energy use increases by a factor of 3.6 under the A1B scenario, 2.6 under A2, 2.3 under B2, and 2.1 under B1. This does not mean that Hg emissions will increase by these same amounts, however; the penetration of improved technologies and fuels, sectoral transformations, and socioeconomic development all combine to moderate the growth of Hg emissions. In addition, the adoption of new emission control technologies for other speciessparticularly FGD for SO2 controlshas a mitigating effect on Hg emissions as well. Table 4 summarizes Hg emissions under each scenario. We estimate that global Hg emissions will grow from 2480 Mg in 2006 to 4856 Mg in under the A1B scenario, to 3906 Mg under A2, and to 2630 Mg under B2. Under the B1 scenario, Hg emissions decline slightly to 2386 Mg. The change relative to 2006 levels ranges, therefore, from +96% to -4%. Developments in Asia largely dictate the variation across scenarios. Rapid expansion of coal-fired electricity ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 8, 2009 generation in Asia is the major determinant of the growth, as shown in Figure 2, where the share of coal-fired power plants in global Hg emissions could rise to almost 50%. The increase in Hg emissions from coal combustion is difficult to mitigate. Though rapid deployment of FGD is partially successful in the environmentally strong scenarios like B1 and B2, the limited ability of FGD to remove Hg inhibits the effectiveness of emission control. Figure 3 shows the full range of emissions, present and future, by major emitting category. It also shows the sensitivity of total Hg emissions to the efficiency of Hg removal by FGD in the range of % removal, for the highest (A1B) and lowest (B1) scenarios. The range of total emissions covered by this sensitivity analysis is Mg, or a range from a reduction in total Hg emissions of 36% to an increase of 106%. This sensitivity analysis shows how important it will be to develop ways to improve the Hg removal efficiency of coal-fired power plants in future developing-country applications. And though the effect on absolute emission levels is greatest in the developing world, Figure 4 shows that, relative to total emission levels in each region, the greatest benefits to improving Hg removal efficiency in coal-fired power plants are obtained in the U.S. and OECD Europe where electricity generation comprises the largest share of

5 FIGURE 4. Sensitivity of total Hg emissions under the A1B scenario in eight regions and the world to Hg removal efficiency of FGD in the range %, indexed to the base A1B assumption of 40% removal. Note that the lines for the U.S. and OECD Europe are essentially superimposed. total emissions. In these two regions, improving Hg removal efficiency to the extreme value of 100% reduces total Hg emission levels in by almost 70%. At present, a number of add-on sorbent injection technologies are under development with the specific objective of capturing Hg released during coal combustion. One promising technique is Activated Carbon Injection (ACI), in which powdered activated carbon is injected into the fluegas stream before the particulate matter control device in order to bind the Hg and allow it to be captured (32). Demonstrations of ACI suggest that a performance of 95% Hg removal may be achievable, though actual removal efficiencies in some of the tests were significantly less than 95% (33). If we assume that all coal-fired power plants employ ACI at 95% removal in, then the range of Hg emissions over the four scenarios would be Mg, or a change relative to 2006 levels of -33% to +40%. However, because ACI has not yet reached the stage of commercial deployment, it would be inappropriate to include it in the main forecasts. Finally, we examine the effect of these future scenarios on relative emissions of the different species of Hg, as this will have an important bearing on the trends for long-range transport vs local Hg deposition. Applying the source-specific Hg speciation profiles listed in Table 6 of Streets et al. (7) to each of the 152 source categories, we obtain speciated emissions of Hg0, divalent Hg (Hg2+), and particulate Hg (Hgp) for each emissions case. The results are illustrated in Figure 5. For 1996 the shares of primary emitted species at global scale were: 67% Hg0, 25% Hg2+, and 7% Hgp. We see that by 2006 the share of elemental Hg has already begun to decline slightly. By we see a large decline in the share of elemental Hg and an increase in the share of divalent Hg, though there is some variation across scenarios. Shares of Hgp remain relatively constant at 5-7%. Under A1B we find the following shares: 47% Hg0 and 49% Hg2+. Under B1, the shares are 56% Hg0 and 40% Hg2+. This shift from Hg0 to Hg2+ is largely driven by the adoption of pollution control technology on coal-fired installations. The implications are clear: there will be a relative trend toward reduced long-range transport and enhanced local deposition of Hg. In absolute terms, the change in emissions from 2006 to for Hg0 is from -19% to +38% (also shown in Figure 5), whereas for Hg2+ the range is from +38% to +243%. The forecasting of future emissions is a challenging task. There are significant uncertainties in each of the steps involved and some of them are hard to quantify. In this work we have utilized the IPCC socioeconomic scenarios and their FIGURE 5. Variation of species fractions in each of the six cases (left axis, lines) and absolute values of Hg0 emissions (right axis, bars). associated energy and fuels requirements, imposed on them the technology characteristics that affect emissions of Hg, and used our own parametrizations of Hg emission rates. The uncertainty in estimates of present-day global Hg emissions is believed to be on the order of ( 25-30% (4, 5). Consistent with this, we have estimated the uncertainty of estimates of Hg emissions in China at ( 44% (7) with uncertainties as high as a factor of 3-4 for some industrial processes. The uncertainty in estimates of Hg emissions is perhaps less than might be expected. This is because the quantity of Hg (just like sulfur) in the original fuels and ores constrains the final emission estimate. Uncertainties are higher for species that are mostly formed in the combustion zone and therefore sensitive to combustion conditions (e.g., CO, NO x, BC). Future developments in complex human systems are inherently unpredictable (27). Two significant sources of uncertainty in this regard are the evolution of the underlying driving forces and the nature of mitigation policies that governments might adopt in response to environmental degradation. Thus, the scenario approach is the commonly used way to bound the range of possible futures. However, we can be sure that the uncertainty in the estimates of future emissions along any of the trajectories is greater than the ( 25-30% estimate for current emissions. In this work we have chosen four IPCC scenarios that effectively bound future conditions. The A1B scenario, reflecting as it does unfettered economic growth with no special attention to environmental protection, represents an upper bound on emissions, while the B1 scenario, with its emphasis on global solutions to environmental problems, represents a lower bound. Using these brackets, we can say that Hg emissions will likely be higher in than in The worst case would be roughly a doubling in emissions from 2480 Mg to 4860 Mg. The best case would be a reduction of just 4% under base-case conditions. This could be improved to 33% by full application of ACI or to 36% by complete elimination of Hg emissions from coal-fired power plants. Coal-fired power plants, we feel, are the key target for Hg emission control. With a great expansion of electricity demand throughout the developing world, much of which is likely to be fueled by coal (according to the IPCC scenarios), this source category dominates Hg emissions in. Though flue-gas desulfurization for SO 2 control will achieve some reduction in Hg emissions as a cobenefit, it is insufficient. We recommend the urgent development of specific techniques for Hg control, such as ACI, before these new power plants are built. Acknowledgments This work was funded by the U.S. Environmental Protection Agency s STAR Program on the Consequences of Global Change for Air Quality, as part of collaboration with Harvard VOL. 43, NO. 8, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

6 University. We are grateful for the support and collegial advice of Darrell Winner (U.S. EPA) and Daniel Jacob (Harvard University). The submitted manuscript has been created by UChicago Argonne, LLC, Operator of Argonne National Laboratory ( Argonne ). Argonne, a U.S. Department of Energy Office of Science laboratory, is operated under Contract DE-AC02-06CH Literature Cited (1) Pacyna, J. M.; Pacyna, E. G. Global Emissions of Mercury to the Atmosphere. Emissions from Anthropogenic Sources; Arctic Monitoring and Assessment Programme (AMAP): Oslo, Norway, (2) Pacyna, E. G.; Pacyna, J. M. Global Emissions of Mercury from Anthropogenic Sources in Water Air Soil Pollut. 2002, 137, (3) Pacyna, J. M.; Pacyna, E. G.; Steenhuisen, F.; Wilson, S. Mapping 1995 Global Anthropogenic Emissions of Mercury. Atmos. Environ. 2003, 37 (S1), (4) Pacyna, E. G.; Pacyna, J. M.; Steenhuisen, F.; Wilson, S. Global Anthropogenic Mercury Emission Inventory for Atmos. Environ. 2006, 40, (5) Wilson, S. J.; Steenhuisen, F.; Pacyna, J. M.; Pacyna, E. G. Mapping the Spatial Distribution of Global Anthropogenic Mercury Atmospheric Emission Inventories. Atmos. Environ. 2006, 40, (6) Nelson, P. F. Atmospheric Emissions of Mercury from Australian Point Sources. Atmos. Environ. 2007, 41, (7) Streets, D. G.; Hao, J.; Wu, Y.; Jiang, J.; Chan, M.; Tian, H.; Feng, X. Anthropogenic Mercury Emissions in China. Atmos. Environ. 2005, 39, (8) Wu, Y.; Wang, S.; Streets, D. G.; Hao, J.; Chan, M.; Jiang, J. Trends in Anthropogenic Mercury Emissions in China from 1995 to Environ. Sci. Technol. 2006, 40, (9) Pacyna, E. G.; Pacyna, J. M.; Pirrone, N. European Emissions of Atmospheric Mercury from Anthropogenic Sources in Atmos. Environ. 2001, 35, (10) Pirrone, N.; Costa, P.; Pacyna, J. M.; Ferrara, R. Mercury Emissions to the Atmosphere from Natural and Anthropogenic Sources in the Mediterranean Region. Atmos. Environ. 2001, 35, (11) Pai, P.; Heisler, S.; Joshi, A. An Emissions Inventory for Regional Atmospheric Modeling of Mercury. Water Air Soil Pollut. 1998, 101, (12) Gbor, P. K.; Wen, D.; Meng, F.; Yang, F.; Sloan, J. J. Modeling of Mercury Emission, Transport and Deposition in North America. Atmos. Environ. 2007, 41, (13) Dabrowski, J. M.; Ashton, P. J.; Murray, K.; Leaner, J. J.; Mason, R. P. Anthropogenic Mercury Emissions in South Africa: Coal Combustion in Power Plants. Atmos. Environ. 2008, 42, (14) Telmer, K.; Veiga, M. World Emissions of Mercury from Small Scale Artisanal Gold Mining and the Knowledge Gaps about Them. In Mercury Fate and Transport in the Global Atmosphere: Measurements, Models and Policy Implications; Pirrone, N., Mason, R., Eds.; United Nations Environment Programme: Geneva, Switzerland, (15) Wilhelm, S. M. Estimate of Mercury Emissions to the Atmosphere from Petroleum. Environ. Sci. Technol. 2001, 35, (16) Friedli, H. R.; Radke, L. F.; Lu, J. Y. Mercury in Smoke from Biomass Fires. Geophys. Res. Lett. 2001, 28, (17) Friedli, H. R.; Radke, L. F.; Lu, J. Y.; Banic, C. M.; Leaitch, W. R.; MacPherson, J. I. Mercury Emissions from Burning of Biomass from Temperate North American Forests: Laboratory and Airborne Measurements. Atmos. Environ. 2003, 37, (18) Friedli, H. R.; Radke, L. F.; Prescott, R.; Hobbs, P. V.; Sinha, P. Mercury Emissions from the August 2001 Wildfires in Washington State and an Agricultural Waste Fire in Oregon and Atmospheric Mercury Budget Estimates. Global Biogeochem. Cycles 2003, 17, doi: /2002gb (19) Cinnirella, S.; Pirrone, N. Spatial and Temporal Distribution of Mercury Emissions from Forest Fires in Mediterranean Region and Russian Federation. Atmos. Environ. 2006, 40, (20) Selin, N. E.; Jacob, D. J.; Park, R. J.; Yantosca, R. M.; Strode, S.; Jaeglé, L.; Jaffe, D. Chemical Cycling and Deposition of Atmospheric Mercury: Global Constraints from Observations. J. Geophys. Res. 2007, 112, D02308, doi: /2006jd (21) Selin, N. E.; Jacob, D. J.; Yantosca, R. M.; Strode, S.; Jaeglé, L.; Sutherland, E. M. Global 3-D Land-Ocean-Atmosphere Model for Mercury: Present-Day Versus Preindustrial Cycles and Anthropogenic Enrichment Factors for Deposition. Global Biogeochem. Cycles 2008, 22, GB2011, doi / 2007GB (22) Jaffe, D.; Prestbo, E.; Swartzendruber, P.; Weiss-Penzias, P.; Kato, S.; Takami, A.; Hatakeyama, S.; Kajii, Y. Export of Atmospheric Mercury from Asia. Atmos. Environ. 2005, 39, (23) Weiss-Penzias, P.; Jaffe, D.; Swartzendruber, P.; Hafner, W.; Chand, D.; Prestbo, E. Quantifying Asian and Biomass Burning Sources of Mercury Using the Hg/CO Ratio in Pollution Plumes Observed at the Mount Bachelor Observatory. Atmos. Environ. 2007, 41, (24) United Nations Environment Programme. Mercury Fate and Transport in the Global Atmosphere: Measurements, Models and Policy Implications; Pirrone, N., Mason, R., Eds.; UNEP: Geneva, Switzerland, (25) Streets, D. G.; Bond, T. C.; Lee, T.; Jang, C. On the Future of Carbonaceous Aerosols. J. Geophys. Res. 2004, 109, D24212, doi: /2004jd (26) RIVM. The IMAGE 2.2 Implementation of the SRES Scenarios: A Comprehensive Analysis of Emissions, Climate Change and Impacts in the 21st Century; RIVM CD-ROM publication ; National Institute for Public Health and the Environment (RIVM): Bilthoven, The Netherlands, (27) Nakicenovic, N., et al. Emissions Scenarios: A Special Report of Working Group III of the Intergovernmental Panel on Climate Change; Cambridge University Press: Cambridge, UK, (28) Intergovernmental Panel on Climate Change (IPCC). Climate Change 2001: The Scientific Basis. Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change; Houghton, J. T., et al., Eds.; Cambridge University Press: UK and New York, (29) U.S. Environmental Protection Agency. Compilation of Air Pollutant Emission Factors, AP-42, Fifth Edition, Vol. I: Stationary Point and Area Sources; EPA: Washington, DC, (30) U.S. Environmental Protection Agency. Mercury Study Report to Congress, Vol. II: An Inventory of Anthropogenic Emissions in the United States; EPA-452/R ; EPA: Washington, DC, (31) Brown, T. D.; Smith, D. N.; Hargis, R. A.; O Dowd, W. J. Mercury Measurement and Its Control: What We Know, Have Learned, and Need to Further Investigate. J. Air Waste Manage. Assoc. 1999, 1 97; 29th Annual Critical Review. (32) U.S. Environmental Protection Agency. Controlling Power Plant Emissions: Mercury-Specific Activated Carbon Injection (ACI); available at merc_specific.htm. (33) U.S. Environmental Protection Agency. Control of Mercury Emissions from Coal-Fired Electric Utility Boilers; available at ES802474J ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 8, 2009