EPA Docket Center, Room B-108
U.S. EPA West
United States Environmental Protection Agency
1301 Constitution Avenue N.W.
Washington DC 20460
Attn: Docket ID No. OAR-2003-0056
Proposed National Emission Standards for Hazardous Air Pollutants; and, in the Alternative, Proposed Standards of Performance for New and Existing Stationary Sources: Electric Utility Steam Generating Units
Submission filed by Environment Canada on March 30, 2004
Contents
Environment Canada is pleased to submit comments to the United States Environmental Protection Agency (EPA) in response to the Federal Register notice of January 30, 2004.
Environment Canada commends EPA action to reduce mercury emissions from coal fired power plants and nickel emissions from oil fired boilers. Our comments are confined to mercury only.
Mercury is a toxic, persistent, bioaccumulative substance. Human exposure to mercury — primarily by eating contaminated fish — may cause neurological and developmental damage, particularly in fetal and adult nervous systems. Thousands of Canadian lakes currently have fish consumption advisories, primarily due to mercury. This is a particular concern for subsistence fishers and Aboriginal people in the far North who consume large quantities of fish as part of their traditional lifestyles. In some northern communities, mercury levels in blood are greater than the Health Canada acceptable level.
Environment Canada supports the efforts being made by the EPA to limit emissions of mercury. Our modelling indicates that 10% of the mercury that is deposited in Canada each year comes from U.S. sources, with that figure climbing to 38% in the Great Lakes Region, home to more than nine million Canadians. As a result, mercury reductions in the U.S. are needed to help protect the environment and human health in Canada.
At the same time, Environment Canada is concerned that both the Maximum Achievable Control Technology (MACT) proposal and Phase I of the cap-and-trade proposal fall short of the emissions reductions that are achievable with current and emerging control technologies.
The Canadian view is based on:
- a review of current literature, which suggests that capture of mercury from bituminous and sub-bituminous coal-fired power plants (and possibly lignite-fired plants) on the order of 60-90% is achievable;
- consensus agreement among all Canadian provinces, territories and the federal government to set, by 2005, a Canada-wide Standard for mercury emissions from coal-fired power plants that will target "the national capture of mercury from coal burned in the range of 60-90%," by 2010; and
- a recommended mercury standard for Alberta power plants based on activated carbon injection and fabric filters, by 2009.
Specifically, Environment Canada recommends and urges the U.S. EPA to:
- consider a more stringent MACT standard for mercury that would result in the national capture of mercury from coal burned in the range of 60-90%, or
- set a lower Phase I cap (for 2010) in the cap-and-trade option that would result in the national capture of mercury from coal burned in the range of 60-90%.
This would be comparable to the mercury Canada-wide Standard for coal-fired power plants that is being developed by the Canadian Council of Ministers of the Environment.
Top of Page
Environment Canada is pleased to respond to the request of the United States Environmental Protection Agency (EPA), in the Federal Register notice of January 30, 2004, for comments on the proposals to regulate mercury emissions from power plants under the U.S. Clean Air Act.
Environment Canada strongly welcomes the EPA's efforts to reduce mercury emissions from the coal-fired electric power generation sector. Canada and its provincial and territorial governments have already determined that a mercury standard for this sector is needed in Canada as well.
This document describes Environment Canada's concerns regarding mercury, and Canadian actions to reduce mercury releases. It also includes a high level review of mercury control technology, and outlines why Environment Canada believes that the EPA could go further to set a more aggressive mercury standard. Finally, the document describes the impact of the mercury proposal on Canada, and shows that a more aggressive standard would reduce transboundary flows and thus be more protective of the Canadian and American environment.
Top of Page
Health and environment
Mercury is a toxic, persistent, bioaccumulative substance. Biological processes, such as bacterial activity in plants and sediments on lake bottoms, in rivers and oceans, can transform elemental mercury to methyl mercury, one of the most bioaccumulative and toxic forms. Levels of methyl mercury in animals increase as it moves up the food chain, from plankton to large fish, birds and mammals, including humans.
Human exposure to mercury — primarily by eating contaminated fish — may cause neurological and developmental damage, particularly to fetal and adult nervous systems. Methyl mercury ingested by a pregnant woman, or metallic mercury that may enter her body from breathing contaminated air, can be passed from the maternal blood to the developing fetus by crossing the placental barrier. Methyl mercury can also accumulate in an unborn baby's blood to a concentration higher than the concentration in the mother, with possible greater adverse health effects to the developing systems in the fetus.
Although levels of total mercury have begun to decline in Aboriginal blood and hair in Canada's North, there is still significant variability between different communities, depending on a number of factors. These include: location, lifestyle, proportion of traditional foods in the diet, and genetic factors. Of the Native populations in Canada's North, Caucasians, Dene, Métis and others are registering total mercury blood levels below the Health Canada "acceptable" level of <20 µg/kg bw/day. However, the Inuit population in Baffin Island and in other regions of the North display significantly higher levels of total blood mercury than their northern counterparts. Total blood mercury levels recorded in Nunavik (northern Quebec), for example, were observed at up to 16% greater than the Health Canada "acceptable" level.
In contrast to the Health Canada total blood mercury guidelines, a recent re-evaluation of mercury in the U.S. concluded that a benchmark dose of 5.8 µg/L maternal blood (10X safety factor) is considered adequate to protect the population. When using this value rather than the Health Canada level of <20 µg/kg bw/day, 79% of the highest exposed population in Nunavik exceeded the U.S. standard, which is considerably higher than the value of 16% identified using the Health Canada standard. As well, proportionally higher levels were seen in Inuit populations in Baffin, Inuvik, Kitikmeot and Kivalliq than calculated under the Health Canada guideline.
Mercury's effect on wildlife is also harmful. Fish-eating predators such as loons, merganser ducks, osprey, eagles, herons and kingfishers can have very high concentrations of mercury. Mercury has been detected in Common Loons from Alaska to Atlantic Canada, and blood concentrations have been correlated with levels in prey fish species. A survey of mercury in loons from five regions across the U.S. and Canada has shown that blood mercury concentrations increased from west to east, with the highest levels in southeast Canada. High levels of mercury are suspected of impairing the loon's reproductive success as well as causing growth-related problems. These problems inevitably lead to an increased death rate and a decreased birth rate, resulting in a reduction in the abundance of natural populations.
In addition, mercury has been found in predatory mammals such as otters from south central Ontario. It is thought that elevated mercury levels in otters may cause early mortality due to toxicity and behavioural changes. While the reproduction and behaviour of bird species is generally affected by exposure to methyl mercury, mammals most often suffer neurological effects. The severity of the toxic effects will depend on the degree of exposure, and may range from a slight impairment to reproductive failure or death.
In Canada, federal, provincial and territorial agencies have issued fish consumption advisories limiting consumption of specific fish species from certain lakes. Thousands of advisories on fish species from individual water bodies are put in place every year due to mercury contamination of fish. Province wide advisories are in effect in New Brunswick and Nova Scotia, while other provinces have advisories for specific lakes and/or species. Over 90% of fish consumption advisories are due to mercury, which is a particular concern for subsistence fishers who consume large quantities of fish as part of their traditional lifestyles. This is particularly true in the far North, where mercury cools, condenses and settles in the colder Arctic ecosystem, resulting in its biocentration in organisms higher in the food chain, including humans.
Mercury from foreign sources
Unlike some pollutants, mercury emissions may remain in the atmosphere anywhere from hours to years. As a result, a cloud of mercury, known as the global pool, has been building up over the years. Furthermore, mercury that was deposited many years ago may volatilize and re-enter the atmosphere. These emissions can end up in ecosystems far from their point of origin, such as in Canada's Arctic, where there are no local point sources.
Global atmospheric emissions of mercury from major anthropogenic sources are estimated to be over 2 200 metric tonnes per year. In 1999, Canada's share was about 8 tonnes and the U.S. share was 106 metric tonnes (or 117.3 U.S. short tons). The largest source of atmospheric mercury in both Canada and the U.S. is coal-fired power plants, with 1999 emissions of approximately 2 metric tonnes and 43.5 metric tonnes (or 48 U.S. short tons), respectively.
Canada is a net receiver of mercury, with some 10% of the 138 tonnes we receive annually coming from the U.S., based on the 1995 global emissions inventory. In the Great Lakes region, that figure climbs to 38% of total mercury deposition coming from U.S. sources.
Since most of the mercury that is deposited in Canada each year comes from foreign and natural sources, Canada cannot solve its mercury problem on its own. In addition to reducing its domestic sources of mercury, Canada needs other countries, including the U.S., to do the same.
Top of Page
Federal government
Canada has taken action for over thirty years to reduce mercury releases to the environment. In 1972, the federal government issued a regulation under the Fisheries Act to limit the discharge of mercury in effluents from mercury-cell chlor-alkali plants. In 1978, the federal government passed another regulation, under the Canadian Environmental Protection Act, to limit mercury releases to the air from mercury-cell chlor-alkali plants. These regulations have since been updated. As a result of these regulations, coupled with provincial regulations and a shift away from chlorine bleaching at pulp and paper mills, mercury releases from this sector alone have fallen from 70 tonnes in the 1970s to 47 kilograms in 2001.
The federal government also participates in a number of activities with other countries, including the U.S., to reduce mercury:
- the Great Lakes Bi-national Toxics Strategy;
- the North American Regional Action Plan on Mercury, under the Commission for Environmental Cooperation;
- the Arctic Council Action Plan mercury project;
- the United Nations Economic Commission for Europe Heavy Metals Protocol;
- the United Nations Environment Programme Global Mercury Programme;
- the New England Governors/Eastern Canadian Premiers Mercury Action Plan.
Canadian Council of Ministers of the Environment
In 1998, the Canadian Council of Ministers of the Environment (CCME) — which comprises the 14 Environment Ministers of the federal, provincial and territorial governments — agreed to set mercury Canada-wide Standards for the remaining large uncontrolled sources. As a result, between 2000 and 2002, the CCME endorsed a series of mercury Canada-wide Standards for base metal smelters, waste incinerators, mercury-containing lamps, and dental amalgam waste. These Canada-wide Standards are to be met by 2010 or earlier, depending on the standard.
The CCME has also committed to develop a Canada-wide Standard by 2005, to reduce mercury emissions from the coal-fired electric power generation sector by 2010 and "to explore the national capture of mercury from coal burned in the range of 60-90%." The CCME has agreed that "provincial application of the national target or standard may vary, with some provinces doing more and some less, depending on the control technologies for the different coal types (especially lignite). Alternatively, the standard may vary depending on the coal type or other relevant factors." The CCME has also agreed that "the capture of mercury includes all efforts to reduce mercury releases and may be achieved through a continuum of actions, from pollution prevention through emissions controls."
The federal, provincial and territorial governments are working together at the CCME toward setting this Canada-wide Standard. The focus at present is to analyze mercury data from Canadian coal-fired power plants (e.g., the mercury content in coal and ash, and mercury stack emissions, and the resulting mass balance) and review current and emerging control technology.
Alberta
In June 2002, the Alberta Minister of Environment asked the Clean Air Strategic Alliance (CASA) to develop an approach for managing emissions from the electricity sector in Alberta. CASA is a multi stakeholder, not-for-profit organization. The members include representatives from industry, health and environmental groups, and the provincial and federal governments.
In December 2003, and by consensus, CASA tabled its report with the Alberta Minister of Environment. On March 4, 2004, the Government of Alberta announced that it was accepting and adopting the CASA recommendations.
CASA based its recommended emission standards for sulphur dioxide, nitrogen oxides and particulate matter on Best Available Control Technology Economically Achievable (BATEA) levels, but did not set a specific capture limit for mercury, because there was no established BATEA level for mercury at the time. Instead, the key CASA recommendations for mercury are:
- In a multi-stakeholder setting, review the latest BATEA in 2005, and set a standard based on BATEA that existing coal-fired units must meet by the end of 2009. The regulations or standards will be established under Alberta's Environmental Protection and Enhancement Act.
- If a BATEA level is not identified in 2005, then, by 2009, existing coal units must make a financial level of effort equivalent to installing fabric filters and activated carbon at an injection rate to be determined as part of the 2005 review.
- New or transitional units that have fabric filters would only be expected to meet the activated carbon component of this level of effort commitment by 2009.
- For those units designated to shut down in the 2012-2017 time frames, no mercury controls are required.
As a result of these recommendations, CASA predicts that emissions from coal fired power plants in Alberta would drop by 400 kg — a 50% reduction — by 2009.
Top of Page
The Government of Canada has been following the work in the U.S. to determine methods to capture mercury from coal-fired power plants using various control technologies. Canada appreciates the significant resources put forth by the U.S. Government and industry in looking at the mercury control issue. We offer the following comments on the work that has been conducted to date.
Current technology
The EPA's Information Collection Request (for mercury emissions data from coal-fired power plants) has revealed that some of the controls used to address acid rain and smog also capture mercury in varying degrees. Generally, mercury removal rates are higher in units fired with bituminous coal, followed by units fired with sub-bituminous, and lower for units fired with lignite coal. In the U.S., more than 90% of the coal combusted for power generation is bituminous and sub-bituminous.
The following list provides examples of pollution controls that also capture mercury:
- Coal washing: Coal washing is widely used to remove sulphur and ash from coal before it is burned. Mercury removals from near zero to about 60% were reported for the washing methods that are used in the U.S.
- Electrostatic precipitators (ESPs): ESPs are designed to capture particulate matter, a component of smog. Mercury removals across cold-side ESPs averaged 27%, compared to 4% for hot-side ESPs.
- Fabric filters: Fabric filters are also designed to capture particulate matter. Removals for fabric filters were higher than ESPs, averaging 58%.
- Flue gas desulfurization (FGD): FGDs are often known as scrubbers, and are designed to control sulphur dioxide, a gas that causes acid rain. Both wet and dry flue gas FGD systems removed 80-90% of the gaseous mercury, but elemental mercury was not affected.
- Fluidized-bed combustors with fabric filters: Designed to control sulphur, these systems showed high levels of mercury removals, averaging 86%.
- Selective catalytic reduction (SCR) and selective non-catalytic reduction (SNCR): SCRs and SNCRs are used to control nitrogen oxides, a gas that may form ozone. Test results on mercury control via SCRs and SNCRs were inconclusive; additional full-scale tests are in progress.
Equipment vendors have also recommended that a mercury MACT rule be set at 90% mercury removal for bituminous coal-fired units and 70% removal for sub-bituminous-fired units.
Emerging technology
The U.S. Department of Energy (DOE) continues to fund emerging mercury control technology for coal-fired power plants. In 2001, DOE's Office of Fossil Energy issued a press release saying: "The department wants to develop a wider array of mercury control options for power plants that can reliably reduce emissions by 50 to 70 percent by 2005 and 90 percent by 2010." To that end, a number of technologies have been tested.
The following list is incomplete, but serves to illustrate some of the promising new controls for mercury:
- Sorbent injection: Sorbent injection is the mercury control technology closest to commercialization for power plants. It involves directly injecting activated carbon, or some other sorbent, into the flue gas stream of a utility boiler. The mercury in the flue gas is adsorbed onto the sorbent, and the sorbent is then collected in downstream particulate control equipment. Mercury removal depends on many factors; preliminary results show removal levels greater than 90% for boilers burning bituminous coals.
- Carbon filter beds: Carbon filter beds are capable of removing high mercury concentrations from waste incinerators. The Eastman Chemical Company's coal gasification plant in Kingsport, Tennessee, achieves 90%+ mercury removal in a carbon absorbent bed. However, carbon filter beds have not been tested for mercury removal at pulverized coal-fired power plants.
- In Situ: In situ sorbent removal captures mercury in the flue gas by using thermally activated sorbent produced in situ. The sorbent consists of semi-combusted coal, which is extracted from the furnace, injected into the flue gas downstream of the air pre-heater, and then captured in a particulate control device. Pilot-scale tests indicate that mercury removal efficiencies of up to 90% are achievable.
While acknowledging the variability in the data from the Information Collection Request, Environment Canada observes that current and emerging pollution controls are able to capture mercury in ranges that are considerably higher than those proposed by the EPA. Although the cost of emerging, commercial-scale mercury control technologies remains a question, the U.S. experience has shown that actual costs of meeting air pollution regulations are often much lower than anticipated.
The above findings suggest that the EPA could set a more aggressive mercury standard by:
- setting a MACT standard based on current as well as emerging mercury control technology; or
- establishing a Phase I cap (for 2010) that is lower than the proposed cap, based on the co-benefits of reducing sulphur dioxide and nitrogen oxides emissions as required to meet the proposed Interstate Air Quality Rule. The Phase I cap should be set at a level that could be met with current as well as emerging mercury control technology.
Top of Page
Environment Canada was not able to model the impact on Canada of the proposed mercury rule within the 60-day comment period. However, Environment Canada has been able to calculate a very approximate estimation of the positive impact — i.e., a reduction in mercury deposition in Canada — of the MACT option and the cap-and-trade option. It must be stressed that these estimations are very rough, as they:
- are based on Canada's modelling work with the 1995 global emissions inventory;
- assume that the spatial distribution of reduced coal-fired power plant emissions is the same as total U.S. emissions; and
- assume a linear relationship between emissions reductions in the U.S. and its impact on the deposition in Canada and the Great Lakes Region.
Since Environment Canada believes that the U.S. mercury rule could be more stringent, Environment Canada also did an approximate estimation of the benefits of a 50% emissions reduction from U.S. coal-fired power plants by 2010, and compared that to the impact of the current proposals.
Regional impact
The impact of U.S. emissions over Canada has a significant spatial distribution; it is higher over eastern Canada and decreases toward northern Canada, where the impact of European and Asian sources is large. Due to this regional distribution, Environment Canada did a very approximate estimation of the impact of the EPA's proposed mercury rules on the Great Lakes Region, where roughly 38% of annual mercury deposition is from U.S. sources. More than nine million Canadians live in the Great Lakes Region.
Both the MACT option and Phase I of the cap-and-trade option would reduce total mercury deposition by roughly 4% in the Great Lakes Region. That figure climbs to roughly 7% with a more aggressive rule that cuts emissions in half (see Table 1). By 2018, Phase II of the cap-and-trade option would reduce total mercury deposition by about 10% in the Great Lakes Region.
Table 1: Estimated reduction in mercury deposition in the Great Lakes Region
| |
MACT option |
Phase I of cap-and-trade option |
Hypothetical option |
Phase II of cap-and-trade option |
| Resulting mercury emissions |
34 tons* by 2007/08/09 |
34 tons by 2010 |
24 tons by 2010 |
15 tons by 2018 |
| Decrease in mercury emissions (from 48 tons in 1999) |
14 tons |
14 tons |
24 tons |
33 tons |
| Sector % emission reduction (from 48 tons in 1999) |
30 % |
30 % |
50 % |
70 % |
| National % emission reduction (from 117.3 tons in 1999) |
12 % |
12 % |
20 % |
28 % |
| % reduction in total mercury deposition in the Great Lakes Region |
4 % |
4 % |
7 % |
10 % |
Top of Page
Environment Canada commends the U.S. for proposing mercury regulations for coal-fired power plants.
The CCME is currently developing a Canada-wide Standard for mercury emissions from Canadian coal-fired power plants, and is exploring the national capture of mercury from coal burned in the range of 60-90%.
Based on the CCME goal, and Canada's determination to reduce mercury deposition as a result of foreign emissions, Environment Canada urges the U.S. EPA to:
- consider a more stringent MACT standard for mercury that would result in the national capture of mercury from coal burned in the range of 60-90%, or
- set a lower Phase I cap (for 2010) in the cap-and-trade option that would result in the national capture of mercury from coal burned in the range of 60-90%,
which would be consistent with the mercury Canada-wide Standard for coal-fired power plants that is being developed by the CCME.
All information requests should be directed to:
Barry Stemshorn
Assistant Deputy Minister
Environmental Protection Service
Environment Canada
351 St. Joseph Boulevard
Gatineau QC K1A 0H3
Canada
Top of Page
Environmental effects
Environment Canada - Atlantic Region. 1998. Mercury in Atlantic Canada: A Progress Report. Sackville.
Environment Canada. 2003. Mercury, Fishing for Answers. Hull.
Wren, C.D., P.M. Stokes, and K.L. Fischer. 1986. Mercury levels in Ontario mink and otter relative to food levels and environmental acidification. Can. J. Zool. 64: 2854-2859.
Health effects
Health Canada Mercury Issues Task Group. 2003. Mercury: Your Health and the Environment: A Resource Tool. Ottawa.
Indian and Northern Affairs Canada. 2003. Northern Contaminants Program Canadian Arctic Contaminants Assessment Report II — Human Health, pp. 21-24, 64-66, 103. Ottawa.
Emissions, deposition and modelling
Dastoor, Ashu P., and Yvan Larocque. 2004. Global circulation of atmospheric mercury: a modelling study. Atmospheric Environment 38: 147-161.
Environment Canada. 1999, 2000, 2001. National Pollutant Release Inventory.
Pacyna, E.G., and J.M. Pacyna. 2002. Global emissions of mercury from anthropogenic sources in 1995. WASP 137 (1): 149-165.
United Nations Environment Programme, Chemicals. 2002. Global Mercury Assessment. Geneva.
What Canada is doing
Canadian Council of Ministers of the Environment. 2003. Notice on a Canada wide Standard for Mercury Emissions from Coal-fired Power Plants.
Clean Air Strategic Alliance. 2003. An Emissions Management Framework for the Alberta Electricity Sector: Report to Stakeholders. Edmonton.
Control technology
Environmental Working Group. 1999. Mercury Falling, An Analysis of Mercury Pollution from Coal-Burning Power Plants. Washington, D.C.
Feeley, Thomas J., James T. Murphy, Jeffrey W. Hoffman, Evan J. Granite, Scott A. Renninger. 2003. "DOE/NETL's Mercury Control Technology Research Program for Coal-Fired Power Plants," EM (October 2003), pp. 16-23.
Massachusetts Department of Environmental Protection. 2002. Evaluation of the Technological and Economic Feasibility of Controlling and Eliminating Mercury Emissions from the Combustion of Solid Fossil Fuel. Boston.
Pavish, John H.a, Everett A. Sondreala, Michael D. Mannb, Edwin S. Olsona, Kevin C. Galbreatha, Dennis L. Laudala, Steven A. Bensona. 2003. Status review of mercury control options for coal-fired power plants. Fuel Process. Technol. 82 (2-3): 89-165. (aEnergy & Environmental Research Center, University of North Dakota, 15 North 23rd Street, P.O. Box 9018, Grand Forks ND 58202-9018. bUniversity of North Dakota, Chemical Engineering Department, P.O. Box 7101, Grand Forks ND 58202-7101.)
United States Department of Energy, Office of Fossil Energy. 2001. "Meeting Mercury Standards: DOE Selects 6 Projects to Develop Cost-Saving Technologies for Curbing Mercury Emissions from Coal Power Plants." Techline (News), June 2001.
United States Environmental Protection Agency. 1999. Progress Report on the EPA Acid Rain Program. Washington, D.C.
Working Group for the Utility MACT Formed Under the Clean Air Act Advisory Committee, Subcommittee for Permits/New Source Review/Toxics. 2002. Recommendations for the Utility Air Toxics MACT: Final Working Group Report. Washington, D.C.
Also see:
| 
