Greenwire
2.15.06
MERCURY: EPA study links fallout in Ohio to nearby coal-burning plants
Darren Samuelsohn
Nearly 70 percent of the mercury in rain collected at an Ohio River Valley monitoring site originated from nearby coal-burning industrial plants, according to new U.S. EPA-funded research.
The findings, which researchers hope to publish this spring, apparently contradict the Bush administration's position that links most U.S. mercury fallout to power plants in other countries.
The study -- conducted in 2003 and 2004 in Steubenville, Ohio -- is the first in which scientists used rain samples and meteorological data to track mercury from smokestacks to monitors, said Matthew Landis, an EPA researcher and the lead investigator in the mercury study.
EPA issued regulations for controlling power plants' mercury emissions in March 2005 that were based on computer models that estimated mercury deposition nationwide. The agency's most widely trumpeted modeling result was this: Only 8 percent of the wet mercury deposition nationwide comes from U.S. power plants.
Assuming that the bulk of U.S. wet deposition of the toxic metal comes from smokestacks abroad, the administration adopted a market-friendly, cap-and-trade program to facilitate domestic mercury reductions. The alternative was to force most of the nation's 1,300 coal- and oil-fired power plants to install their own pollution controls.
Critics of the administration's mercury rule say Landis' findings show for the first time that deposition from regional and local smokestacks are much higher than EPA models suggest. To protect public health, they say, power plants should be required to install mercury-specific emission control equipment rather than be allowed to buy pollution credits from cleaner-burning facilities elsewhere.
"This is a very important study that makes the case for hot spots caused by local large sources, mostly coal plants," said Praveen Amir, a senior scientist for the Northeast States Coordinated Air Use Management, a nine-state coalition of air pollution officials.
An EPA source who has not been given official clearance to speak to the press on the record said, "What we've said to the public is the 8 percent number. We've basically hidden the large local deposition from sources as a way to justify the trading program."
Significance of Steubenville
Landis and his former graduate school professor, University of Michigan atmospheric scientist Gerald Keeler, launched the Steubenville research in the wake of EPA's 1997 report to Congress that listed U.S. power plant emissions as the largest single source of mercury pollution. The report also highlighted how mercury is a significant threat to unborn children whose mothers eat fish with tissue containing large concentrations of the toxin.
The study's goal was to determine whether emissions from regional coal combustion could be more than just a secondary source of mercury.
Much of the continent's mercury data was collected by a network of 69 North American mercury deposition monitors. The monitors -- run primarily by independent scientists and academics -- are designed to measure background levels of mercury and not specifically placed near industrial sources. They also do not account for trace elements, Landis said in a recent interview.
Landis and Keeler picked Steubenville -- famous for being the birthplace of singer and actor Dean Martin -- for their work because it is within air transport distance of many major Midwestern coal-burning facilities. Within 125 miles are more than a dozen major coal combustion sources, including Ohio Edison and American Electric Power Co. power plants and petroleum coke manufacturers.
Steubenville also has a deep history in international air pollution and health research. Harvard University has tested Steubenville's residents dating back to the early 1970s in a well-known and frequently cited study on the link between air pollution and human health. Many of the major regulatory decisions on air pollution -- including EPA's current federal soot and ozone limits -- have been based on information gathered in Steubenville.
The mercury scientists set up their equipment on the mountain campus of Franciscan University, which is 1,500 feet above the Ohio River and industrial riverbank of Weirton, W.Va. A large white cross overlooks the measurement site and an adjacent student parking lot.
Landis and Keeler collected more than 160 precipitation samples there over two years using an automated device about the size of an office photocopier. The machine's trap door opens at the first trace of rain or snow.
The scientists compared their samples with meteorological data about wind patterns and the origin of the precipitation. And they traced mercury to coal combustion sources based on trace elements found in the samples, including sulfur, nitrates and selenium. Those three elements originate only at a coal burning, as opposed to nickel, chlorine or other elements that would originate at an oil-fired power plants.
After running the data through EPA models, Landis found as much as 67 percent of the mercury in Steubenville originated from coal-combustion sources. From weather data, he determined the mercury had traveled three days at most, or about 400 miles.
Also noteworthy, Landis said his results appear to contradict the findings of the Electric Power Research Institute. The group -- which receives funding from industry, the Energy Department and other sources -- has suggested that ionic mercury may be transformed chemically into a form of elemental mercury that wafts into the atmosphere and circles the globe.
If industry's reported findings were accepted, sources said, U.S. regulators would have less incentive to push for domestic mercury reductions. But if the mercury tends to stay in its reactive form, as Landis' research indicates, then EPA would have a greater stake in addressing local emissions.
The Landis-Keeler study is ongoing. Similar rain collection devices and weather tracking programs are in place in northern Vermont, at five sites in Michigan and near Tampa, Fla. Landis and Keeler also plan this year to track dry mercury deposition, a more complex subject given the difficulty of drawing samples from air as opposed to taking them from water.
When finished at the end of the year, the research will have cost more than $1.1 million. EPA has funded the study through appropriations specifically marked for work related to power plant emissions and the administration's "Clear Skies" legislation.
The findings have a margin of error of 14 percent, and they also do not specify which power plant or other coal-combustion source the mercury originated from.
EPA 'quite confident' in mercury rules
Top EPA officials have been aware of the Landis-Keeler study for at least a year -- to about the date the administration was required under a legal deadline to issue the final mercury rule, March 15, 2005.
Tim Oppelt was serving as EPA's Office of Research and Development's acting director when he was briefed by Landis on April 27, 2005 -- about six weeks after the agency finalized the rule. Then-EPA air chief Jeff Holmstead had a similar briefing last July. In all, Landis said he attended at least a half dozen sessions in Washington to present his data to senior EPA officials. He also provided more details of his study at an October air pollution event in Beijing.
Sources familiar with Landis' efforts said EPA senior officials ordered an external peer review before he could submit the study to a scientific journal. The step was necessary given the high-stakes findings and the political repercussions that could follow, sources said.
The peer review -- which cost about $20,000 to the Research Triangle Institute in North Carolina -- ended last December without any major changes recommended in the study's findings. A new manuscript was submitted for publication to a journal last month. Landis declined to release a draft of the article until it has been peer reviewed by the publication.
Senior Bush administration EPA officials say they are not surprised by Landis' findings. Jason Burnett, EPA's point person on mercury issues, said the agency has known all along that the industrial Midwest stands out as a region where mercury emissions would be driven up by regional sources.
EPA models have showed deposition levels in Steubenville of 45 percent, with levels east of the city of more than 70 percent, he said. And even with such levels taken into account, Burnett said the administration's regulations will address the pollution as best as possible.
"The evidence is there will continue to be a large amount of mercury, but the power plant part of it will be reduced significantly because of these rules," he said.
Burnett's explanation squares with how the Bush administration justified its move to the market-based approach nearly a year ago. At press briefings, EPA last March went on the public relations offensive to argue that regulators are limited in their ability to reduce mercury in the United States because of global emission patterns and also the production patterns of the seafood industry.
EPA also said specific mercury control technologies were not available yet for deployment across the country.
Holmstead told reporters then that EPA did not anticipate the trading program would leave local areas vulnerable to high deposition levels. "We don't think there will be any hot spots, we're quite confident of that," he said. "A cap-and-trade approach can always get a bigger reduction at a lower cost."
Landis' presentation to Oppelt clearly indicated that he wanted the information presented to agency decisionmakers to ensure they knew what was being studied. But he also acknowledged last spring that he could not deliver the work before EPA's final rule was due because he was still waiting for final analysis.
"We think its important for policymakers to have the data as soon as we can generate it," he said of the delays in releasing the information.
Study entered into public record
EPA's acknowledgment that deposition levels are likely to be higher in the Midwest has not satisfied its critics. They contend that EPA has said little publicly about the Landis-Keeler study until it was prompted by environmentalists and the press. They say the agency's silence raises questions about whether the administration would prefer to keep the information out of the public eye until after the rulemaking docket is closed and all subsequent reviews were complete.
"It's one thing to say we're withholding the results because it's going through a peer review process, it's another thing to acknowledge that the research is happening," said Ann Weeks, an attorney with the Clean Air Task Force.
Environmentalists first heard about the Steubenville study last summer.
They commented on it in December and entered Landis' Power Point presentation to Oppelt into the public record. Weeks said the report puts EPA on the spot to issue an official reaction to the information.
After interviewing Keeler, the Michigan Department of Environmental Protection also cited the Steubenville study in its own comments submitted to EPA.
EPA has not released a copy of the Holmstead briefing materials.
Greenwire has filed a Freedom of Information Act request to obtain it.
The lack of documentation has also frustrated critics from industry. They have been forced to hold their fire in challenging the study. Leonard Levin, mercury expert at the Electric Power Research Institute, said in an interview that he was skeptical of the Landis-Keeler findings.
Echoing the reaction of EPA officials, Levin said the study is eliciting shrugs given power plant emissions have long been the source of pollution within the river valley. He also said he doubted the validity of the mercury findings given the difficulty in tracking emissions from any one power plant to a deposition site. He also said any debate on the study is premature until it is officially published.
"Give us something technical to chew on," Levin said. "We don't have anything."
EPA says the Landis-Keeler scientific findings were not made public because of the need for external peer review -- a process that slowed the research's submission to a journal for publication by about six months.
Burnett also said EPA would be open to considering the study's results as it sorts through ways that states will implement the mercury trading program.
Administration critics are pushing to make Landis' Steubenville research another stepping stone in their effort to influence policy.
If EPA does not address Landis' findings by a mid-May deadline for administration rehearing, the mercury rule's critics say litigation challenging the rule will almost certainly include a mention of the study. Oral arguments in the case are expected later this year in a federal appeals court in Washington. Plaintiffs challenging EPA's rule include 15 states, environmentalists and several Maine-based Native American tribes.
This is not the first time that questions have been raised about EPA's mercury rule and the analysis behind it. Critics say the administration has appeared to ignore or downplay other key studies on mercury, including the toxic pollutant's potential cardiovascular health effects. The agency's inspector general issued a report in February 2005 that said the administration made decisions during the rulewriting process that were based on preconceived ideas in favor to the market-based trading system.
Greenwire and the Washington Post have reported EPA issued rule language copied verbatim from industry-sponsored memos. In one instance, suggestions to adopt the cap-and-trade approach had come from a Washington law office where Holmstead and his top legal aide, William Wehrum, had worked before joining the administration. Holmstead left EPA last summer, and the air office is currently under the acting command of Wehrum.
Last month, a spokeswoman for the EPA inspector general confirmed that another report is under way on the administration's mercury rule. Landis was among the EPA officials who has been interviewed for the investigation.
