"Treatment dollars could be saved in [total maximum daily loads (TMDLs)] if it were more clearly understood how much of the mercury put out by a [WWTP] actually becomes methylmercury," said Robin Landeck Miller, associate and natural waters fate and transport operations leader at HydroQual (Mahwah, N.J.). "The question isn’t easy to address, because how much mercury becomes methylated is controlled by site-specific environmental conditions."
Trace amounts of natural mercury exist in rocks and soils, but the major sources of air-derived mercury are anthropogenic, according to Mark Brigham, a U.S. Geological Survey (USGS) hydrologist with the agency’s National Water Quality Assessment (NAWQA) program.
The 2009 NAWQA report Data on Mercury in Water, Bed Sediment, and Fish From Streams Across the United States was generated from both fish-tissue sampling conducted from 1998 to 2005 and a national survey of mercury in 300 U.S. streams, which includes data on fish tissue and stream-bed sediment.
Of the 290 sites sampled in the NAWQA study, 60 to 70 were known to be mining-influenced — regions of historic mercury and gold mining, which uses mercury in its process. Other mercury sources include waste incineration and combustion of coal for power and other industries. In sample sites where industrial and mining sources of mercury are lacking, "the atmosphere is the predominant source" of mercury, Brigham noted.
Runoff a Key Factor
David Krabbenhoft, a USGS research scientist and co-primary investigator of the Mercury Experiment To Assess Atmospheric Loading in Canada and the United States (METAALICUS), has spent more than 9 years studying how mercury behaves in a watershed.
Krabbenhoft said the driving force behind this major USGS mercury study is to answer a question from the 1996 International Mercury Meeting in Hamburg, Germany: "What good is it going to do to control emissions now?"
Research conducted in the experimental lakes area north of Ontario at Lake 658 examined land surface runoff. Krabbenhoft noted that two-thirds of the mercury that enters this region is from anthropogenic activities. USGS ran various mercury-loading tests in the area to quantify response changes of freshwater systems.
"No one’s ever done this before," Krabbenhoft said.
After a 2-year comprehensive literature review, Krabbenhoft’s team began dosing the entire Lake 658 watershed with specific mercury isotopes in the wetland, forest, and lake to measure what was already in the environment and "what’s continuing to come in," he said. The team added mercury by a factor of four at the same rate for 7 years.
Methylated mercury quickly made its way into the food web, according to Krabbenhoft. However, during the first few years of study, other USGS observations seemed to be at odds. Namely, researchers found that 70% of the mercury entering Lake 658 was transported via runoff, but the mercury added to the land didn’t immediately load into the water.
"The source of the conflict is twofold," Krabbenhoft said. USGS learned that a portion of precipitative mercury that falls on land is re-emitted into the atmosphere while some gets hung up in vegetation. Through the course of the study, it took 3 to 4 years to see the team’s land-applied isotope show up in Lake 658.
While mercury continued to build in the food web, after 4 years, the rate accumulated in fish tissues and in Lake 658’s environment looked as if might plateau — and by year 7 of testing, it had.
The second part of the study, currently under way, focuses on how freshwater systems, including the food web, respond when mercury is reduced. For the last 2 years, the research team has been monitoring mercury reduction in Lake 658 in the absence of mercury dosing.
Without the addition of mercury, "systems are responding quickly to sources being shut off," said Krabbenhoft, who expects to have more data in a few years.
Maximum Standard for Mercury
While USGS does not provide more than consumer guidance from these studies — the agency advises following state rules about freshwater fish consumption — the U.S. Environmental Protection Agency (EPA) is looking carefully at this work. According to Ellen Kurlansky, of the EPA Office of Air and Radiation, when many are asking, "‘Why spend money to abate mercury source emissions?’ the METAALICUS project has shown that it will do good."
EPA is developing a maximum standard for mercury under the Clean Air Act. The agency was poised to undertake this effort in 2000, but the previous administration instead developed the Clean Air Interstate Rule and the Clean Air Mercury Rule. The latter was vacated by the U.S. Court of Appeals for the D.C. Circuit, and the former was remanded to EPA.
In early 2010, EPA’s "Information Collection Request for National Emissions Standards for Hazardous Air Pollutants for Coal-Fired Electric Utility Steam-Generated Units" will begin requiring approximately 800 coal-fired plants to test their stacks and provide the agency with data.
In addition, EPA will regulate coal-fired power plant emissions for the "maximum reduction achievable." New plants will be required to meet the performance of the best-performing unit, and older plants will have to meet the performance of the average best-performing 12%, Kurlansky said.
EPA also is working on maximum mercury standards for portland cement, chlor-alkali plants (there are four left that still use mercury cells), industrial boilers, and iron and steel foundries.
Kurlansky noted that while the process of setting a maximum standard is divorced from other agency research, the USGS data will be useful in EPA’s cost and benefit analysis. METAALICUS shows that "if you change the deposition, the fish adjust rather rapidly," Kurlansky said. "That’s very important in a cost–benefit world."
Nitrates a Remedy?
A twist in the mercury saga is that in a recent first ecosystem-level study, nitrates in fresh water seem to limit methylmercury production.
Charles Driscoll, a professor of environmental engineering at Syracuse University (Syracuse, N.Y.), has been studying biogeochemical control of nitrate on methylmercury production at Lake Onondaga, which had been home to a chlor-alkali facility that used mercury cells to process chlorine from 1880 to 1980.
The lake is also naturally enriched in sulfate, and with low dissolved oxygen, conditions were ideal for the study. "It’s a methylmercury soup," Driscoll said.
However, the nearby Metropolitan Syracuse WWTP came under court order to improve treatment to control ammonium — 90% of the lake’s quantity was derived from the WWTP, said Driscoll. The plant began nitrifying its wastewater, which nearly tripled the amount of nitrates in the lake.
Driscoll’s research included analyzing temporally and vertically detailed water-column profiles and developing a long-term data set for methylmercury and reduction–oxidation species.
According to a 2009 article co-authored by Driscoll (Svetoslava G. Todorova, Charles T. Driscoll Jr., David A. Matthews, Steven W. Effler, Mark E. Hines, and Elizabeth A. Henry, "Evidence for Regulation of Monomethyl Mercury by Nitrate in a Seasonally Stratified, Eutrophic Lake," Environmental Science & Technology [September], pp. 6572–6578), "[a] recent study indicated that emission controls on sulfur dioxide have decreased [methylmercury] production and mercury accumulation in fish. Our results suggest changes in [nitrate] leaching could also influence [methylmercury] production and bioaccumulation of [methylmercury] in aquatic biota."
Driscoll said the nitrates "out-compete" the sulfur-reducing bacteria, so the change at the plant "basically shut down" sulfur-reducing bacteria and methylmercury in the lake.
Opportunities for Balance
For some heavily affected freshwater resources, Driscoll’s discoveries could offer a measure of methylmercury control.
Honeywell International (Morris Township, N.J.), which is remediating a former AlliedSignal Inc. property on the shores of Lake Onondaga, is pilot-testing adding nitrate as a treatment option to reduce sulfate-reducing bacteria in order to control mercury. (Honeywell and AlliedSignal merged in 1999.) Driscoll explained that Honeywell initially was going to pipe in oxygen, which is very costly. "Monitoring the nitrate gives complete flexibility," he said.
Driscoll noted that nitrate additions are used in Europe to control phosphorus, and the research could be relevant in such environments as the Everglades. But he said the "best way" to address mercury-contaminated lakes is to limit carbon, mercury, or sulfate.
Miller noted that "a lot of unanswered questions" remain regarding the mercury methylation process. While wastewater plants that nitrify wastewater may reduce mercury methylation, those that nitrify and denitrify could be worsening the mercury problem. "Trying to balance removing nutrients versus exacerbating methylation really comes down to trying to manage bacterial populations," she said.
Miller, who assists government agencies with developing TMDL models and is preparing to work on an upcoming mercury TMDL for the New York–New Jersey Harbor, said that while there are numerical models that can calculate system response and "you could certainly measure mercury," loadings and concentrations must be better understood.
"It’s not clear yet how these simultaneous things are going to be considered," Miller said. "We have to err on the side of caution."
— Andrea Fox, WE&T
©2009 Water Environment Federation. All rights reserved.