UCAR > Communications > Staff Notes Monthly > July 2001 Search

July 2001

Wildfires and mercury pollution: A smoking gun?

Mercury emissions from this boreal forest fire near Hearst, Ontario, Canada, last July were measured with instruments aboard a Twin Otter aircraft. (Photo courtesy Ian MacPherson, National Research Council of Canada).

Hans Friedli. (Photo by Carlye Calvin.)

What do forest fires have to do with mercury in fish? Hans Friedli and Larry Radke (senior research associates in ASP and ATD, respectively) are trying to find out.

In the 1950s, photographs from Japan showed the neurological damage inflicted on children whose mothers had eaten fish contaminated by mercury from an industrial spill into Minimata Bay. In the United States, concerns about elevated mercury levels over the last several decades have led the Environmental Protection Agency (EPA) and the Food and Drug Administration (FDA) to issue warnings to vulnerable populations to limit consumption of fish, especially from the Great Lakes. This past December, the EPA recommended that mercury emissions from coal-fired power plants be curbed, and the Bush administration has included mercury on the list of pollutants it is considering for increased regulation. Most industrial sources of mercury have been phased out in the United States, but a few continue, including electrical switches and thermometers.

But, as Hans and Larry have learned, the classic model of point- source pollution affecting people downstream or downwind may be only part of the story.

Gaseous elemental mercury in the atmosphere travels the globe for about a year before being deposited on land or water. About 6,500 tons (5,900 metric tons), all well mixed, are circulating at any one time. "So a concentration of mercury downwind of a coal-fired power plant may not be much different from that upwind," says Larry. About half the atmospheric mercury got there from natural sources (in soil, oceans, and volcanoes) and the other half through human activity. The EPA estimates that 41 tons (37 metric tons) is contributed annually from U.S. coal-fired plants. Mercury is removed from the atmosphere through oxidation, which creates ionic mercury in the free troposphere and in clouds; it then rains or falls out as wet or dry deposition to the surface.

Mercury in fish

The livers of mammals, including humans, can process moderate amounts of mercury, but it becomes toxic at high doses. Victorian hat makers, like Lewis Carroll's fictional Mad Hatter, suffered from mercury poisoning, and Hans points out that indigenous gold miners in Amazonia are exposed to toxic levels of mercury today.

Mercury becomes most dangerous when ionic mercury is deposited onto soils or water bodies and becomes methylated by microbial or abiotic processes. Methylmercury, the neurotoxic form, enters the food chain and becomes concentrated at higher and higher levels as larger fish and mammals eat smaller fish and biota.

The highest levels of gaseous ionic mercury ever observed were reported this spring from NOAA's Point Barrow Observatory in Alaska; indigenous Arctic populations consuming the region's game and fish are thought to be at severe risk. Long-range atmospheric transport brings the mercury to the Arctic, where chemical reactions tied to springtime ozone depletion lead to elevated levels of ionic species and deposition to the surface.

Whether in the Arctic or elsewhere, it's the reactive ionic species that, through precipitation or dry deposition, fall on land, water, and vegetation. For trees, "Wet deposition is most important," says Hans. "[Mercury is] picked up by the surfaces—the leaves or needles—and it stays there."

Mercury in smoke

Hans and Larry became interested in smoke from biomass burning as a potential source of atmospheric mercury when their wildfire research called their attention to the high levels of mercury threatening Arctic Inuit communities. (For more on wildfire research at NCAR, see the August 2000 issue of Staff Notes Monthly; URL is below.) The elevated levels of sulfur dioxide coming off forest fires in the Los Angeles basin suggested that other pollutants absorbed by trees might also be liberated in a wildfire's intense heat.

To conduct laboratory tests, they enlisted financial support from EPRI and cooperation from the Meteorological Service of Canada, which provided a state-of-the-art Tekran mercury vapor analyzer and the methodology to measure mercury contained in smoke particles. Their work with Julia Lu (MSC) will be reported in a forthcoming paper in Geophysical Research Letters.

Since most of what burns in a wildfire is foliage and ground litter, not tree trunks, volunteers collected samples from seven forests across the continental United States. Those samples were set alight at the U.S. Forest Service Fire Science Laboratory's burn facility in Missoula, Montana. All the coniferous and deciduous samples contained mercury at levels ranging from 14 to 71 nanograms per gram of fuel. Ground litter had the highest concentrations, reflecting accumulation during annual or biannual cycles before the leaves or needles were shed.

When the team ignited the samples, their sensors immediately detected mercury, and plenty of it. All samples released nearly all the mercury they had stored—from 94 to 99 percent. Most was released as gaseous elemental mercury, although up to 10% was attached to aerosol particles.

The authors extrapolated their findings to global biomass burning and estimated the contribution at up to 800 tons (730 metric tons) per year, or 25% of all anthropogenic sources.

A complicated budget

Mercury from naturally occurring sources in geothermal mineralized zones is another factor contributing to the atmospheric budget. According to Hans, researchers are now finding that in zones with high concentrations of mercury, such as those in the U.S. Southwest, "the fluxes are much higher than expected—about three times higher."

Last summer, Hans and Larry were back in the air, flying over a wildfire in Quebec. "The emissions were higher than in the lab experiment," Hans says, "presumably because mercury in real fires is also emitted from heated soil, a source not yet considered in our experiments."

Hans also flew with the Tekran analyzer aboard the NSF/NCAR C-130 during ACE-Asia flights this spring (see the Spring 2001 issue of UCAR Quarterly). Based on source data published by Chinese researchers, the team expected to find mercury distributed at "some level above background," perhaps around 0.6 nanograms per cubic meter increase above a global background estimated at 1.4–1.6 ng/m3. Instead, "there was tremendous variety in the observations." For some plumes, "we went up to 5 nanograms." For Hans, "this was the biggest surprise" of the project. By working with the ACE-Asia meteorologists, modelers, and scientists making other tracer measurements, Hans and Larry identified plume sources, including the Shanghai industrial area. Over Japan "we saw mercury emissions from areas with high concentrations of industry and also from the Miyake Jima volcano."

This summer the team plans to measure mercury emissions from a large wildfire in cooperation with Peter Hobbs (University of Washington) and also will participate in a prescribed burn in Saskatchewan's Prince Albert National Park. They'll be gathering fuel and ash and will use the Thermacam deployed in earlier wildfire experiments to track fire intensity and behavior.

All this suggests that the mercury budget is far from solved, and it won't be a simple matter to keep the toxin out of the world's fisheries. But, Hans muses, perhaps this work is helping put the global picture in better perspective.

• Zhenya Gallon

On the Web:
"Burning clues: Wildfire research team lights up Washington," Staff Notes Monthly, August 2000
"ACE-Asia finds plenty to study," UCAR Quarterly, Spring 2001
FDA Consumer Advisory on Mercury in Fish
EPA National Advice on Mercury in Fish Caught by Family and Friends

In this issue... Other issues of Staff Notes Monthly

UCAR > Communications > Staff Notes Monthly > July 2001 Search

Edited by Bob Henson, bhenson@ucar.edu
Prepared for the Web by Jacque Marshall
Last revised: Tue Jun 26 15:04:26 MDT 2001