Do trees share blame for global warming?

Cows burp it, pipelines and landfills leak it, and vast amounts lie frozen beneath the ocean floor. Methane is ubiquitous - as fuel for heating and cooking and as a source of concern for atmospheric scientists. Molecule for molecule, methane packs thousands of times more punch as a "greenhouse gas" than carbon dioxide does.

Until now, scientists tracking debits and credits in the globe's methane "budget" figured they had a pretty good handle on where the gas comes from - mostly from microbes breaking down organic material in places where oxygen is relatively scarce.

Enter Frank Keppler. Working with colleagues from Northern Ireland and the Netherlands, Dr. Keppler has discovered that plants may give off significant amounts of methane just by growing. And the amount they give off appears to rise with temperature. The results have stunned many researchers because no one expected methane to form biologically out in the open air, where oxygen abounds.

It's not that there's more methane in the atmosphere, but that some of it is coming from a wholly unexpected source. The results imply that, at best, this new source of methane may need to be taken into account as nations try to curb carbon-dioxide emissions by planting trees. Would increased methane emissions erase the gains against CO2? At worst, the results imply that thawing tundra in the Arctic is not the only worrisome source of methane in a warming world.

The experiments Keppler and his colleagues performed grew out of the team's effort to measure the gases that plants give off only in tiny amounts. When they looked at emissions from dead leaves, "we saw a pattern of methane" along with other gases, says Keppler, a scientist at the Max Planck Institute for Nuclear Physics in Heidelberg, Germany. Others had detected methane from rice plants, but thought the rice merely acted as a minipipeline for methane formed in the muck in which rice grows.

Keppler put fresh and dried plant materials, as well as young plants, in special chambers. He removed possible sources of contamination - including microbes - and found that from bananas and sugar cane to European ash and Spanish moss, the material yielded methane.

From individual plants, the amounts are small: from 12 to 370 billionths of a gram. (One gram is about .04 ounces - the weight of two small paper clips.) But the collective effect could be large. The team roughly calculates that, globally, living plants may contribute from 10 to 30 percent of global methane emissions.

The phenomenon appears to be connected somehow to the presence of pectin in plants. For humans, pectin is used to set jellies and jams. For plants, it serves as a kind of glue for cementing cells together.

The results, published in the Jan. 12 issue of the journal Nature, have drawn astonished reactions and skepticism from researchers, particularly regarding the extrapolations of global emissions.

"This needs to be confirmed," says Michael Keller, a scientist with the US Forest Service's International Institute of Tropical Forestry in Rio Piedras, Puerto Rico, and a visiting scientist at the University of New Hampshire. "Until we have the basic mechanisms" for creating methane in plants - "or at least we understand the controls," such as nutrients, heat, or moisture - "it's hard to do reliable extrapolations."

Still, he says, the results may help explain the high methane emissions he and others have found over tropical forests using ground and satellite measurements. Other scientists recently have reported increased emissions over Arctic-river flood plains in eastern Siberia, invoking a variation of the "rice pipeline" hypothesis to explain them.

Given all the scrutiny plants have undergone, one of the open questions is how researchers could have missed these emissions. Keppler speculates that because the methane emissions are so small, they wouldn't have been detected in field studies. Any signal would have been swamped by much larger natural background levels. And microbial sources have been so well established that no one has looked for another mechanism.

For some researchers, the evidence Keppler and his team presents is sufficiently convincing to begin working them into computer models of the globe's greenhouse-gas budget - especially the potential implications for land-use changes. To do that, scientists will need to see how emissions might vary with plant species, says Alex Guenther, a senior scientist at the National Center for Atmospheric Research in Boulder, Colo.