Updated 8/13/2012 10 a.m.

Last summer, visitors to Lacamas Lake, a reservoir outside the town of Vancouver, Wash., may have seen some strange devices floating on the reservoir's surface.

"It look(ed) like an alien lander," says Washington State University doctoral student Bridgit Deemer.

The mysterious objects were traps designed to catch air bubbles spurting up from the reservoir's surface. Deemer and an undergraduate student, Maria T. Glavin, were trying to learn more about what was in these bubbles. They suspected they held a whole lot of methane, a greenhouse gas many times more potent than carbon dioxide.

While dams are often seen as a clean energy generators, the reservoirs behind them can also be a significant source of climate changing gases. To understand how this works, you first need to understand a process called stratification. As Deemer explains: "During summer, reservoirs (and lakes) heat up on the surface, but below this hot layer they also have a cooler lower layer of water."

Way down deep in that cold lower layer, at the bottom of the reservoir, there's almost no oxygen. Microbes that live in that anoxic environment produce methane, which stays trapped at the bottom of the reservoir all summer. In a natural lake, as the warm top layer cools in autumn, it slowly begins to mix with the cold lower layer.

As this happens, the methane from the bottom gradually rises to the surface, which allows it to mix with oxygen. As the methane oxidizes, it becomes available to other microbes, who use it up, and in turn, excrete carbon dioxide.

When a reservoir is drawn down, though, there is a sudden change in water pressure. Deemer and Glavin hypothesize that the pressure change allows methane bubbles to form and quickly rise the the surface of the water. When this happens, the methane rises too quickly to be available to the microbes that would normally use it up. Instead, it bubbles out.

That's where the traps come in. They're essentially an upside-down funnel submerged just underneath the water, says Deemer. "The fat end catches the bubble and funnels it into this long, narrow tube (that rises above the water surface), where a pressure sensor can record exactly when that bubble formed." She goes on: "The bubbles accumulate in this really narrow tube and you have to go out there and pull them out with a syringe." Deemer and Glavin would suck out the air bubbles and drive them back to their laboratory, where they measured the concentration of methane in the bubbles. When I spoke with her Friday, Deemer didn't recall the exact numbers, but said they were very high.

At one point during the study, says Deemer, methane bubbles were effervescing from the reservoir so quickly that their traps filled up overnight, and she and Glavin had to empty them daily. This prolific bubbling coincided with the late summer drawdown of Lacamas Lake, when managers were spilling lots of water from the dam.

This finding, that the reservoir emitted high amounts of methane in early September, when it was being rapidly emptied, points to a potential need for changes in reservoir management to limit greenhouse gas emissions.

"What we observed were these huge events where bubbles of methane formed and escaped the lake very quickly," says Deemer. Other researchers have suggested reservoirs are a globally significant source of greenhouse gases; Deemer's results suggest this may be the case.

Managers might be able to mitigate this emission source, though, by drawing down reservoirs later in the year, after the top and bottom water layers have already mixed.

"There are a ton of things involved in how a reservoir is managed, but the hope is that this could be a part of climate change mitigation plans for some managers of dams," says Deemer.

Deemer and Galvin presented their results at the Ecological Society of America meeting in two posters, but they are not yet published. The researcher was careful to point out she does not expect all reservoirs to behave the same. Some, probably those that receive more nutrient runoff from cities or farms, will likely be lower in oxygen and thus have more methane to release. Reservoirs that behave more like rivers than lakes will not stratify, and thus won't release methane this way. And newer reservoirs, which may have more available nutrients, could behave differently than older ones.

Deemer plans to expand her methane studies to other reservoirs this year, and is also looking into releases of nitrous oxide, another greenhouse gas.

For now, though, she says:

"I think it would be okay to just focus on the bubbles"

Stephanie Paige Ogburn is the online editor at High Country News.

Images of air bubbles (although not from a reservoir) courtesy Flickr user Phil Whitehouse.

Image of Lacamas Lake dam spillway courtesy Flickr user Jonathan James.

This article was updated at 10 a.m., August 13, 2012, to clarify the manner in which reservoir drawdown leads to methane bubbles being released t0 the atmosphere instead of used up by microbes. An earlier version indicated drawdown allowed the layers of cold and warm water to mix more quickly. This is not the case.