Two days before Halloween, Kurt Waltzer appeared before the House subcommittees on Energy and the Environment and said something bizarre. Waltzer, managing director of the Clean Air Task Force, a nonprofit focused on combating climate change, told the congressmen that oil extraction, if done with a certain technique, could help the United States reduce carbon dioxide emissions – even if the nation keeps burning coal.

More surprisingly, a number of environmental groups share Waltzer's enthusiasm for "enhanced oil recovery," or EOR. The process works by pumping carbon dioxide into aging oil wells to help retrieve stranded pockets of crude. The greenhouse gas stays in the ground. Oil companies have practiced EOR for decades in places like the Permian Basin, on the Texas-New Mexico border. But only recently have environmental groups like the Natural Resources Defense Council and the Center for Climate and Energy Solutions taken an interest in it. "It's a win-win-win," says Patrick Falwell, a fellow at the center. "You're expanding domestic energy production, advancing carbon capture and sequestration, contributing to economic growth, and also reducing carbon dioxide emissions." (Actually, that's four wins.)

But serious concerns remain about the environmental costs of burning all that crude – and about whether the CO2 will stay put. As climate activist Bill McKibben put it: "Anything that helps recover more oil is a bad idea."

Power plants are the United States' most prolific producers of carbon dioxide, contributing 40 percent of our emissions. The Obama administration wants to reduce that number to fight climate change. In September, the U.S. Environmental Protection Agency released draft rules that would cap CO2 emissions from new power plants. If those rules become law, new coal plants could be built only if they capture and safely store carbon emissions.

Unfortunately, so-called carbon capture and sequestration technology is currently prohibitively expensive. The EPA says it probably won't require existing power plants to install it, and most proposed plants that plan to use it are subsidized by the Department of Energy. Still, there are successful examples: The Great Plains Synfuels Plant in North Dakota captures about half of its carbon emissions, and ExxonMobil's LaBarge, Wyo., natural gas processing plant traps 4 million metric tons of the stuff each year. Both plants sell the carbon for enhanced oil recovery.

There is a huge demand for CO2 in the oil industry. Historically, oil companies in the Permian Basin have tapped natural underground deposits of CO2 in Colorado and New Mexico and piped the pressurized gas hundreds of miles to the oilfields. There, it's injected underground, where it pushes the stranded oil towards producing wells. But for several years, CO2 demand has outpaced supply. This shortage is considered the single largest obstacle to expanding enhanced oil recovery in the basin.

Dave Hawkins, the Natural Resources Defense Council's director of climate programs, would like to use carbon captured from power plants to meet that demand, instead of "doing that rather idiotic thing of pulling CO2 out of the ground and pipelining it and sticking it back in." The Department of Energy estimates that EOR could permanently store the CO2 emissions from 93 large coal-fired power plants operated for 30 years.

Before depleted oilfields can be safely transformed into CO2 sinks, however, Hawkins and others say the EPA must increase its oversight to ensure the pressurized CO2 doesn't cause earthquakes or leak out, which is likely what happened in a Wyoming oilfield last fall, when six dead ducks were found floating in a creek bubbling with CO2 . "If you want to claim the carbon benefit of sequestration," says NRDC scientist George Peridas, "then I think we should set the bar higher."

The issue of "carbon benefit" is where EOR gets controversial. To be a meaningful storage solution, says ClimateProgress editor Joe Romm, EOR would need to be used to bury massive amounts of CO2, which would also require extracting "a staggering amount of oil." Indeed, a 2009 EOR study estimates that every metric ton of CO2 injected underground produces oil that, when burned, emits between 3.7 and 4.7 metric tons of CO2. Other estimates are lower, but the amount emitted still exceeds the amount stored.

"Does it produce more oil out of the ground?" Waltzer asks. "Yeah, it does." Still, if EOR helps bring about the widespread adoption of carbon capture technology, ultimately resulting in a net decrease of emissions, he's OK with burning more oil in the short term. "You're paying a little bit now to get a big payoff later."

Besides, EOR is the only real market for CO2 , and the only cost-effective way to sequester it, says energy consultant Vello Kuuskraa. Although storing carbon without producing oil is better from a climate perspective, Kuuskraa says, it's unrealistic without a hefty carbon tax. "There's no economic value. It's waste disposal." Plus, "You still have to import that barrel of oil from the Middle East."

The possibility of substituting domestic for imported oil, and extending the life of existing oilfields rather than drilling new ones, is another reason some environmentalists support EOR. "It is much better for the environment to have that oil come from oilfields that are already developed," says NRDC's Hawkins. "EOR doesn't mean additional barrels of oil. It just helps those barrels of oil come from (better) places," with the added benefit that CO2 is sequestered in the process.

Ultimately, Waltzer hopes EOR can lower the cost of carbon capture and advance sequestration technology, thereby opening the door to other forms of storage that don't involve oil drilling. "If there was ever a chance for a big idea to succeed in our current political climate," he wrote last year, "EOR is it."