On February 16, the Kyoto Treaty will take effect, following Russia's ratification last November. For the next seven years, the 132 signatory nations will strive to curb their emissions of carbon dioxide (CO₂) and other heat-trapping gases in an effort to put the brakes on climate change. But how best to achieve this goal in the long term is unclear. Several emissions heavyweights--including the U.S., which produces nearly a quarter of the world's CO₂--have refused to abide by Kyoto, and international representatives meeting in Buenos Aires in December failed to agree on a successor to the agreement, which will expire in 2012.

One approach that is gaining currency among environmental scientists is carbon dioxide capture and storage (CCS), a form of carbon sequestration in which CO₂ is removed from the waste gas of power plants, typically by absorbing it in a liquid, and subsequently burying it deep underground, hence keeping the gas out of the atmosphere. Abandoned oil fields and briny aquifers can impound much more carbon than the plants and soil envisioned in some sequestration schemes can, and are more practical--at least for now--than deep ocean storage.

Indeed, CCS technology is already in use. Since 1996, the Norwegian company Statoil has been stripping about a million tons of CO₂ a year out of natural gas from the Sleipner West field under the North Sea and injecting it at high pressure into a saline aquifer. Avoiding Norway's tax on CO₂ emissions balanced most of the costs of the project. An even larger project, begun in 2000, takes CO₂ from North Dakota and sinks it into an oil field in Weyburn, Saskatchewan, Canada. Here, selling the extra oil that the injected C0₂ squeezes out largely offsets the costs. These efforts and others show that the sequestered CO₂ can be monitored, and that it largely stays confined underground.

But these cases notwithstanding, CCS is expensive. Capturing even a modest part of the 25 billion tons of CO₂ emitted each year, experts say, will require economic incentives. On December 8, the bipartisan National Commission on Energy Policy, which includes academics, industrialists and environmentalists, proposed a broad course of action for the U.S. The commission recommended a mandatory system of tradable permits to emit carbon dioxide and other greenhouse gases. (A similar mandatory system had been proposed in the McCain-Lieberman Climate Stewardship Act, which was defeated in the Senate in 2003.) To minimize economic disruption, the commission advised that the government cap the cost by selling extra permits starting at $7 per metric ton of CO₂ in 2010 with a 5 percent annual increase. In contrast, in Europe, which initiated trading of carbon credits on January 1 to meet its Kyoto commitments, credits already sell for about $11 to $12 per ton.

But even the higher European credit prices aren't going to be enough to promote widespread adoption of CCS, which currently costs $40 to $100 a ton. "There's still a gap between the marketplace and the technology," says Howard Herzog of the Laboratory for Energy and the Environment at the Massachusetts Institute of Technology.

The best way to make CCS cheaper is for government to provide incentives to use it, asserts David Hawkins of the environmental group Natural Resources Defense Council. "Learning by doing is the thing that drives the cost down, not R&D dollars," he adds. Between 2000 and 2030, power plants with a lifetime potential of more than 1,000 billion tons of CO₂ could be built, including many in coal-rich China, Hawkins observes. Extracting CO₂ from traditional coal plants is much less efficient than from gasification plants, where coal is first turned to a gas and reacted with water to form CO₂ and hydrogen. Because the new plants are somewhat more expensive and less reliable, "there's a real danger of lock-in" to plants that make CCS difficult, he warns.