Many things will have to happen if we are to lower greenhouse-gas emissions enough by 2050 to avoid catastrophic climate change. One of those things, according to a number of projections, is the large-scale deployment of carbon capture and sequestration technology, or CCS. In this process, carbon dioxide emissions from such sources as fossil-fuel power plants and industrial facilities are collected, compressed, transported to a storage site, and injected deep underground for permanent storage. Indeed, with demand for fossil fuels still soaring, the International Energy Agency predicts that one-fifth of the carbon dioxide reductions necessary by 2050 will have to come from CCS (shown in blue in the chart above).

The problem is that large-scale CCS is poorly understood and prohibitively expensive. As a result, there are not yet any facilities aimed at controlling power plant emissions. In the absence of government incentives, technology for injecting carbon dioxide underground is used mainly by oil companies as a way to extract hard-to-reach oil; these and other projects are injecting about 20 million metric tons of the gas annually. By the IEA’s estimate, annual carbon dioxide sequestration will need to jump by a factor of about 15 by 2020 and 120 by 2030, and at least 110 more big CCS facilities must come online in the next eight years. Although 65 are in planning or construction phases, building one can take more than a decade. This problem might not be one we can bury.