ECO-ENV-0022-Stock iStockphoto/Jan Rysavy

After years of deadlock over climate policy, Congress appears poised to enact the first federal limits on greenhouse gas emissions this fall. Yet a growing number of climate scientists and scholars believe that such efforts are likely to be too little, too late to stop warming--and that, consequently, a broader view of our climate policy options is needed.

On Monday, the National Academy of Sciences will convene a workshop to explore the question of "geoengineering" the climate--that is, influencing the global environment in ways that would restore balance to the global energy system and cool the planet until emissions reductions take effect.

Why even consider such an extreme and potentially dangerous response to climate change, just when emissions limits are likely to be enacted? Such an idea may seem outlandish at first blush, but it deserves serious study.

Despite the progress we may see in the coming years, the mathematics and politics of rapid greenhouse gas reductions remain remarkably daunting. China's emissions exploded in the past decade; if left unchecked, they alone will equal the current global level by 2030. Worldwide energy consumption is expected to increase 50 percent from 2005 levels by 2030, with the bulk of the growth coming in the developing world.

Many climate scientists believe that a significant degree of warming is already "locked in" by past emissions and that greenhouse gas concentrations have already reached potentially dangerous levels. To avoid warming, therefore, global emissions would have to be halted immediately--and existing emissions would have to be removed from the atmosphere as well. Not a likely prospect.

Even if the international climate treaty due to be negotiated in Copenhagen in December is vastly more stringent and effective than the Kyoto Protocol, it will take decades to eliminate net global emissions. Warming seems inevitable; the only questions are its timing, distribution and severity. The effects may prove to be modest--but they could be severe or perhaps catastrophic.

Even strong advocates of limiting emissions have concluded that global emissions controls are likely to take effect too slowly and too unevenly to avoid substantial risk of severe damage--and that it would be prudent to pursue research on geoengineering. White House science adviser John Holdren recently explained that "we have to look at the possibilities and understand them--including their shortcomings--because if other approaches to mitigation fall short, the geoengineering approach will end up being considered."

Could we really cool the planet? Although the science is nascent, evidence suggests that we could.

The Earth is warmed by two forces: solar radiation, which enters the atmosphere, and the greenhouse gases that trap it there. If greenhouse gases rise to dangerous levels, it might be possible to stabilize the climate by reducing the amount of solar radiation that reaches the Earth's surface or reflecting more of it back into space, restoring balance to the global energy system. The science is in its infancy, but there is intriguing preliminary evidence that it could be done--quickly, effectively and affordably, three attributes no other climate policy can claim.

The most promising ideas take their proof of concept from nature. Scientists noted that the 1991 eruption of Mount Pinatubo in the Philippines cooled the planet for two to three years by roughly half a degree Celsius. There are ways of artificially reproducing this effect. For instance, ultra-fine sulfur particles injected into the upper atmosphere could deflect 1 or 2 percent of incoming sunlight--almost unnoticeable but enough to cancel out the warming expected to occur this century. This would not halt the slow acidification of the oceans caused by elevated levels of carbon dioxide, but it could cool the planet and drastically reduce the heat-related damages we would otherwise experience, buying time for emissions reductions to take effect.

There are other potentially effective geoengineering techniques that deserve serious study. For instance, low-altitude marine stratocumuli clouds, which cover about 25 percent of the world's oceans, also reflect sunlight. Research suggests that it might be possible to increase the reflective abilities of these clouds by spraying a fine mist of seawater into the air. A fleet of roughly 1,500 ships (estimated cost: $2 million per ship) might be able to increase the reflectivity of these clouds by 10 percent, enough to counteract anticipated warming.

There is considerable scientific evidence to support the key concepts--but no federal resources to support research. The National Academy of Sciences, NASA and the Energy Department have all concluded that geoengineering could be (as the National Academy put it) "feasible, economical and capable." All three recommended further study. The Bush administration declined to pursue a five-year, $64 million geoengineering research program that its Energy Department proposed. The Obama administration should revisit that decision.

A geoengineering system would of course be controversial, but the policy question we face today is simple: Should the federal government conduct research on geoengineering? The scientific and engineering challenges involved in geoengineering the global climate for decades, and the potential consequences of success or failure, are extraordinary; all the more reason to begin a research program commensurate with the scale and significance of the task.

Geoengineering is not a substitute for mitigation, and it raises potentially serious environmental and ethical issues. It could, however, protect us from the worst effects of warming for the many decades it will take for emissions reductions to become effective. We may ultimately decide that geoengineering's risks are too great--but undertaking a research program now would give future policymakers the opportunity to make decisions about geoengineering from a position of knowledge rather than ignorance and desperation.

Samuel Thernstrom is a resident fellow at AEI, director of the AEI Press, and codirector of AEI's geoengineering project.