Summary paper prepared for the National Commission on Energy Policy, Task Force on Geoengineering
If the term geoengineering is restricted to deliberate manipulation of the entire global climate, the most evident possibility derives from a remarkable feature of nature. For whatever deep or arbitrary reason, sulfate molecules injected into the stratosphere can offset the radiative forcing effect of the principal greenhouse gas, carbon dioxide, with spectacular efficiency. A single kilogram of sulfate particles of appropriate size could negate the thermal impulse imparted by at least two hundred thousand and perhaps three hundred thousand kilograms of carbon dioxide. The four order of magnitude difference means that the principal measure of global warming – average surface temperature – could be reduced by half a degree centigrade within a single year at modest cost. Volcano eruptions have demonstrated that effect. Many countries and even private organizations could almost certainly emulate what they do. There may also be alternative methods of accomplishing the same result, but those currently discussed are much more speculative. The sulfate option is the dominant immediate issue.
There is, of course, a major catch with that option. The thermal impulse of carbon dioxide may be offset by stratospheric sulfates, but it would not be removed and is far more enduring. Sulfates precipitate out of the atmosphere in the course of a year or two. Carbon dioxide molecules and their radiative effect persist for more than a century. Stratospheric sulfates would not remove the danger of ocean acidification posed by carbon dioxide, and their supposedly beneficial effect on annual average surface temperature would be accompanied by localized changes to precipitation and other weather patterns that are currently unpredictable and are likely to remain so. Hence any advantage resulting from a reduction of average temperature could be negated by the aggregate consequences of ocean chemistry and local weather changes that cannot be precisely anticipated and may or may not be accurately recorded after the fact.
Until recently the inherent uncertainties entailed in global climate manipulation were considered to be grounds for categorical proscription. “If it were done when it were done,” said the infamous Lady Macbeth about the prospective murder of Duncan, “then it were well it were done quickly.” Messing with the global climate was assumed to be the equivalent of murdering the king, and the presumption was that ultimate grief would overwhelm fleeting glory, as in the iconic play. As evidence accumulates on the effects of average temperature increases, however, particularly regarding rapid changes in the dynamics of glacial ice, there have been prominent second thoughts. The possibility of some catastrophic effect occurring on a schedule more rapid that any feasible process of mitigation has inspired fears of a climate emergency that would justify immediate intervention. Mainstream scientists have concluded that geoengineering techniques should at least be investigated, and that in turn has posed the problem of prudent regulation.
The situation is essentially unprecedented. The average surface temperature on earth is already at a level that has not occurred for several thousand years, and the increase projected to result over the current century would take it to a level not experienced for tens of millions of years. The rate of increase projected for this century would be substantially more rapid than any period on the geological record. The current state of knowledge is not sufficient to determine the consequences to scientific standards of confidence, but there are some cataclysmic possibilities. Most alarming, perhaps, would be a surging release of frozen gas hydrates from arctic tundra or ocean deposits. The radiative effect of methane is a factor of 21 greater than that of carbon dioxide, and a surging release could generate a positive feedback cycle with truly monumental consequences. That is currently judged to be too unlikely to be a practical concern, but the state of understanding on which that judgment is based cannot provide categorical reassurance.