In June 1991, Mount Pinatubo in the Philippines erupted with extraordinary force, blasting millions of tonnes of sulphur dioxide high into the stratosphere. Over the following months, a fine haze of tiny particles spread around the globe, reflecting a little of the sun's light back into space. For a year or two, average temperatures across the planet fell noticeably. To most observers, this was simply a curiosity of nature. To a small number of scientists, however, it hinted at something far more provocative: the possibility that humanity might one day cool the Earth deliberately.
That idea lies at the heart of geoengineering, a term that covers a range of large-scale, intentional interventions in the climate system. Broadly speaking, the proposals fall into two families. The first, often called solar radiation management, aims to reflect a small fraction of sunlight away from the Earth, either by releasing reflective particles into the upper atmosphere or by spraying fine sea salt into low clouds to make them brighter. The second, carbon dioxide removal, seeks to draw the gas that is already warming the planet out of the air and lock it away, whether by planting forests on a vast scale, by building machines that capture it chemically, or by fertilising the ocean so that plankton absorb more of it.
Why should such ideas be taken seriously at all? Supporters point out that, even if emissions were cut sharply tomorrow, the carbon already in the atmosphere would go on warming the planet for decades. Solar techniques, in particular, are attractive precisely because they could act quickly and, in principle, relatively cheaply. 'Nobody I know regards this as a substitute for cutting emissions,' says Dr Hannah Okafor, a climate modeller who studies the risks. 'The question is whether we ought to understand it properly, in case we one day face a crisis and someone is tempted to try it without that understanding.'
Critics, by contrast, regard the whole field with deep unease, and their objections are not easily dismissed. Reflecting sunlight would do nothing to stop carbon dioxide from dissolving in seawater and making the oceans more acidic. It might also alter rainfall patterns, and the regions most affected could well be those least responsible for the problem. Then there is the danger known as termination shock: if a programme of particle release were stopped suddenly, after years of masking the warming, temperatures could rebound rapidly, leaving ecosystems little time to adapt.
Perhaps the most widely discussed concern, however, is not technical but psychological. Economists call it moral hazard: the worry that the mere promise of a technological fix might weaken the will to make difficult changes now. If governments and industries come to believe that a cheap solution is waiting in reserve, the argument goes, they will feel less pressure to phase out fossil fuels. Supporters reply that this fear is speculative, and that ignorance is hardly a safer basis for decisions than research.
Carbon removal attracts less alarm, yet it raises difficulties of its own. Forests take decades to mature and can burn down, releasing what they stored; machines that pull carbon from the air remain expensive and consume considerable amounts of energy; and adding nutrients to the sea could have knock-on effects on marine life that are hard to predict. Even so, many climate scientists now accept that some form of removal will probably be needed alongside rapid cuts in emissions if the most ambitious climate goals are to be met.
Underlying all of this is a question of governance that no laboratory can answer. The climate is shared by every nation, yet a single country, or even a wealthy individual, might in theory be able to launch a solar scheme alone. Who would decide how much cooling is enough? Who would be held accountable if a monsoon failed? Without international agreement, many fear that the technology could become a source of conflict rather than a solution.
Rarely has a scientific idea provoked such a mixture of hope and anxiety. What most researchers do agree on is that geoengineering is no silver bullet. At best, it might buy time or ease some of the damage while the real work of reducing emissions goes on; at worst, it could become a dangerous distraction. Which of these it turns out to be will depend less on the cleverness of the engineering than on the wisdom of the people who choose whether, and how, to use it.