A scientist from the Los Alamos National Laboratory in New Mexico, which was at the centre of the Manhattan Project during the Second World War, yesterday proposed a plan to cool down the Earth's climate should international efforts to limit emission of greenhouse gases fail.
Dust, iron fillings and artificial clouds could all play a part in slowing global warming.
Greg Canavan, a senior scientific adviser at Los Alamos, told the annual meeting of the American Association for the Advancement of Science that 'climate engineering' may be the only option left if governments fail to control emissions of man-made carbon dioxide.
'Engineering the Earth's climate is coming within reach. If I had to make a guess, all the conditions for doing something are probably only about three decades off,' he told the Independent on Sunday.
Dr Canavan is among a growing number of scientists who believe climate engineering may be possible by injecting particles into the atmosphere to deflect incoming sunlight, or sprinkling iron filings into the oceans to stimulate the growth of plankton that can absorb excess carbon dioxide from the atmosphere.
One idea being treated seriously is to use 16in naval guns to shoot 20 million tons of dust annually into the upper atmosphere to mimic the eruption of the Mount Pinatubo volcano in the Philippines, which caused global cooling.
Scientists estimate this amount of dust would reduce global heating by about 1 watt per square metre, enough to offset the same amount of heat from the greenhouse gases emitted by human activity over the past few decades.
Another idea is to try to create clouds to provide more cover over the oceans or sprinkling iron filings in the oceans to stimulate the growth of microscopic plants, an idea that was tested last year with unpromising results, according to Richard Barber, an oceanographer at Duke University, in Durham, North Carolina.
Mixing and circulation in the oceans 'greatly complicated' the estimates of the effect on absorbing carbon dioxide, he said. The Los Alamos laboratory is keen to promote the technology of monitoring climate engineering because the network of 30 to 100 satellites needed for the job could also double up as an early-warning system, Dr Canavan said. 'There is enough synergy in sensors, satellites and data- handling that climate engineering and defence applications might be able to share technologies.'
Dr Canavan said that no large-scale climate engineering should be done unless a network of satellites was in place to monitor the effects and scientists were convinced the benefits outweighed the risks.
Climate scientists are sceptical about trying to create a 'technological fix' to mend a sick planet, but admit there may be little choice if greenhouse gases continue to be released.
Stephen Schneider, a leading atmospheric scientist at California's Stanford University, said: 'I'd prefer to start to lower the human impact on the Earth through conventional, less exotic means - even if I do reluctantly agree that study of geoengineering potential is probably needed, given our growing impact on the planet.'
He said that in a worst-case scenario, when global warming rapidly accelerates in the first half of the next century 'we could find ourselves in 2047 with nine billion people hooked on the remaining reserves of soft 'dirty' coal in China, India and the US.
'Facing social and economic collapse if we go cold-turkey off that fossil fuel, we may find ourselves throwing dust into the atmosphere whether we like it or not.'Reuse content