Fixing the climate

Technological schemes to combat global warming are viewed as wacky or impractical. But they belong in the mainstream debate on climate change
December 16, 2009
John Latham and Stephen Salter’s idea of spraying up sea salt is one of the proposed schemes that might cool the planet

If you want to know why many people take a dim view of geoengineering—a catch-all term for technological interventions to cool the climate system—look no further than Richard Branson. In October, Branson told the Wall Street Journal that “if we could come up with a geoengineering answer… then Copenhagen wouldn’t be necessary. We could carry on flying our planes and driving our cars.” But the idea that such schemes—for example, giant mirrors in space, artificially-brightened clouds or vast airborne Hoovers sucking up carbon dioxide—are a reasonable response to carbon emissions that can allow humanity to carry on polluting would strike most people as mad, bad or both. Doug Parr, chief scientist at Greenpeace, even said in 2008 that many of these ideas were “outright dangerous.”

If so, Richard Branson isn’t the only dangerous man. In recent months climate fixes have become quite the buzzy topic. In September 2009 the Royal Society released the first major report on the subject by a national academy of science. Around the same time “sceptical environmentalist” Bjørn Lomborg brought together five eminent economists to assess potential climate change solutions. The group concluded that research into geoengineering offered a better “cost-benefit ratio” than any other approach. The Intergovernmental Panel on Climate Change (IPCC) will discuss many aspects of the subject when the next of its hugely influential Assessment Reports appears in 2013, having given it only a few pages in its 2007 report. The IPCC’s chair, Rajendra Pachauri, has talked of carbon dioxide removal technologies probably being necessary in the long run. Less significantly, but with far higher visibility, economist Steven D Levitt and journalist Stephen Dubner’s book Superfreakonomics served up various unconvincing and breathless claims on the topic with a side order of needless error, leading to epic levels of vituperation in the blogosphere, and many delightfully snarky reviews.

To understand why this debate excites such passion, start with Branson’s idea that geoengineering could permit people to keep on burning fossil fuels. Here environmentalists and scientists agree that no geoengineering scheme can simply cancel out the effects of rising emissions. Greenhouse warming has distinctive effects on the climate that cannot just be reversed by reducing incoming sunlight; removing carbon dioxide from the atmosphere while continuing to emit more of it cannot in the near term reduce concentrations. So the one all-but-universally agreed opinion on geoengineering is that while it may ameliorate some aspects of climate change, it is not an alternative to reducing emissions. A second round of objections runs that geoengineering would be used to delay tough action on emissions reduction today in favour of technological fixes tomorrow. So not only should we doubt airline-owners like Branson, but we should treat any scientific research on the topic as inherently dangerous. “It is naïve to think that politicians who have failed to deliver on mitigation targets will not jump at the opportunity for a ‘techno-fix,’” says Jim Thomas of Canadian environmental group ETC.

Yet you don’t have to be an expert in deconstruction or denial to sense that behind these geoengineering-is-not-an-alternative arguments lies the fear that, at some level, it really is one. The Royal Society’s report’s press release read: “Stop emitting CO2 or geoengineering could be our only hope.” Its authors thus framed the technology as an unsavoury “plan B” that merited study only to use as an “emergency response” to the imminent collapse of the Greenland ice sheet or similar catastrophe. But what is a Plan B if not an alternative, even if it is not an alternative that you want? To understand geoengineering better, concentrate not on what it isn’t, or what you don’t want it to be; look instead at what it is, and what it could become. Geoengineering is not an alternative, but it can be an addition. This neglected set of ways in which people can alter climate should be part of mainstream debate on climate change, studied and assessed as a part of the whole. And that is going to require a far greater level of research. Despite all the public discussion there are only a few dozen people in the world contributing to the scientific literature.

Much of the “otherness” of geoengineering stems from the fact that it seems too futuristic and unprecedented. Yet global warming is itself an increasingly deliberate activity. No one pumps oil from 5km below the seabed, refines it into gasoline and sells it through a retail network by accident. Selling—or buying—fossil fuels does not take place for the purpose of climate change, but that change can no longer be seen as an unintended consequence. What is more, other climate-changing processes already exist, and are under our explicit control. As a recent study chaired by David Lee of Manchester Metropolitan University confirmed, the world’s shipping industry provides fairly significant global cooling. Ships burn fuel high in sulphur, and so emit lots of sulphur dioxide. This forms tiny particles of sulphuric acid, that enrich the water-droplet content of low-lying clouds over the oceans. More droplets mean whiter clouds, and whiter clouds mean more sunlight reflected into space. The shipping industry thus works, in Lee’s words, as a form of “inadvertent geoengineering.”

This example illustrates the complexity of climate change. Far from being a unitary problem to be solved, it is a range of causes, effects, meanings and implications. Along with carbon dioxide, there are the greenhouse gases methane, ozone, nitrous oxide and more. Then there are the problems of deforestation, irrigation, urbanisation, nitrogen deposition, nitrogen run-off, chemical smogs, and so on. These items interact in forbiddingly complex ways, not only with each other but also with the economy, and with people’s values and ideological stances. It makes much more sense to create a home for geoengineering within this complexity than to keep it out. “The climate is complicated,” says Tim Lenton, a professor of Earth systems science at the University of East Anglia who works in this field. “Why would we try to control it using just the one knob?”


Which knobs should we seek to twiddle? Pull back from the complexity, and imagine our climate to be a bathtub full of energy. The energy flows in from the sunshine tap in the form of visible light, and it flows down the cosmic plughole as infrared radiation—that is, waste heat. Normally the input will match the output, but greenhouse gases have partially blocked the drain. The level of energy in the bath is rising, so the climate is getting hotter. Present responses to climate change are adaptation (goggles and a snorkel) or mitigation (reducing the rate at which the drain gets further blocked). If mitigation succeeds in the way most people envisage—by reducing emissions to more or less zero over the course of this century—the energy will stop rising, but it would not fall back to its old level. The drain might not be becoming more clogged, but it would not be becoming unclogged either.

Geoengineering offers two ways to empty the bath. You can reach out a toe to turn the sunlight tap, or you can find a plunger and unclog the drain. “Tap” methods reflect sunlight away by putting aerosols into the stratosphere or brightening clouds, painting roofs white and so on. “Drain” methods almost all concentrate on removing carbon dioxide from the atmosphere, either by chemical engineering or by planting trees and crops, which must be disposed of so that the carbon they store during photosynthesis does not get released again.

For almost as long as people have been worrying about global warming, they have been talking about these kinds of geo-engineering solutions. Back in the 1970s Mikhail Budyko, a Soviet climatologist, first suggested putting a veil of sulphate particles into the stratosphere by burning sulphur in jet planes, thus reproducing the cooling hazes known to follow volcanic eruptions. Around the same time, Freeman Dyson, a physicist given to futuristic speculation, worked out a plan to suck up carbon emissions with new forests while also developing alternative energy on the scale required to cut emissions to zero. By the early 1990s, more or less all the forms of geoengineering now under discussion—from the more exotic space mirrors to brightened grasslands and deserts, new forests, fertilising the oceans to create algae blooms and sucking the carbon dioxide out of the sky—had been looked at.

But the research did not take off. Many researchers shunned the topic fearing it a diversion from the real problem. There was a fear, too, of unexpected consequences—what, for example, might a stratospheric veil do to the ozone layer?—not to mention the hubris of it all. So from the 1990s to the mid-2000s, very few scientists worked on the problem and there was very little funding for it. And both those things remain largely true today. As a result, such interest as there has been has often come from outside traditional science. In 2007, Richard Branson launched his Virgin Earth Challenge, a $25m prize for the development of a successful carbon dioxide removal system. Nathan Myhrvold, former chief technology officer of Microsoft, invented a scheme that pumped aerosols into the stratosphere using airship-supported hosepipes (see box p33), which also formed the basis for the controversial chapter in Dubner and Levitt’s Superfreakonomics.

That said, there has been rising academic interest in the past few years, which can be traced to a 2006 paper by Paul Crutzen, a Nobel-prize winning chemist and perhaps the world’s most influential atmospheric scientist. In the early 1970s, Crutzen’s studies revealed the fragility of the ozone layer. In the 1980s he raised the possibility that a nuclear winter might result from the smoke and soot produced by a nuclear war. His suggestion in 2000 that human influence on climate is so great that the industrial revolution marked the start of a new geological era—the anthropocene—crystallised how many researchers felt. Crutzen’s scientific reputation gave geoengineering an imprimatur it had never possessed before.

While his paper did not add much substance to what had already been written in the previous ten years, its novelty lay in its suggestions of how recent events could give the idea more traction. The 1991 eruption of Mount Pinatubo in the Philippines pumped 10m tonnes of sulphur into the upper atmosphere. Climate monitoring after this event proved the theory of sulphate-aerosol cooling. Crutzen also cited research revealing that emissions acted to acidify the ocean, damaging ecosystems and potentially dooming coral reefs. This gave new weight to the argument that carbon dioxide emissions had to be controlled, regardless of geoengineering.

The greatest fillip for geoengineering research came from the fact that after more than a decade of political pressure for emissions reduction, little progress had been made. There was a real possibility that emissions control would not happen in time, so another solution was needed. All these points were repeated in the subsequent smattering of reports, journals and meetings that have raised the subject’s profile.

Crutzen’s most original contribution in that paper, though, is often overlooked. Humans, he pointed out, already produce sulphate aerosols that cool the world. Burning fossil fuels, on land or sea, releases tens of millions of tonnes of sulphur dioxide. These form a haze of little particles that diffuse and reflect away sunlight, just as Pinatubo did. But this happens in the lower atmosphere, where the particles get into people’s lungs and damage their health. Rich countries started to scrub sulphur out of their emissions from the 1960s onwards, to reduce a death toll reckoned in the hundreds of thousands. Yet this process of atmospheric cleaning also lets in more sunlight, turning up warming. (Crutzen quoted a study arguing that worldwide air-quality controls could warm the planet by almost 1 degree Centrigrade.)

A choice between warming and deaths due to sulphate is a diabolical one. But, as Crutzen says, there is no need to choose. In the lower atmosphere rain washes the air clean in days; in the dry stratosphere, aerosols stay aloft for years, spreading worldwide. A comparatively small amount of sulphur in the stratosphere could thus preserve the cooling currently provided by a lot of sulphur lower down. Crutzen calculated that the cooling resulting from the 55m tonnes of sulphur emitted into the lower atmosphere every year could be provided instead by 1-2m tonnes of sulphur in the upper atmosphere. Health impacts would be vastly reduced while cooling would be preserved and spread more evenly.

This is not to say that such a “stratospheric veil” would not have implications. It would change the pattern of warmth in the climate, and the winds and ocean currents the Earth uses to move warmth around might behave differently in response. It would also change the water cycle: sunlight is what drives evaporation, and if it is turned down a little, the water cycle will slow down, with less evaporation and less rainfall. This effect is not necessarily dangerous, but it is worth watching—and the thicker the veil, the more problematic it would be. In a world with no emissions reductions and an ever thickening veil to balance an ever thickening greenhouse, this hydrological effect would become ever more extreme. That is one of the reasons why Ken Caldeira, a respected Stanford climate scientist who pioneered geoengineering modelling, insists that any such scheme happen only alongside emissions cuts.

But Crutzen did not imagine thickening the veil continuously. He didn’t even imagine making it thick enough to deal with all the warming associated with carbon dioxide, as most studies have done. He only imagined it thick enough to offset the cooling lost from other pollution reduction. Accepting that geoengineering can be worthwhile even with quite small effects is an important part of understanding its potential application. For example, choosing crops with lighter leaves is a minor geoengineering proposal that would have only a small effect at the global scale, and is therefore discounted by many people. But it should have an effect, and one felt most strongly in summers in the temperate north, where it could hedge against heat waves.

Turning down the sunshine tap is only half the geoengineering story. The other half, clearing the drain, can be done in numerous ways: trees, energy crops, blooms of algae, chemical engineering or hastening the natural reactions between rocks and carbon dioxide that go on in the slow reaches of geological time. Carbon dioxide could be stored in depleted oil fields, in aquifers, or in the ocean depths; biomass could be ploughed into soil in the form of charcoal, or chemically reclaimed carbon dioxide could be transformed into solid carbonate rocks. This range of techniques fits our moral intuitions; if you make a mess, you should clean it up. It also offers hope on ocean acidification. Because of this, such geoengineering by means of carbon dioxide removal is becoming positively mainstream, at least for long-term planning. It offers a way to reduce temperatures from whatever peak they might reach. Nasa climate scientist James Hansen along with other environmentalists (and an increasingly high-profile group called argue that the long-term stability of the icecaps needs a lower carbon-dioxide level than today’s. If so, then this can only be guaranteed by the use of some sort of removal technology.

If humans already had the technology to remove carbon dioxide from the atmosphere as fast as they put it in, a large part of the climate problem would be dealt with. But there is nothing like that capacity at present. Artificial ways of capturing carbon dioxide from the air are still in their infancy. Geological storage of carbon dioxide has yet to be demonstrated on large scales. Forestry schemes large enough to make a big dent require plantations the size of countries. In very rough terms, a removal system capable of dealing with the world’s emissions would be on a similar scale to the industries that mine, pump, refine and transport the fossil fuels responsible for those emissions, and would use similar amounts of energy. Mining, gas and oil are big industries.

While in the near term there is no way that carbon removal can take up more than a tiny fraction of emissions, in the long term such technology will get cheaper and more efficient. But believing that carbon dioxide emitted today can be mopped up later is potentially dangerous. All the time carbon is in the atmosphere it heats the world. So, just as you get a mismatch in the spatial pattern of warming if you heat the world with greenhouse gases while cooling it by turning down the sunshine tap, you can also get a mismatch in time if you heat the world with greenhouse gases and only later take them away by unblocking the drain. Emit a tonne of carbon dioxide today and mop it up in 50 years, and you still have the 50 years of warming to cope with.

Still, in the long term carbon dioxide removal is likely to play a significant role, if it gets cheap enough. Schemes to turn down the sunshine tap could make a difference sooner, as they would be relatively easy to deploy. But that does not make them a cheap solution. A stratospheric veil might cost the turnover of one large global company, rather than a whole industry—just the kind of reasonable price that endeared such schemes to the superfreakonomists, and Bjørn Lomborg’s five economists. But to look at the costs of the veil alone misses the bigger picture. If you accept that there is no merit in a world where the greenhouse effect gets stronger and the stratospheric veil gets thicker (and the climate anomalies caused by the mismatch between the two get worse) then the costs of geoengineering become additional to the costs of emissions reduction, not a substitute for them.

This is not to say that geoengineering could offer no economic advantages. Shortly after Paul Crutzen’s 2006 article, Tom Wigley, a respected climate scientist then at the National Center for Atmospheric Research in Boulder, Colorado, looked at ways of combining emissions reduction and sunlight reduction. Wigley suggested that sulphates might be squirted into the stratosphere in the near term as a way to slow the rate of warming and buy time for the massive and costly industrial shift to alternative energy. Wigley’s “buying time” approach has not enjoyed much enthusiasm from other researchers, who fear that it will reduce the sense of urgency needed to drive emissions elimination. The Royal Society report spoke for many in treating geoengineering techniques only as an insurance policy. But this is also inconsistent. Rejecting the Wigley scenario reflects a view that political decision-making cannot summon the nuanced, self-disciplined approach needed to geoengineer a little without losing your commitment to reducing emissions a lot. The “Plan B” scenario rests on a political process with characteristics just as unlikely: it requires schemes to be researched in depth but to stay unused until (but only until) some unspecified assessment commanding international political assent deems disaster imminent but not unavoidable. Good luck with that.

Whichever scenario you prefer, both require research into how to make the various schemes work and what their effects might be. Some of this can be done with models, but field trials will also be desirable, especially with schemes that aim to affect clouds. Here Britain may have something to offer. The number and size of the water droplets in a cloud depends on the number of tiny particles called “condensation nuclei” in the air. Other things being equal more droplets makes for a brighter, whiter, more reflective cloud. Over the oceans, where there is a dearth of condensation nuclei, clouds are less white. In 1990 a British cloud physicist (and poet), John Latham, suggested that adding condensation nuclei in the form of a fine mist of sea salt might brighten maritime clouds enough to cool the climate. No one paid much attention, but a decade or so later Stephen Salter, a wildly inventive marine engineer at Edinburgh University, got wind of the idea and talked to Latham about schemes whereby such particles might be made. He hit on the idea of a fleet of thousands of wacky-looking sailing ships, using electricity from underwater turbines to spray up the required mist of tiny particles.

Set against the alternative of pumping or lifting a million tonnes of sulphur into the stratosphere every year, the idea of simply spraying up sea salt seems eminently feasible. Add the amiable presence of elderly British academics and hardware that looks like something out of Thunderbirds, and the idea appears charming. Real questions remain as to whether it would work—not just questions about how such cooling, if applied to relatively small patches of the ocean, would affect the global climate, but also more basic questions about whether the particles would actually get into the clouds, and whether the clouds would get whiter. Yet a limited field trial could be carried out quickly, and with minimal if any environmental risk—albeit probably not without some objections.

Such trials look even more plausible if put in the context of another cooling process that is being reversed: that of sulphur emissions from shipping. As mentioned before, the global shipping industry burns sulphur-rich fuel, and the sulphate particles from that fuel produce much of their cooling effect by brightening clouds. The industry’s carbon emissions—roughly the same as Germany’s—are likely to be regulated before too long, but are nevertheless rising. Its sulphate emissions, meanwhile, came under new regulation in 2008, and will be pretty much phased out by 2020. So relatively obscure shipping regulations designed to protect public health near ports and shipping lanes will inadvertently commit the world to a significant extra warming. Looked at this way, the case for trialling systems that deliberately try to restore a similar cooling, with no damage to health, seems strong.

Crafting policy approaches to geoengineering that treat it neither as pariah nor panacea will be hard—perhaps, in the end, impossible. Alan Robock, a climate scientist at Rutgers University in New Jersey, who works on geoengineering ideas in an adversarial but fair way, has put together an impressive list of “20 reasons why geoengineering may be a bad idea.” Some of these can be firmed up or crossed off the list by further research, and he has deleted three since the list was first published in 2008. But other problems on Robock’s list are more permanent and troubling—none more so than the question of who controls the process. To see geoengineering as just another form of climate change, as I think we should, is not to deny that it has distinct features. Stratospheric aerosol options, in particular, may be cheap and practical enough to be undertaken by a single country, like China, or even conceivably, a single rich individual, like Richard Branson. In short, climate change could be undertaken unilaterally.

This is a real risk, and one to take seriously. But it is perhaps not as unusual as it seems. Countries, especially large ones, already have powers beyond their borders. But a repertoire of tactics exists to restrain their use. An individual, or a small country, that tried to geoengineer could easily be forced to desist, militarily if need be. For bigger powers there is less that can be done. But there are already lots of tensions between large powers—over trade, economics, nuclear proliferation and indeed climate. Geoengineering might sharpen such questions, but climate issues are already effectively in the hands of China, America and Europe.

The lack of a system of governance for something as potentially powerful as geoengineering is alarming. But so is the lack of an effective system of governance for most of the rest of climate change. In trying to put together such a system, as at the Copenhagen summit, people and governments have accepted the idea that something can be done, that at least some of the responses of Earth’s system can be predicted, that the risks of climate change can, in some way, be governed, and that there are ways of choosing between better outcomes and worse ones. They are, in short, seeking to manage the risks of climate change. To research, judiciously, new technical means by which to do so is not to change the game, but simply to expand its possibilities. It is a natural outcome of taking climate change seriously—not as a single problem with a specific solution, but as the context in which the next century’s history is going to be made.