Skip to main content

The Ass’s Dilemma: Can Man Engineer the Climate?

ISSUE:  Spring 2009

In 1816 the summer never came. Across New England and the Maritimes, late snows and repeated frosts led to crop failures, livestock deaths, and severe food shortages. In Europe, which was struggling to recover from the extended cataclysm of the Napoleonic Wars, torrential rain and abysmal cold resulted in the collapse of the German wheat crop, leading to meager harvests across the continent. Authorities struggled to quell riots in England and France as mobs raided storehouses, and troops were dispatched to protect grain shipments. In Switzerland, the poor resorted to eating cats and lichen, and the streets of Zurich were swarmed with beggars. Historian John Post has called it the last great subsistence crisis of the Western world, but today 1816 is generally remembered—when it is remembered at all—as “the year without a summer.”

A Painting of a Volcanic Eruption
Lithograph of Krakatoa from The Eruption of Krakatoa and Subsequent Phenomena (1888).

At the time, the cause of the strange weather was a mystery, but today we know that the main culprit was a massive volcanic eruption that occurred the previous spring on the other side of the globe. Mount Tambora, on the island of Sumbawa in Indonesia, blew its top in April 1815. The largest observed eruption in recorded history, it was far more explosive than the better-known Krakatoa event of 1883 and ejected ten times the volume of debris. The eruption killed at least ten thousand people instantly, entombing a culture and wiping out a language in the process. In the aftermath of the explosion, a column of ash rose nearly thirty miles skyward, and the resulting dust cloud spread in a wide band of latitude across the northern hemisphere. It was this cloud, and more specifically, the fine aerosols that formed as sulfuric acid condensed in the upper atmosphere, that caused temperatures to drop by backscattering solar radiation into space.

The definitive account of the Year without a Summer is the 1983 book Volcano Weather, by the late Henry Stommel, former director of the Woods Hole Oceanographic Institute, and his wife, Elizabeth. A straightforward mix of history and science, the book steers clear of conjecture and takes pains to debunk some of the sensationalist claims surrounding the events, but it ends on an oddly speculative note. In a brief coda entitled “Dust vs carbon dioxide,” the Stommels observe that while increases in atmospheric carbon dioxide due to “continued burning of the world’s coal reserves” was having a warming effect on the globe, “manmade dust and aerosols” were, at the same time, offsetting that effect. Finally, they submit:

It seems that the demonstration that Mount Tambora and other volcanoes from time to time have given us of how the temperature of the world can be altered by an almost imperceptible amount of dust points to a mechanism that in the next century may save us all from sweltering under the pall of the carbon dioxide that will by then envelop us.

It’s worth remembering that when Volcano Weather was published, the idea that human activity was raising the global thermostat, although hardly new, had not yet entered the mainstream consciousness. Even among environmentalists, far more attention was being paid to issues like acid rain caused by sulfur emissions from coal-fired plants, and the specter of “nuclear winter,” the hypothesis that, in the wake of an all-out nuclear war, clouds of dust and soot issuing from incinerated cities would cast Earth into an extended period of cold and darkness. Against that backdrop, the blithe suggestion that we could somehow use “manmade dust and aerosols”—the very stuff of acid rain and nuclear winter—to counteract an imperceptible rise in temperatures would have struck many as lunacy. A quarter-century later, now that global warming is the foremost environmental issue in most minds, it still does.

And yet, in recent years, a handful of notable scientists have advanced the idea that we may well have to engineer the climate if we hope to avoid catastrophic consequences from warming. The most prominent of these, Dutch atmospheric chemist Paul Crutzen, won a Nobel Prize in 1995 for his work in heading off another global catastrophe—the hole in the ozone layer. Today, thanks to global efforts to phase out certain chemicals, scientists are optimistic that the ozone layer is on the mend. Similarly, clean air regulations have been successful in reducing airborne particulates and sulfur compounds, which adversely affect human health and the environment.

That’s good news, except that, as the Stommels noted, those particulates have been masking the effects of the buildup in greenhouse gases. Eliminate the pollution entirely and models show that the world could warm by as much as .8 degrees Celsius across most of the planet, and a whopping 4 degrees in the Arctic. That’s on top of current warming projections. It is precisely this “policy dilemma” around which Crutzen framed a 2006 essay published in the journal Climatic Change, in which he argued that “research into the feasibility and environmental consequences of climate engineering” should be immediately undertaken and that discussion of the topic “should not be tabooed.” Specifically, the Nobel laureate proposed mimicking the cooling effect of volcanoes by lofting sulfate particles into the stratosphere as a possible “escape route against strongly increasing temperatures.”

In an editorial comment that accompanied Crutzen’s essay, Ralph Cicerone, president of the National Academy of Sciences, acknowledged that many atmospheric and climate scientists had opposed the publication of the paper “even after peer review and revisions, for various and sincere reasons that are not wholly scientific.” The overall sentiment was perhaps best reflected in a response by NASA climate modeler Gavin Schmidt, who criticized Crutzen’s paper on the blog RealClimate, even before it appeared. After running down geoengineering’s dubious, almost Strangelove-ian pedigree, Schmidt dismissed Crutzen’s idea as “unlikely to gain much traction”:

Maybe an analogy is useful to see why. Think of the climate as a small boat on a rather choppy ocean. Under normal circumstances the boat will rock to and fro, and there is a finite risk that the boat could be overturned by a rogue wave. But now one of the passengers has decided to stand up and is deliberately rocking the boat ever more violently. Someone suggests that this is likely to increase the chances of the boat capsizing. Another passenger then proposes that with his knowledge of chaotic dynamics he can counterbalance the first passenger and indeed, counter the natural rocking caused by the waves. But to do so he needs a huge array of sensors and enormous computational resources to be ready to react efficiently but still wouldn’t be able to guarantee absolute stability, and indeed, since the system is untested it might make things worse.

So is the answer to a known and increasing human influence on climate an ever more elaborate system to control the climate? Or should the person rocking the boat just sit down?

It is generally accepted that stabilization of carbon dioxide concentrations in the atmosphere will require a reduction in current manmade CO2 by 60 to 80 percent, and many politicians, including President Obama, have made that target part of their rhetoric. But while seemingly everyone agrees that society should be hitting the brakes on fossil fuel consumption, we appear to be stepping on the gas instead. Despite enactment of the Kyoto Protocol, which had the modest goal of cutting emissions to pre-1990 levels, actual worldwide emissions have grown by 37 percent since 1990. Furthermore, the increase in CO2 concentrations has been accelerating in recent years, largely due to the explosive industrialization of China, which has been building new coal-fired power plants at the rate of one a week. As Elizabeth Kolbert soberly observed in her 2006 book, Field Notes from a Catastrophe: “Over the next fifteen years, the size of China’s economy is expected to more than double. This projected growth, most of which will be fueled by coal, overwhelms not just all conservation projects that are underway in the United States, but also any that could be imagined.”

China and other developing nations are exempt from Kyoto, and under the Bush administration, the US–historically, the leading contributor to the buildup in greenhouse gases–withdrew from the treaty, pointing to China’s unbridled growth as an excuse to do nothing about its own emissions. The Obama administration will no doubt reengage in diplomatic efforts, but the growing economic crisis, combined with falling oil and coal prices, does not bode well for the negotiations at this year’s climate summit in Copenhagen. Furthermore, diplomacy is only part of the story. Fundamentally, global warming is an energy problem, and solving it will require nothing less than a complete and radical transformation of the world’s energy infrastructure, something most energy experts believe will take decades, if it happens at all.

As Paul Crutzen stressed in his essay, “The very best would be if emissions of the greenhouse gases could be reduced so much that the stratospheric sulfur release experiment would not need to take place. Currently, this looks like a pious wish.”

In October 2007, an opinion piece ran in the New York Times under the headline, “How to Cool the Globe.” Written by Ken Caldeira, a contributor to the Intergovernmental Panel on Climate Change (IPCC), the op-ed put forth the idea that by “tossing small particles into the stratosphere (above where jets fly)” we could potentially “cool the earth within months.” The author added that, “even if we could stop adding to greenhouse gases tomorrow, the earth would continue warming for decades—and remain hot for centuries,” before ending with a blunt rhetorical challenge. “Which is the more environmentally sensitive thing to do,” he asked, “let the Greenland ice sheet collapse and polar bears become extinct, or throw a little sulfate in the stratosphere? The second option is at least worth looking into.”

That “second option”—most often referred to as stratospheric sulfate injection—is hardly the only climate engineering proposal on the drawing board. Other schemes range from the merely goofy-sounding to the wildly implausible. They include: fertilizing the ocean with iron to stimulate carbon-absorbing plankton blooms; sending ships to sea equipped with huge fans to whip up banks of sea spray and enhance marine clouds; and launching trillions of tiny reflectors into geostationary orbit to shield the planet from the sun. Stratospheric sulfate injection is related to the last two schemes in that it would attempt to manage Earth’s energy balance by modulating incoming solar radiation. It differs, however, in two significant respects: first, it could (at least theoretically) be deployed rather cheaply (suggested methods of delivering particles to the upper atmosphere include naval artillery, airplanes burning sulfur-doped fuel, and hoses tethered to high-flying balloons); second, nature has already run the experiment.

When Mount Tambora erupted, news still traveled on the trade winds and meteorological observations were scattershot. But when Mount Pinatubo, in the Philippines, erupted in 1991, the process unfolded under the watchful eyes of satellites, and scientists around the world could study the effects in real time. Pinatubo was no match for Tambora, but it was still massive—the largest eruption of the twentieth century. The explosion sent some twenty million tons of sulfur dioxide into the stratosphere, and the dust veil encircled the globe in three weeks. Unlike Tambora, which erupted toward the end of a centuries-long cold period known as the Little Ice Age, Pinatubo blew amid rapid warming. In its wake, the global average temperature, which had been climbing steeply, suddenly fell by roughly half a degree Celsius—a drop accurately predicted by climate models. The cooling was only temporary, however. Within a couple of years, the mercury was climbing once again. To date, the fourteen warmest years on record have all occurred since 1990.

One bright, warm day last March, I drove from Berkeley to the Stanford campus in Palo Alto to meet Ken Caldeira at the Department of Global Ecology at the Carnegie Institution for Science, where he runs a lab focusing on Earth’s carbon cycle, ocean acidification, and, of course, geoengineering. Caldeira is in his early fifties but looks younger. His round, youthful face is topped by a fringe of curls, and although he laughs easily, the mirthful countenance is tempered by heavily-lidded eyes that, at times, give him a sad, almost hangdog look. As we walk to a nearby cafeteria, he tells me that the first time he heard of stratospheric sulfate injection, he was incredulous. “I thought to myself, ‘Oh, there’s no way this is going to work!’ So we did the simulations, we modeled it, thinking we could just dismiss the idea and move on. To our surprise, it actually worked quite well. And a lot of the reason has to do with sea ice.” Over lunch, he explained that Arctic sea ice not only reflects sunlight away from the planet, but it also insulates the atmosphere from the ocean. “So, in polar winter, if there’s no sea ice, all the heat can rush out of the ocean. And in the summertime if there’s no ice, the heat can rush back into the ocean. But if you turn down the sun enough, so to speak, such that sea ice restores back to more or less its normal extent, well … what you find is that the response of the climate system is largely influenced by the feedbacks, not the details of the forcing.”

A “forcing” is what climate scientists call anything that alters the natural energy balance of Earth’s atmosphere. Volcanic eruptions are a negative forcing since the aerosols they produce subtract energy from the system by blocking sunlight. Carbon dioxide emissions, on the other hand, are a positive forcing, as they enhance the natural greenhouse effect and trap more heat in the atmosphere. As surface temperatures increase due to that forcing, melting ice leads to what is called a “positive feedback loop”: the more the ice melts, the more heat the ocean absorbs and the more ice it can melt —and so on, until it’s gone. Filtering out a small fraction of incoming sunlight would, if Caldeira’s models are correct, keep that vicious cycle from unspooling.

If “turning down the sun” sounds like a patently bad idea for plant life, Caldeira has run the numbers on that too. “What we find is that if you reduce the sunlight by about 2 percent—which is about what we figure it takes to compensate for a doubling in CO2 concentrations from the pre-industrial era—then it turns out your photosynthesis goes down by about 2 percent. The relationship is more or less linear. But CO2 has a fertilizer effect which is thought to be much greater than 2 percent, so the net result is that with more CO2 plus geoengineering, you get more photosynthesis than you would get otherwise. Without the fertilization effect, you get more or less a linear reduction … except not even that is true.” After Mount Pinatubo, he says, evidence shows that the biosphere took up more carbon, probably because the more diffuse light filtering through Pinatubo’s dust veil penetrated deeper into the forest canopy than usual. “Of course, that could be disruptive to ecosystems too, but I think at this point, it’s hard to deny that the high-CO2 world with climate engineering would be more like the pre-industrial world than the high-CO2 world without it. It’s not an improvement on the natural world, it’s just an improvement on too much CO2.”

Ultimately, Caldeira says, the natural world is his main concern. “I don’t share the views of people like Al Gore, who talk about global warming as a threat to civilization. I think humans are basically an invasive generalist species and that we already occupy everything from the Equator to the Arctic Circle. Climate change is pretty small relative to those variations. There’ll be some adaptive costs, but we’ll survive. On the other hand, I think there’s a massive extinction unfolding on a time scale of a couple centuries, and I think that if there are geologists in fifty million years, they’ll see us as the cause. But okay, let’s say Gore’s vision turns out to be correct, … if you had a famine, say, in the developed world, then politicians would be called upon to act quickly and what else could they do but geoengineering? And that’s another reason to have a research program, because if it turns out it really is a bad idea… .”

He spoke quickly and freely, but his sentences often trailed off like this before resuming on an unrelated point. I took the opportunity to ask whether he had ever met Gore personally. “I have,” he said laughing. “When I mentioned geoengineering to Gore, he made a cross with his fingers, you know, like ‘Eeee-villll.’ He exorcised me.”

“I’m not a fan,” Inez Fung said flatly when I asked her opinion of stratospheric aerosol injection. A petite woman in a pageboy haircut, Fung is a prominent figure in climatology—professor of atmospheric science at the University of California, Berkeley, co-director of the Berkeley Institute of the Environment, and, like Caldeira, a contributor to the IPCC, which, in 2008, shared the Nobel Peace Prize with Al Gore. I met her in her office last April in Berkeley’s McCone Hall. She sat in her office chair, arms folded tightly across her chest, one foot bobbing impatiently under the desk.

“You don’t think it could serve as an insurance policy in the event of abrupt climate change?” I asked.

“Heavens, no. What abrupt climate changes are you worried about?”

“Well, what if the Greenland ice sheet starts to go far faster than anticipated? Would it make sense then to decrease temperatures in the hopes of stopping that process, since we’re talking about something like a twenty-foot rise in sea level? Would the ramifications be so serious that we could justify . . . ?”

“I think a much more important thing is to say sea level will rise,” she answered. “Preventing it is just delaying it. I mean all this geoengineering is just delaying . . . It’s gonna go sooner or later.”

“But wouldn’t it at least buy us time?”

“What are we doing in the interim if the whole strategy is to buy time? If we just continue to squander energy, I wouldn’t support it.”

Her foot bounced under her desk.

“Geoengineering is not science fiction, okay?” she continued. “How do we test it? How do we know that it would work? The scientist’s responsibility is not just to propose wild ideas. The scientist’s responsibility is to say, ‘How do we test them?’”

Her frustration seemed to grow as our interview progressed, and when I finally gathered my things to leave, I thanked her for putting up with my questions. She said, “No, no, it’s important not to just look at what is the last resort and ignore responsible action. It’s very American to want a quick fix, but the energy problem is the principal challenge for humankind—the two energy problems: not just the squandering of energy, but also the imbalance in energy access in the world.”

“Sure,” I answered, “but recognizing that doesn’t do anything to solve the problem of climate change.”

She took a deep breath before answering wearily. “There is no solving the problem. There is no solving the problem. All it is is slowing the symptoms.”

A couple weeks after I spoke with Dr. Fung, I drove back to Stanford to see Ken Caldiera again, mindful that the carbon dioxide emanating from my tailpipe would persist in the atmosphere for hundreds, even thousands of years. As one geologist has expressed it, the lifetime of fossil carbon in the atmosphere can effectively be thought of as “300 years, plus 25% that lasts forever.” By contrast, the lifetime of sulfates in the stratosphere is a few years at most. In terms of deploying aerosol injections, this is both blessing and curse. On the one hand, it means that any experimental deployment could be halted quickly if unforeseen consequences arose, and the effects would be short-lived. On the other hand, once a decision was made to continue, there would be no turning back—not for a thousand years, at least. As CO2 concentrations continued to climb, any interruption in maintenance of the atmospheric dust veil would lead to very rapid warming and a potentially crippling shock to the planet. This was just one of the objections detailed by a scientist named Alan Robock in an article called “20 Reasons Why Geoengineering May Be a Bad Idea,” published in the Bulletin of the Atomic Scientists.

I asked Caldeira about Robock’s article as we sat in his office. “A lot of Alan’s objections are really objections to CO2 emissions—that, oh, for example, geoengineering doesn’t cure ocean acidification. Well, we knew that. Nobody ever claimed it did. But geoengineering’s not causing that. It’s CO2 emissions. And then I think another category of his objections is just that it’s not perfect. When the Bulletin asked me to write a response, I started with the Churchill quote about democracy being the worst system of government, except for all the others. I mean, we all know geoengineering’s bad, now tell me what’s better—something you can actually do. Of course, the danger of me talking like this is that, if we really are at some kind of tipping point in terms of actually solving the energy problem, then maybe talking about this stuff is enough to prevent a consensus from forming on reducing emissions.”

“And is that something you worry about?” I asked.

“I do,” he said, then studied the floor for a moment before continuing. “It’s one of the things I worry about most, this idea that by decreasing the downside risk of warming, you remove the impetus to reduce emissions. It’s just like if you have flood insurance, you’re more likely to live in the floodplain. On the other hand, CO2 emissions have been going up more rapidly than in any of the IPCC scenarios, and the concentrations are growing faster than ever. Europe’s emissions went up 1 percent last year despite their supposed commitments to reducing them. And so, if the people who are committed to reducing emissions still have their emissions going up and the rest of the world hasn’t even committed to it, then what remains is just this basic game theoretic problem of how do you get the whole world to sacrifice now, locally, for the broader common good where most of that benefit will accrue decades into the future?”

The question hung in the air until, suddenly, Caldeira brightened. “But anyway,” he said, “we have to make a decision. Where do we want to go for lunch? It’s basically two choices here: burritos or salad.”

I reminded him we had burritos the last time I visited.

“Okay, then let’s go to the salad place just so we don’t get too redundant.”

When I told him we didn’t have to mix it up on my account, he said, “No, no, it’s good, we just need a reason. It’s like Buridan’s Ass. You know the story of Buridan’s Ass?”

I confessed I didn’t.

“No? Well, it’s this philosophy problem where you have an ass that’s equidistant from two piles of food, and he starves to death in the middle.” I pondered that image for a moment before asking him to elaborate on what he called the “game theoretic problem.”

He began by explaining that most economists think it would cost around 2 percent of GDP to stabilize carbon dioxide concentrations, then proposed a thought experiment. “Let’s imagine we already had a zero-emissions economy and let’s say you went to people and said, ‘Okay, we can increase our income by 2 percent and, by the way, all we have to do is acidify the oceans, and heat up the planet, and melt the icecaps, and get rid of polar bears’ … I don’t think most rational people would go for that. And so, that tells me the cost isn’t really the issue, other than the cost being too high for any one actor to pay. You’re basically stuck in the Tragedy of the Commons. And the main thing about geoengineering is that it’s cheap enough that it doesn’t require everyone’s cooperation. And it’s something you can actually do. I mean, it’s easy for everybody to sit around the table and agree that it would be better for all of us to stop emitting CO2. Then we’ll all drive home, right? Agreeing on it doesn’t make it happen. And you can say well, geoengineering’s not as good as emissions reduction in many demonstrable ways. True. But then the question becomes: Is it better than nothing? And I think there’s at least a chance that it’s better than nothing.”

After a long pause, he added, “But if I had to say, just simply on game theoretic grounds, what’s most likely to happen? I think the most likely thing is we don’t do anything. We don’t reduce emissions, and we don’t do geoengineering.” We were silent for a second, then suddenly both of us launched into a kind of half-crazed laughter. It didn’t last long, but for a few moments, the two of us stood there in hysterics, braying like donkeys.


This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.

Recommended Reading