Uncovering flaws in the climate models
The graph flashed up on the screen for only a few seconds, but it set alarm bells ringing. Had I read it right? The occasion was a workshop on climate change at the Hadley Centre for Climate Prediction, held in Exeter in mid-2004. The room was packed with climate modelers from around the world. Even they raised a collective eyebrow when the graph sank in. If carbon dioxide in the atmosphere doubled from its pre-industrial levels, the graph suggested, global warming would rise far above the widely accepted prediction of 2.7 to 8.1°F. The real warming could be 18°F or even higher. Surely some mistake? Too much wine at lunch? No. This was for real.
Till now, climate modelers have graphed the likely effect of doubling carbon dioxide levels using what is known in the trade as a bell graph: the best estimate—about 50—falls in the middle, and probabilities fall symmetrically on either side. So the chance that the real warming will be 8. 1°, for instance, is the same as that it will be 2.70. But the graph of likely warming that James Murphy, of the Hadley Centre, was displaying on an overhead screen that morning looked very different. The middle point of the prediction was much the same as everybody else's. But rather than being bell-shaped, the graph was highly skewed, with a long "tail" at the top end of the temperature range. It showed a very real chance that warming from a doubling of carbon dioxide would reach 10, 14, 18, or even 21°F.
Carbon dioxide is widely expected to reach double its pre-industrial levels within a century if we carry on burning coal and oil in what economists call a business-as-usual scenario. But nobody has seriously tried to work out what 18 degrees of extra warming would mean for the planet or for human civilization. It would certainly be cataclysmic.
Let's be clear. Murphy was not making a firm prediction of climatic Armageddon. But neither was this a Hollywood movie. The high temperatures on the display, he said, "may not be the most likely, but they cannot be discounted." Nor was Murphy alone with his tail. The meeting also saw a projection by David Stainforth, of Oxford University, that suggested a plausible warming of 21°F. Six months later, this new generation of scarily skewed distributions started turning up in the scientific journals. Unless the editors take fright, these figures will probably become part of the official wisdom, incorporated into the next report of the IPCC.
So what is going on? For one thing, modelers have for the first time been systematically checking their models for the full range of uncertainty about the sensitivity of the climate system to feedbacks that might be triggered by greenhouse gases. Assessing those efforts for the IPCC was the main task of the Exeter meeting. And what has emerged very strongly is that clouds, which have always been seen as one of the weakest links in the models, are even more of a wild card than anyone had imagined. The old presumption that clouds will not change very much as the world warms is being turned on its head. There may be more clouds. Or fewer. And their climatic impact could alter. It is far from clear whether more clouds would damp down the greenhouse effect, as previously thought, or intensify it. Being mostly of an age to remember 1970s Joni Mitchell songs, the climate scientists in Exeter mused over coffee that they had "looked at clouds from both sides now." And they didn't like what they saw.
An assessment of the sensitivity of global temperatures to outside forcing —whether changes in sunlight or the addition of greenhouse gases—has been central to climate modeling ever since Svante Arrhenius began his calculations back in the 1890s. This assessment mostly revolves around disentangling the main feedbacks.
The three biggest feedbacks in the climate models are ice, water vapor, and clouds. We have already looked at the effect of melting ice on the planet's albedo. It explains why the Arctic is warming faster than elsewhere and giving an extra push to global warming. Water vapor, like carbon dioxide, is a potent greenhouse gas, without which our planet would freeze. The story of what will happen to water vapor is a little less clear cut. A warmer world will certainly evaporate more water from soils and oceans, and this process is already increasing the amount of water vapor in the atmosphere, amplifying warming. In the standard climate models, extra water vapor in the air at least doubles the direct warming effect of carbon dioxide. But it's when we come to clouds that the calculations get sticky.
A lot of water vapor in the air eventually forms clouds. At first guess, you might say that clouds would have the opposite effect of water vapor, shading us from the sun's rays and keeping air temperatures down. They do that on a summer's day, of course. But at night they generally keep us warm, acting like a blanket that traps heat. Globally, these two effects—or, rather, their absence—are most pronounced in deserts. Where there are no clouds, the days are boiling, but the nights can get extremely cold, even in the tropics.
The temperature effects of clouds turn out also to depend on the nature of the clouds. Their height, depth, color, and density can be vital, because different clouds have different optical properties. The wispy cirrus clouds that form in the upper atmosphere heat the air beneath, because they are good at absorbing the sun's rays and re-radiating the heat downward, whereas the low, flat stratus clouds of a dreary summer's day are good at keeping the air below cool.
Researchers still know surprisingly little about how many and what sort of clouds are above our heads. For instance, it has only recently emerged that there may be many more cirrus clouds than anyone had thought. Many are almost invisible to the naked eye, but nonetheless seem to be highly effective at trapping heat. Some studies suggest that, taken globally, the cooling and warming effects of clouds currently largely cancel each other out, with perhaps a slight overall cooling effect. But nobody is sure. And even small changes in cloudiness could affect planetary albedo substantially. If a warmer world tipped clouds into causing greater warming, the effects could be considerable.
So what is the prognosis? Again, a first guess is that extra evaporation will make more clouds, because a lot of the water vapor will eventually turn into cloud droplets. But even that may not be so simple. Evaporation doesn't just lift water vapor into the air to create more clouds; it also burns off clouds, leaving behind blue skies. And greater evaporation can also make clouds form faster, so that they fill with moisture faster, make raindrops faster, and dissipate faster. So, in a greenhouse world, fluffy cumulus clouds that we are used to seeing scudding across the sky all day could instead boil up into dark cumulonimbus clouds and rain out, leaving behind more blue skies.
Bruce Wielicki has been trying to figure out the answer to such questions during more than twenty years of cloud-watching at NASA's Langley Research Center, in Hampton, Virginia. He says that satellite data suggest that clouds probably still have an overall cooling effect on the planet; but, especially in the tropics, there is a trend toward clearer skies. Since the midi 980s, the great tropical convection processes that cause air to rise where the sun is at its fiercest have intensified. As a result, storm clouds are forming and growing more quickly in those areas. This may be increasing the intensity of hurricanes across the tropics. Less obvious is Wielicki's discovery that the storm clouds not only form more quickly but also rain out more quickly. That leaves the tropics drier and less cloudy as a whole.
Many researchers see the phenomenon as strong evidence of an unexpected positive feedback to global warming. But Wielicki is cautious about what is behind his discovery of clearer tropical skies. We need to know, because the tropics are where an estimated two thirds of the moisture in the atmosphere evaporates—an important element in the planet's thermostat. "Since clouds are thought to be the weakest link in predicting future climate change, these new results are unsettling—the models may be more uncertain than we had thought," says Wielicki. His own guess is that clouds may be two to four times more important in controlling global temperatures than previously thought.
And that takes us back to the graphs on display in Exeter, where Murphy and Stainforth reached much the same conclusion as Wielicki in their new modeling projections of possible future warming. To make his graph,
Murphy took a standard climate model and tweaked it to reflect the new range of uncertainties for cloud cover, lifetime, and thickness. His model responded by delivering much higher probabilities of greater-than- expected warming. "Variations in cloud feedback played a major role in the predictions of higher temperatures," he said. Susan Solomon, who as chair of the IPCC's science working group will be the final arbiter of what goes into its 2007 assessment of climate change, concurs. The biggest difference between models that give high estimates of global warming and those that give lower ones, she says, is how they handle cloud feedbacks.
Who is right? Are fears about a strong positive feedback from clouds warranted or not? One way of finding out is to test how the different models reflect the real world today. The IPCC is currently using this approach more widely to help weed out poor models from its analysis. Murphy has no doubt about what the outcome will be. The models that predict low warming "have a lot of unrealistic representations of clouds," he says. "The weeding process suggests higher temperatures." That is not proof, but it is worrying.
Clouds are not the only thing changing the reflectivity of Earth's atmosphere. Planet Earth is becoming hazier; the wild blue yonder is not so blue. The problem is pollution spreading across the Northern Hemisphere and much of Asia, blotting out the sun. The issue is not just aesthetic. Nor is it just medical, though millions of people die from the toxic effects of this pollution every year. It is also climatic. While some parts of the world are seeing temperatures soar, some of the world's most densely populated countries have seen temperatures drop. Climatologists who have spent many years warning about global warming are reaching the conclusion that we may need to be at least as concerned about the effects of this localized cooling.
The pollutants of concern here are normally lumped together under the name aerosols, but they are of many types and come from many sources. The culprits include operators of power stations in Europe, farmers burning crop stubble in Africa and trees in the Amazon, steel manufacturers in India, and millions of women cooking dinner over millions of open cooking stoves across the tropics. Most of these activities produce greenhouse gases, but they also produce aerosols in the form of smoke, soot, dust, smuts of half-burned vegetation, and much tinier but highly reflective sulfate particles. Depending on their characteristics, these aerosols reflect or absorb solar radiation. In fact, most do both, in varying quantities. But with one important exception that we shall return to, the dominant effect is cooling. The result is that some parts of the planet, from central Europe to the plains of India and the Amazon jungle to eastern China, have missed out on global warming either permanently or at certain times of the year.
A global cooling to counteract global warming might seem a good idea. Sadly, things are not so simple. The competing forces are pulling the climate system in two different directions that may not so much counteract as inflame each other. Certainly they introduce a new element of uncertainty in atmospheric processes. But although many countries are trying to reduce their emissions of smog-making aerosols, for excellent public-health reasons, the cleanup will lift the "parasol of pollution" over those countries. The likely result will be a burst of warming that could happen within days of the pollution's clearing.
We can see evidence of this already in central Europe. Fifteen years ago, countries like Poland, Czechoslovakia, and East Germany reeked with the smell of burning fossil fuels from the old Soviet-style heavy industries. Chimneys belched, and smog was endemic. The region where the three countries met became known as the "black triangle." The pollution was having a local cooling effect more than twice as great as the warming effect of greenhouse gases. Since the fall of the Berlin Wall, the old polluting industries have mostly shut down, and the air has cleared. More sun penetrates the smog-filled landscape, and central Europe has warmed correspondingly. In the past fifteen years, temperatures there have risen at three times the global average rate.
This real-world experiment shows clearly the power of aerosols to cool Earth's surface. And it raises another question for the future: How much warming is being suppressed globally by aerosols? "We are dealing with a coiled spring, with temperatures being held back by aerosols," says Susan Solomon, chief scientist for the IPCC. "If you shut off aerosols today, temperatures would increase rapidly, but we don't yet know exactly how much, because we don't know how coiled the spring is." The best guess until recently was that aerosols were holding back a quarter of the warming, or about o.36°F. In other words, a greenhouse warming of 1.4 degrees since pre-industrial times has been reduced by aerosols to a current warming of 1 degree. But critics say this calculation is little more than a guess, and the first efforts at a more direct measurement of radiation changes caused by aerosols suggest that the spring may be much more tightly coiled.
I was present at one of the first meetings where these ideas were discussed in detail. The occasion was a workshop of climate scientists held at Dahlem, a quiet suburb of Berlin, in 2003. The meeting was discussing "earth system analysis," and the man who brought the issue to the table was the distinguished Dutch atmospheric chemist Paul Crutzen, whose brilliant and creative mind first divined many of the secrets of chemical destruction of the ozone layer. Back in the 1980s, Crutzen had stumbled on the notion that during a nuclear war, so many fires would be burning that the smoke "would make it dark in the daytime" and "temperatures would crash." That insight has led to continued analysis of the role of everyday aerosols in climate and to his conclusion, argued in Dahlem, that aerosols could be disguising not a quarter but a half to three quarters of the present greenhouse effect. "They are giving us a false sense of security," he said. Past calculations of the cooling effect of aerosols, he said, had been inferred by comparing the warming predicted by climate models with actual warming. The aerosol cooling effect was reckoned as the warming that had "gone missing." But as the modeler Stephen Schwartz, of the Brook-haven National Laboratory, put it on another occasion, "that approach assumes that we know that the climate model is accurate, which of course is what needs to be tested."
After dinner in Dahlem—over a few Heinekens, as I remember—Peter Cox, a hard-thinking, hard-drinking climate modeler then at the Met Office in England, did some back-of-the-coaster calculations about what Crutzen's conjecture might mean for future climate. He became rather absorbed. A couple of bottles later, he had come to the conclusion that, if Crutzen was right, the true warming effect of doubling carbon dioxide could be more than twice as high as existing estimates, at 12 to 1 8°F. The following morning, his more sober colleagues registered agreement. I went home and wrote a story for New Scientist, quoting Cox's numbers and the workshop's conclusion that the findings had "dramatic consequences for estimates of future climate change." I was rather excited by it, but the story decidedly failed to interest the rest of the world.
Later Cox, his Hadley Centre colleague Chris Jones, and Meinrat Andreae, of the Max Planck Institute for Chemistry, in Mainz, tested the guesses in more detail, and reached the same conclusions that Cox had on his coaster. They did it by running climate models that assumed either a low greenhouse warming moderated by a small cooling from aerosols, or a bigger greenhouse warming held back by a bigger aerosol cooling. They reported in Nature that the "best fit" involved a warming from doubling greenhouse gases that, without the moderating effect of aerosols, would be "in excess of" 10.8 degrees and "may be as high as" 18 degrees.
"Such an enormous increase in temperatures would be greater than the temperature changes from the previous ice age to the present," wrote the three researchers. "It is so far outside the range covered by our experience and scientific understanding that we cannot with any confidence predict the consequences for the Earth."
Still the world didn't take much notice. I asked Andreae about this strange indifference. "It's always amazing," he e-mailed me, "how many people don't see how important this issue is for the future development of the climate system." The discussion at the Dahlem meeting had rather changed his worldview, he said. "Before the Dahlem meeting, I was becoming kind of climate complacent, in the sense that I was convinced of coming global warming, but felt that it was going to be a couple of degrees and we could deal with that. Also, I felt that the aerosols were doing us a favor in slowing and reducing warming. But after it, I came to realize that the aerosols brake will come off global warming, and also that the aerosol cooling introduces a great uncertainty about climate sensitivity. I'm now in a situation where, as a human being, I hope that I'm wrong as a scientist. If we are right with our current assessment, there are really dire times ahead."
Models are only models, of course. But whatever the precise scale of the current aerosol effect, it would be quite wrong to imagine that it can carry on protecting us from the worst as global warming gathers momentum. That is because aerosols and greenhouse gases have very different life spans in the atmosphere. Aerosols stay for only a few days before they are washed to the ground in rain. By contrast, carbon dioxide has a life span of a cen- tury or more. If, for the sake of argument, we stuck with current emission levels of both aerosols and carbon dioxide, the aerosol levels in the air would stay the same. There would be no accumulation and no increase in the cooling effect. But carbon dioxide levels would carry on rising and produce ever greater warming.
Probably. The trouble is that scientific knowledge is, if anything, even poorer about aerosols than it is about the effects of clouds. Says Stephen Schwartz: "There are many different kinds of aerosols, lots of interactions among them, and unknown issues of cloud microphysics—all of which need to be better understood. This is hard science which I am afraid nobody has come to grips with yet." There is no dispute that some aerosols, such as sulfate particles from coal-fired power stations, predominantly scatter sunlight and reflect it back into space. They increase albedo and cool the planet for sure. Others, though, have some scattering effect but also absorb solar radiation and then re-radiate it, warming the ambient atmosphere. And with them it is harder to be sure where the balance between the two effects lies.
Here the biggest concern is soot, the black carbon produced from the incomplete burning of coal, biomass, or diesel. Scientific understanding of the role of soot is, to be frank, all over the place, as a quick scan of the major scientific journals makes clear. In March 2000, a paper in Science said soot was "masking global warming"; eleven months later another, in its chief rival, Nature, said soot was "generating global warming." Ten months later, presentations at a big U.S. conference of the American Geophysical Union called it variously "a cooling agent" and "the biggest cause of global warming after carbon dioxide." These can't both be right.
The truth seems to be this. A cloud of soot—whether from a forest fire, a cooking stove, or an industrial boiler—shields Earth from the sun's rays, thus cooling the ground beneath. But it also absorbs some of that radiation and converts it to heat, which it radiates into the surrounding air. So soot cools the ground but warms the air. The ground doesn't move, but the air does. The cooling effect, though intense, is mostly located near the pollution source; while the warming effect, though less intense, extends much farther.
There is still great uncertainty about the precise role of soot in global climate. Jim Hansen suggests that it could be responsible for up to a quarter of warming over parts of the Northern Hemisphere. He believes that soot may be the third most important man-made contributor to the greenhouse effect, behind carbon dioxide and methane, and that controlling it offers one of the cheapest, most effective, and quickest ways of curbing global warming. Even so, in those parts of the world where it is produced in large quantities, it is undoubtedly cooling the land. Those parts of the world are mainly in Asia. And now there is a new concern. Could aerosol emissions in India and China turn off the Asian monsoon?
Was this article helpful?
Disasters: Why No ones Really 100 Safe. This is common knowledgethat disaster is everywhere. Its in the streets, its inside your campuses, and it can even be found inside your home. The question is not whether we are safe because no one is really THAT secure anymore but whether we can do something to lessen the odds of ever becoming a victim.