麻豆传媒

PAT MICHAELS, a belligerent sceptic about global warming, is in ebullient mood. 鈥淭he truth is that what we sceptics say is always pilloried by the climate modellers, and then adopted as their own five years later. That would make a good theme for your article.鈥�

Michaels, a climatologist at the University of Virginia, is convinced that the tide is about to turn in his favour, and that the efforts of the mainstream climate modellers, stalwarts of the UN鈥檚 Intergovernmental Panel on Climate Change (IPCC), are about to collapse under the weight of their own inconsistencies. 鈥淭hey can鈥檛 go on forever tinkering with their models, trying to make them fit reality. Ever heard of Ockham鈥檚 razor? It says the simple explanation is usually the best. Apply that in this case and you conclude that the climate just is not as sensitive to the greenhouse effect as they predicted.鈥�

So what exactly is the problem for the climate modellers? Well, something strange has happened to global warming. For almost twenty years, while temperatures at ground level round the world have continued to rise inexorably, the warming has failed to penetrate the atmosphere. In wide areas some three kilometres above Earth, the atmosphere has actually been cooling. This is not what is predicted by the computerised climate models on which all estimates of global warming depend. They all say the warming should spread right through the troposphere, the bottom ten kilometres or so of the atmosphere.

Global warming sceptics have spent almost a decade challenging some of the basic tenets of the climate models. Ever since global warming became headline news in the late 1980s, they have been complaining that the prevailing view is skewed and overstates the problem. Their prime motivation seems to be indignation, coupled with a maverick instinct to buck the latest fashion. But they have also managed to secure some lucrative lecturing fees and consultancy deals with commercial concerns 鈥� such as the coal industry 鈥� who are anxious to undermine international efforts to control emissions of greenhouse gases such as CO₂.

It is a year this month since the US鈥檚 under-secretary of state Timothy Wirth castigated the sceptics at a UN climate conference. They were 鈥渂ent on belittling, attacking and obfuscating climate change science鈥�. But the stubborn failure of most of the troposphere to warm continues to hearten the greenhouse outlaws. Are they right? Has the climatic apocalypse been postponed? Are the modellers on the run from reality?

On a tour of some of the main players, you dodge a constant crossfire of personal and professional abuse. In his home near Boston, the sceptics鈥� guru Dick Lindzen, a meteorologist from the Massachusetts Institute of Technology, accused the IPCC of being dominated by 鈥済uys from the bottom of the heap, such as geographers鈥�.

Back in Virginia, Michaels slams the IPCC scientists for manipulating data, then settles into an explanation of how the key to Vice-President Al Gore鈥檚 challenge for the presidency next time round will be corrupt environmental journalists who he will use to peddle a fraudulent version of climate change.

In the greenhouse wars, the battle is bloody and many a good scientist has been conscripted by both sides. Take John Christy. For ten years he was a Baptist minister in Kenya before he took up science. Now, as professor of atmospheric science at the Global Hydrology and Climate Center, part of the University of Alabama in Huntsville, he catalogues data from satellite instruments operated by the US government鈥檚 National Oceanic and Atmospheric Administration (NOAA).

Since 1979, long before global warming was an issue, NOAA鈥檚 instruments have been measuring microwave radiation released by the atmosphere. The radiation comes largely from oxygen molecules, which release more as they warm, giving an overall picture of temperatures in the troposphere. So far, and in stark contrast with the ground-based meteorological stations, the satellites have apparently picked up little evidence of warming.

Naturally, the sceptics have adopted Christy鈥檚 data to suggest that global warming is a myth. In response, collectors of ground-based data have hit back. Jim Hansen, the NASA climate modeller who first put global warming onto the front pages back in 1988, claims that if the satellites can鈥檛 see evidence of warming 鈥渢here鈥檚 something wrong with their data鈥�.

Both are wrong, says an exasperated Christy. First, his short time-series 鈥� just 18 years 鈥� is skewed by warming at the start, caused by a short-term natural warming in the Pacific, and by cooling towards the end, from the eruption of Mount Pinatubo in the Philippines. Filter out those effects and he reports 鈥渁 very slight warming trend鈥�. But most importantly, says Christy, 鈥渢he satellites don鈥檛 measure surface temperatures, but average temperatures through the troposphere鈥�. Thus the surface and satellite data complement rather than contradict. Put them together, he says, and they show that the surface of the planet is warming, but the bulk of the troposphere, the so-called free troposphere, is not.

Ups and downs

Add archive data from weather balloons recently assembled at the British Meteorological Office by David Parker, another assiduous and independent-minded data-cruncher, and something even more surprising emerges. From the early 1960s to the late 1970s, there was strong warming in the free troposphere, at a time when the surface was cooling slightly. Then everything switched: the surface warmed strongly while at 1.5 kilometres and upward, temperatures were at least stable, if not falling.

Nobody can explain this reversal. Certainly, the temperature record of the past three decades does not match the predictions of the models, agrees Parker. In the models, the surface and the free troposphere are very closely coupled, so their temperatures should move together. But if this is right, why are the temperatures in these two regions moving in different directions? 鈥淭he surface and mid-troposphere appear to be much less coupled than the models assume,鈥� says Parker. 鈥淚f the models don鈥檛 get tropospheric heating right, we are in trouble.鈥�

After years of trying, the greenhouse sceptics finally feel that here they have found the Achilles heel in the climate models. If the models are wrong about how surface warming influences temperatures in the troposphere, they are also likely to be wrong about another fundamental feature: the movement of water vapour between the surface and the free troposphere. And that, argue the sceptics, means the models may have misrepresented, or even have invented, one of the vital mechanisms behind global warming itself: the positive water-vapour feedback.

Feedbacks are what turn the greenhouse effect from a benign curiosity into a potential apocalypse. Even sceptics agree that putting more greenhouse gases, such as CO₂ from burning fossil fuels, into the atmosphere will tend to warm the planet. But even the doubling of CO₂predicted for the late 21st century would only add about 1 掳C to global temperatures. According to the models, however, this initial warming will be magnified by a series of positive feedbacks. In the most important of these, they predict that surface warming will increase evaporation from the oceans and push more water vapour into the atmosphere. Because water vapour is itself a strong greenhouse gas, this would amplify the effect of the CO₂ 鈥� a positive feedback that might roughly double global warming.

But will it? Christy鈥檚 collaborator on the satellite data, Roy Spencer of NASA鈥檚 Marshall Space Flight Center, is doubtful. 鈥淚 don鈥檛 think warming will be as big as people think,鈥� he says. 鈥淭he positive feedback theory assumes that, in practice, a warmer troposphere will actually hold more water vapour.鈥� But he points out that if, as the satellite data suggest, the free troposphere is largely cut off from the surface, water that evaporates from the oceans will not necessarily mean more water vapour in the free troposphere.

This matters, according to Spencer, because water molecules in the troposphere could have a much bigger warming effect than ones that stay close to the surface. Higher up, the air is extremely dry, and Spencer argues that because of this, adding or subtracting relatively small amounts of water there could greatly alter the amount of heat trapped.

Not everybody agrees about the importance of the upper troposphere in driving water vapour feedbacks. One of Britain鈥檚 leading meteorologists Keith Shine, from the University of Reading, says: 鈥淭he surface is at least as important here. Sure, a single molecule is more important aloft, but there are so many more molecules at the surface.鈥�

But if Spencer is right, this would seem to pose a serious problem for the conventional view that water vapour would amplify the effect of CO₂. The extra water vapour generated by warming at the surface would never make it to the regions where it could significantly increase the greenhouse effect.

Clouding the issues

And there is more. Some sceptics also argue that the complex physics of clouds could actually reduce the amount of water vapour that reaches the free troposphere 鈥� and so damp down global warming. The modellers are on weak ground in this discussion because individual clouds are too small to be modelled in any detail inside global climate models, which assume that their dynamics will be unchanged by warming.

The sceptical position here is that warming might make clouds more efficient at producing rain, leaving less water vapour behind to moisten the free troposphere. It works like this: most of the water vapour that evaporates from the oceans does not stay in the air for long. It forms clouds and quickly returns to the surface as rain. But some re-evaporates from clouds and heads off into the clear air of the free troposphere. Change the amount of water that leaves the clouds as rain, says Spencer, and you change the amount of water vapour left to re-evaporate from the clouds, and hence the amount of water in the free troposphere. ( See Diagram.

)

This is where Lindzen comes in. Though he is highly regarded for his innate ability, Lindzen鈥檚 ideas are notoriously difficult to pin down. Colleagues complain that too few of them turn into peer-reviewed papers and too many emerge as invective in newspaper articles. Shine says: 鈥淟indzen is painfully clever. But he keeps changing his arguments.鈥� Even so, John Houghton, co-chair of the IPCC鈥檚 science panel, once described him as the 鈥渕ost serious鈥� of his sceptic foes.

Lindzen argues vehemently that water vapour operates a negative rather than a positive feedback. He says 鈥渁ll the data show鈥� that if clouds are warmer, they will turn a greater proportion of their moisture into rain. Result: less water re-evaporating from clouds, a drier free troposphere and a negative water-vapour feedback. In other words, changes in the tropospheric water vapour would compete with, not reinforce global warming caused by CO₂.

Here, even some fellow sceptics back off. 鈥淚t鈥檚 intriguing, but it鈥檚 a theory, that鈥檚 all,鈥� says Spencer. 鈥淭here is no actual evidence for a negative feedback.鈥� Shine agrees. 鈥淭he idea that precipitation efficiency is better in warm clouds is at best contentious. I think what is alarming us more about Lindzen鈥檚 comments is his assertion that the answer is known.鈥�

But Lindzen鈥檚 opponents know that he is attacking them at a weak point. Evidence of a positive feedback is not overwhelming, either. The IPCC鈥檚 most recent review, its Second Assessment published in 1996, admits to serious gaps in knowledge about water vapour and concedes that feedback 鈥渞emains a substantial uncertainty in climate models鈥�.

Simon Tett, one of the Met Office鈥檚 top modellers and a leading IPCC author, agrees that recent studies suggest the positive feedback may have been overestimated. 鈥淚 believe the upper troposphere is probably drier than the models suggest,鈥� he says. Though the evidence for a negative feedback is, if anything, even weaker, the sceptics still believe things are moving their way.

Lindzen for one argues that if the models get the detail wrong, they will get the big picture wrong, too. But modellers say the proof of the pudding is in the eating. 鈥淲e think there is good evidence that our models reproduce past climate change reasonably well,鈥� says Tett. 鈥淭hat is good evidence that they are basically correct.鈥�

But here too the modellers are being challenged. 鈥淭he simply fact,鈥� says Michaels, 鈥渋s that even at the surface the world is not warming up as much as the modellers say it should.鈥� Hansen agrees: 鈥淢odels driven by greenhouse gases alone give warming about twice as large as observed over the past 150 years 鈥� about 1 掳C rather than the observed 0.5 掳C to 0.6 掳C.鈥� For recent years, says Christy, 鈥渢he models suggest a warming three times what we see鈥�.

Michaels was one of the first scientists to propose an explanation for this. In an article he wrote for 麻豆传媒in the early 1990s, he suggested that sulphate smogs emanating from industrial areas were casting a thin pall over sufficiently large areas to mask much of the warming (鈥淕lobal pollution鈥檚 silver lining鈥�, 23 November 1991, p 40). By late 1994, modellers at the Met Office had adopted the idea and claimed that combining predictions of warming from CO₂ and cooling from sulphate produced a good fit of actual climate change 鈥� a fit that persuaded John Gummer, Britain鈥檚 environment secretary of the day, to pronounce himself convinced that global warming was for real.

Yet today Michaels is a vociferous opponent of this theory. The problem is geography. Sulphate only survives a few days in the air. Since most sulphate is emitted in the northern hemisphere, its cooling influence should be largely limited to that hemisphere. So if sulphate cooling is important, Michaels says, the southern hemisphere should be warming faster than the north. Until the late 1980s, that was what was happening. But since 1987, warming has virtually ceased in the southern hemisphere 鈥� notably in the mid-latitude region between 1 and 1.5 kilometres above the surface, where warming was previously most intense 鈥� while it has surged ahead in the north.

For Michaels, that condemns the sulphate theory to the dustbin. And he scorns modellers who don鈥檛 follow his lead. One of Michaels鈥� most violent verbal attacks on fellow scientists is against Ben Santer of the US government鈥檚 Lawrence Livermore National Laboratory in California. In a paper in Naturelast summer, Michaels implied that Santer bolstered a claim that sulphates were cooling the planet by arbitrarily ending his analysis of temperature trends in 1987, the year the southern warming ceased. Santer鈥檚 data finished in 1987, but Michaels argues that there are other datasets that Santer could have used to extend his findings. In a subsequent letter to Nature, Michaels argued that the models and reality were diverging. 鈥淪uch a result鈥� cannot be considered a `fingerprint鈥� of greenhouse-induced climate change,鈥� he claimed.

And that charge drew blood. For Santer was the main author of a key chapter in the IPCC鈥檚 Second Assessment which concluded, largely on the basis of this work, that there was 鈥渁n emerging pattern of climate response to鈥� greenhouse gases and sulphate aerosols in the climate record鈥�. In other words, the human fingerprint could now be seen in climate change.

But is it that simple? Santer, who admits to being 鈥渢roubled鈥� by Michaels鈥� assault on his integrity, responds vigorously. He denies selective use of the data and says that the datasets available when he wrote his paper for the years after 1987 were not compatible with his own data. He says that the relative cooling of the southern hemisphere since 1987 does not contradict the models. In fact, the models explicitly predict it. His case is that what we are seeing is the interaction between two different effects happening on two different timescales.

Blowing hot and cold

Basically, the first effect, global warming, is bound to happen more slowly in the southern hemisphere than the north. This is because most of the southern hemisphere consists of oceans, which heat up more slowly than the landmasses which dominate the north. But the picture has been confused by the second effect, sulphate cooling, which peaked in the north in the mid-20th century. It slowed warming in the northern hemisphere so much that the southern hemisphere, oceans and all, raced ahead. But since 1987, the growing force of the greenhouse effect has reasserted itself, and the north has again taken the lead. 鈥淭he contention by Michaels that model predictions and observed data differ fundamentally is simply incorrect,鈥� says Santer.

Michaels dismisses this. He says the data actually show that the southern hemisphere is not simply warming more slowly, as Santer maintains. It has actually been cooling since 1987, which no models predict. This, he says, makes a nonsense of both the theory of sulphate cooling and the models of global warming 鈥� a double-whammy that he relishes.

And, if anything, the more recent raw data quoted by both sides tends to bear Michaels out: the past five years was more than 0.2 掳C cooler on average than the previous five. But, says Tett, 鈥渢hat鈥檚 probably just natural variability. You can鈥檛 dismiss a 30-year warming trend on the basis of a blip at the end.鈥� Again, the jury is out.

Meanwhile, the argument is moving on. In a paper published in Sciencein November last year, Tett added ozone depletion in the stratosphere to the equation. It has been clear for some years that ozone depletion in the stratosphere is causing cooling there. More recently, modellers have suggested that this cooling is being transmitted to the upper levels of the troposphere as well. 鈥淏etween 1960 and 1995, adding ozone depletion dramatically improves the fit between reality and the model,鈥� Tett said. This is compelling, in particular because it helps explain why the upper troposphere has failed to warm as the models first predicted.

Michaels thinks there might be something in this. But he scoffs that constant tinkering with the climate models to make them fit reality is deeply unscientific. 鈥淚t鈥檚 like a doctor who prescribes an aspirin for a headache and then claims the body鈥檚 problem was a lack of aspirin. They don鈥檛 have a proper diagnosis.鈥� There is a clash of scientific cultures here.

Stripped of the polemic, Michaels analyses the two possible explanations for the slowness of global warming to date. Either, as the modellers say, the warming is being masked by something else, such as sulphate or ozone depletion. In which case, the mask must eventually slip as the greenhouse effect intensifies. Or, as the sceptics say, the climate is simply less sensitive to the warming effect of greenhouse gases than the models predict. And the positive feedbacks are largely illusory.

The modellers鈥� belief that they can create the future in climate simulations are undoubtedly shaken by the constant revisions to the models. The problem for the sceptics, however, is that they still lack a coherent story about how the atmosphere is working. And whenever they can find any uncertainty in the way that the atmosphere works, they tend to use this to claim that there will be no problem with greenhouse warming.

As Shine says of Lindzen: 鈥淗e always falls back on uncertainty. Sure there is uncertainty, but he then claims that all the uncertainty will work in his direction. Why should it?鈥�

Parker admits that 鈥渢here are a lot of things we don鈥檛 know鈥�. But, he adds, that doesn鈥檛 disprove global warming, or the models. 鈥淪ceptics tend to elevate one element in a complex system above all the others. You cannot do that, however clever you are. You have to integrate every influence to find out what they might mean when all acting together. And the models are the only way of doing that.鈥�

Is there any common ground? Of all people, Michaels insists there could be. 鈥淲hen it comes to it, the modellers and the sceptics are not so far apart,鈥� he says. Indeed, if pressed, Michaels, Lindzen, Spencer and other sceptics suggest a doubling of CO₂ in the atmosphere would raise average temperatures by between 1 and 1.5 掳C. And 1.5 掳C is the bottom end of the modellers鈥� range of predictions.

But Michaels has, as ever, a twist. 鈥淵ou can鈥檛 make a case for a global apocalypse out of a 1.5 掳C warming. It destroys the issue. If politics weren鈥檛 driving this we could all meet on common ground.鈥�

But, of course, he thinks the politics is all on the other side.