Chill in the air

++++++++++++++++++++++++++++++++++++++++++
 In the thin, frigid atmosphere beyond the global greenhouse it's getting colder than ever. And that, says Fred Pearce, spells big trouble down on the ground
 ++++++++++++++++++++++++++++++++++++++++++

THEY CALL IT the "ignoro-sphere" because so little is known about it. Even so, the upper atmosphere is becoming the hot topic in the global warming debate. But not because it is heating up. The uppermost parts of our atmosphere are growing steadily colder and, paradoxically, the greenhouse effect is being held to blame. Until recently, scientists concerned about the greenhouse effect have concentrated on the lowest layer of the atmosphere, the troposphere. This is understandable enough. Although the tropo-sphere is only 12 to 15 kilometres thick, it contains 75 per cent of the mass of the atmosphere. It is where our weather occurs, where our planes largely fly--and where there is growing evidence that the accumulation of greenhouse gases is causing global warming.

Wide-open spaces ~~~~~~~~~~~~~~~~ But in spatial terms, the troposphere is only a small part of the atmosphere, and there is no reason to believe that the greenhouse effect won't be just as important for the layers above the troposphere--the wide-open spaces of the stratosphere, mesosphere and thermosphere. Already, researchers have found that the changes occurring in these higher reaches are far greater than those below. And they are moving the opposite way: cooling rather than warming. Take the stratosphere, the layer immediately above the troposphere. This extends from around 15 kilometres to about 50 kilometres above the Earth's surface and contains the layer of ozone that protects us from the worst of the Sun's ultraviolet rays. Last year researchers warned that if greenhouse gases continue to accumulate in the atmosphere, cooling in the stratosphere could accelerate ozone destruction and yield an Arctic ozone hole as severe as the one over the Antarctic. Meanwhile, we now find that the meso-sphere, between 50 and 90 kilometres up, has been cooling by as much as a degree every year for the past 30 years--ten times faster than anyone had predicted. For some experts, such as Gary Thomas of the Laboratory for Atmospheric and Space Physics at the University of Colorado in Boulder, the cooling of the mesosphere may be the "miner's canary". He believes it is the latest, the biggest and the most unequivocal signal that the global climate really is changing. It's around a decade since climate scientists began to calculate how the greenhouse effect might influence the upper atmosphere. Their logic then was simple: the troposphere is warmed largely by heat radiating from the surface of the planet. As the concentration of greenhouse gases such as carbon dioxide and methane increases in the atmosphere, more of this radiated heat will be trapped close to the ground, warming the troposphere from the bottom up. The more this happens, the less heat will be available for the rest of the atmosphere. The upshot: as the troposphere warms, the upper parts of the atmosphere cool. This effect, known as radiative cooling, would not matter if the heat at the lower levels could gradually diffuse upwards. The problem is, it can't. When warm air rises through the troposphere--in the huge storm clouds of the tropics, for instance--it is forced to a halt at the tropopause, where the troposphere ends and the stratosphere begins. This happens because the ozone in the lower stratosphere is very efficient at absorbing solar heat directly, making the air there warmer than in the upper troposphere. The result is a temperature inversion that blocks the air below in much the same way that a layer of warm air will seal in a city's smog. Warm air rising through the troposphere and entering the stratosphere swiftly finds that it is no longer warmer than the surrounding air. It loses its buoyancy and cannot rise further. So the heat trapped by greenhouse gases in the troposphere cannot be redistributed, and the upper atmosphere cools down. When researchers such as Nobel prizewinner Paul Crutzen of the Max Planck Institute for Chemistry in Mainz, Germany, first included this radiative cooling effect in their atmospheric models a decade ago, it was considered little more than a curiosity. Now it's becoming clear that the cooling has real implications for life on the Earth's surface. In particular, stratospheric cooling contributes to thinning of the ozone layer because the chemical processes that cause ozone to be destroyed most quickly depend critically on temperature. The ozone-eating chemicals from old fridges and air-conditioners do their damage in conjunction with ice particles inside polar stratospheric clouds that form below -80 degrees C. This figure happens to be close to the temperature in winter at the bottom of the Antarctic stratosphere. The Arctic is generally milder, and until now has been spared the kind of severe ozone hole that appears every year in the south. But there are signs that the Arctic is catching up. "Stratospheric climate has changed considerably during the last decades," says Hans-Friederich Graf, a senior scientist at the Max Planck Institute for Meteorology in Hamburg. In particular, the Arctic has grown colder. Researchers had expected some cooling, but until recently they assumed that it would be more than counterbalanced by the reduction in the quantity of ozone-eating chemicals reaching the stratosphere, so that there would be no increased threat to ozone.

Too cool ~~~~~~~~ What has caused alarm is the realisation that the cooling of the stratosphere over the past five years has been greater than predicted, especially over the polar regions in winter. The suspicion is growing that something besides radiative cooling is at work lowering the temperature of the upper atmosphere. So what is happening? One factor is that despite the temperature-inversion lid that the stratosphere puts on the troposphere, the two zones are not entirely cut off from one another. The cold base of the Arctic stratosphere is periodically heated up when slugs of warm, buoyant air from the troposphere break through the barrier. The fear now is that these breakouts have become less frequent. "In the past ten years, there have been only two major warmings in the Arctic during December-February," says Drew Shindell of NASA's Goddard Institute for Space Studies (GISS) in New York. "That is a big change from the 1980s, when there were five." Shindell thinks global warming can be blamed directly for this and for the resulting cooling in the regions where ozone holes form. In April last year, Shindell published a paper in Nature (vol 392, p 589) describing a new analysis of how the atmosphere responds to the greenhouse effect. He suggested that the answer to the extra stratospheric cooling over the Arctic lies in the tropics. Shindell's model predicts that as the world's temperature rises, stronger upward currents of hot air over the warm tropical oceans will create intense heating in the upper troposphere there. This will increase the difference in temperature between the tropics and higher latitudes. Increasing the temperature gradient from tropics to poles, he calculates, will also increase the strength and speed of a strong winter wind, known as the polar night jet, which encircles the Arctic in the lower stratosphere. Strengthening the polar night jet isolates the cold Arctic air from surrounding influences. The warmer tropics and cooler polar regions predicted by this model are just what we have begun to see in the past decade, says Shindell. The radiative cooling has been amplified by changes in atmospheric circulation caused by the greenhouse effect. Shindell predicts that the Arctic vortex will become more like its southern cousin in the next few years. By 2020, he says, the Arctic lower stratosphere will be 8 degrees C to 10 degrees C colder than it would have been without the greenhouse effect. As a result, ozone loss will be double what it would have been, and the repair of the ozone layer as the world reduces emissions of ozone-eating gases will be delayed. While chlorine levels are expected to peak in the stratosphere within the next five years, Shindell says that ozone loss over the Arctic will continue to rise for a further 10 or 15 years. To add a frisson to those who live in the North, Shindell's model runs predict that over Greenland and northern Europe more ozone will be destroyed than anywhere within the Antarctic ozone hole. And where the stratosphere goes, the mesosphere seems to follow. Though the evidence is more fragmentary, the same unexpectedly intense cooling is extending into the mesosphere, only more so. In February this year, specialists on these rarefied regions of the atmosphere met in Pune, India. Though everyone there admitted that these are still early days, there was widespread concern that not enough is known about how changes in the upper atmosphere might affect the planet's inhabited zones. The director of the UN's World Climate Research Programme, Hartmut Grassl, called the upper atmosphere his new number one priority. The mesosphere is just about the least explored part of the atmosphere, says John Plane of the University of East Anglia in Norwich. It is too high for research aircraft, which can reach 20 kilometres, and weather balloons, which can get to 45 kilometres. But it is too low for satellites, which can't easily be maintained in orbits below 140 kilometres.

Trickle down ~~~~~~~~~~~~ It has always been the coldest part of the planet. At its base, where it meets the warmer stratosphere, temperatures are about 0 degrees C. But the thermometer falls to -100 degrees C or lower at the top, the meso-pause, and a steady trickle of data since the 1960s indicates that it is getting colder. These data come from a variety of sources. They include instruments strapped to Russian rockets as they headed into space in the 1960s, and two proxies for temperature: laser radar (lidar) measurements of distinctive layers of metal atoms in the mesosphere, and the height at which radio waves bounce back towards the Earth. There is also evidence that the mesosphere is becoming more cloudy, a sign of colder temperatures that Thomas discovered almost a decade ago. Moreover, the atmosphere seems to be shrinking--another sign of cooling aloft (see "The sky is falling", p 28). Thomas is still analysing the data, but he announced at Pune that the mesosphere appears to have been cooling by as much as 1 degree C a year over the past 30 years. That the mesosphere is getting colder is not unexpected; like the stratosphere it should be subject to cooling as the troposphere warms. It is the size of this cooling that has taken everyone by surprise. Ray Roble of the High Altitude Observatory at the National Center for Atmospheric Research in Boulder, Colorado, predicted in 1989 that a doubling of carbon dioxide and methane levels in the atmosphere--which should take around a century at the current rate--would cool the mesosphere by about 10 degrees C and the thermosphere above it by some 40 degrees C.

Hot bursts ~~~~~~~~~~ Nobody knows why the mesosphere should be cooling so much faster than this, says Thomas. Given the size of the change, he believes it is unlikely to be radiative cooling alone that is to blame, or even a knock-on effect of the cooling stratosphere. "The models predict a decreasing effect with height, whereas the data show the opposite. There are clearly other things involved." The most popular view in Pune was, once again, that circulatory changes were amplifying the radiative cooling. Plane suggests that the cause could, confusingly, be bursts of warm air from the stratosphere pushing upwards on the mesosphere during the summer months when the atmosphere is exposed to long hours of sunlight. As the air rises and the surrounding pressure decreases, it is forced to expand. This requires energy, and as the only energy source is the heat in the air itself, it is forced to cool. The process is known as adiabatic cooling. Of course, the importance of these upward surges of air is still debatable, not least because nobody yet knows what could cause them. However, Thomas has a suggestion. He thinks the culprit may be planetary waves--huge atmospheric waves created by weather systems in the troposphere, with wavelengths measured in thousands of kilometres. Although the waves are barely perceptible at lower altitudes, they cause much stronger air movements in the thinner air of the upper atmosphere. In the mesosphere they become unstable and "break", says Thomas. The energy that they create produces strong winds searing through the mesosphere which suck up air from below, forcing it to cool adiabatically. Thomas and Plane both believe that changes in planetary waves may be the mysterious driving force behind the massive cooling of the mesosphere. Are such changes caused by global warming in the troposphere? We can't be sure, they say, but what else could it be? And there's an extra twist to the plot. As in the stratosphere, cooling in the mesosphere may be influencing the processes that create ozone holes. Plane warns that the air upwelling into the mesosphere will eventually travel to polar regions in winter. The air will then descend, he says, into the stratospheric polar vortex, where it could help lower the temperature still further and encourage the formation of polar stratospheric clouds. Of course, many of these connections remain to be proved--this isn't called the ignoro-sphere for nothing. But whatever the cause, the whole region does appear to be cooling at an alarming rate. And unexplained changes to planetary waves, jet streams and a plethora of little-known features of upper atmosphere circulation are probably amplifying the radiative cooling from the greenhouse effect. Should we care about events so far above our heads? Yes, we should, says Thomas. For one thing, they seem to be a sign that global warming really has arrived. For another, he says, they could change the way we understand the troposphere. "If we don't get our models right for what is happening up there, we could get things wrong down below."

The sky is falling ~~~~~~~~~~~~~~~~~~ PERHAPS THE MOST COMPELLING evidence that the upper atmosphere is indeed cooling is the fact that it seems to be shrinking. There, as anywhere else, cooler gases will take up less room. And last year, scientists from the British Antarctic Survey (BAS) in Cambridge reported in the Journal of Geophysical Research that the sky is indeed falling in. Measurements of the height of certain layers in the mesosphere that reflect radio waves showed that, as one American newspaper put it: "Chicken Little may have been right". Over Antarctica, the top of the mesosphere has descended by about 8 kilometres over the past four decades. Similar observations have also been obtained over Europe. The measurements are not easy. The boundaries of various parts of the atmosphere expand and contract dramatically from day to night as temperatures fluctuate. There are also variations through the seasons and in response to changes in solar activity and the Earth's magnetism. But Martin Jarvis of BAS says the long-term signal is now clear. He has little doubt that radiative cooling caused by the greenhouse effect, perhaps amplified by changes in atmospheric circulation, is to blame. "It is another warning signal," he says. With the greenhouse effect likely to accelerate, few doubt that the atmosphere will continue to get smaller. Ray Roble from the NCAR in Boulder, Colorado, for instance, calculates that the doubling of carbon dioxide levels anticipated within the next century will bring the edge of space 20 kilometres closer. He also believes the region above the mesosphere, the thermosphere, will become less dense. A contracting atmosphere could have unpredictable effects, not least for satellites. Roble estimates the density of the atmosphere at any given height in the thermosphere, where some satellites orbit, could be 50 per cent less within a few decades--and a less dense atmosphere means less drag. This would disrupt satellites' orbits and could require complex redesigns. More worryingly space debris will last longer before burning up in the atmosphere, adding to the hazards for satellites and astronauts.

From New Scientist, 1 May 1999 =A9 Copyright New Scientist, RBI Limited 1999