28 April 1998

Fred Pearce

Global warming creates a lot of hot air. So are we to blame or not?

Climatologists have sunspots before their eyes this month, following claims by Henrik Svensmark of the Danish Meteorological Institute in Copenhagen that solar cycles alone could explain global warming.

Svensmark publicised the ideas, which have been circulating in a small circle for some years, during a meeting last week at Cern, the centre for particle physics research in Switzerland, where researchers want funds for a laboratory experiment to test them. But this week one of the idea’s originators, the British solar physicist Brian Tinsley, told the Guardian that it could not explain recent warming, and that efforts to cut emissions of greenhouse gases should continue.

Svenmark’s argument is more complicated than the simple claim that "sunspots cause global warming". It relies on the observation that the space around planet Earth is a battleground between competing fields of radiation: cosmic rays emanating from deep in space and the more local "solar wind" of our own Sun. When sunspot activity is strong, the solar wind is also strong and it blows away cosmic rays. This much is accepted wisdom.

What Tinsley, now working at the University of Texas at Dallas, and Svensmark add that cosmic rays create clouds in the Earth’s atmosphere. If the Sun blows them away, there will be fewer clouds to shade the Earth from the Sun’s heat. How come? One theory is that by ionising the atmosphere, turning its atoms into charged particles, the rays make it more easy for water vapour to condense into cloud droplets.

But it is only theory, says Giles Harrison, an atmospheric electrician at the University of Reading, who is devising his own experiments in the field. There is as yet no hard evidence that it happens. What Svensmark does claim to have is statistical evidence. Since 1979, satellites have observed a small variation in global cloud cover, of 3-4 per cent, that seems to follow variations in the amount of cosmic rays entering the Earth’s atmosphere. And for the past century, he says, solar activity seems to have risen in line with global temperatures.

For statisticians, such correlations in time mean nothing without a plausible causal mechanism. (Cancer death rates generally rise with TV ownership, but that doesn’t mean TVs cause cancer — or that cancer causes TVs, come to that.) And even if a link is proved, it is still a long way from showing either that it is important or that it absolves greenhouse gases of any blame for global warming.

And that is where climate scientists fall out with the high-blown claims being made for this work. Keith Shine of the University of Reading is no lackey of the conventional wisdom on global warming. Two years ago, he publicly ticked off his colleagues on the UN’s Intergovernmental Panel on Climate Change for pushing the evidence too far in claiming that they had identified the hand of man in global warming.

"It could be a fluke," he said then, and he used the same word this week to describe Svensmark’s correlations. "I’m exasperated about the conclusions being drawn from a very tenuous statistical association that could be explained by many other factors. It is bizarre to make such claims when you haven’t even identified a cause and effect mechanism."

He points out that the statistical link between clouds and cosmic rays is based on satellite data from little more than a single 11-year solar cycle.

"Many natural things could resonate on a similar cycle." Harrison points out that water levels in Lake Victoria correlated with sunspots for two whole solar cycles before deviating and showing no further relationship.

Shine agrees that ionisation could make clouds. But this would probably be mainly in the upper atmosphere, where the bombardment of radiation is most intense and where there is usually a shortage of nuclei on which cloud drops can condense. Unfortunately clouds in the upper atmosphere, such as cirrus clouds, do not cool the Earth. They heat it up, by stopping heat escaping.

Climatologists admit that Svensmark’s work has refocused attention on the influence of the Sun on climate. This is often forgotten, but is hardly scientifically controversial. Astrophysicists reckon that the solar cycle alters the amount of radiation reaching the Earth by about 0.5 watts per square metre of the Earth’s surface. This is significant enough to make graphs of sunspots and global temperatures go up and down in pleasing sync.

But it is a mere ripple compared with the tidal wave of global warming from man-made greenhouse gases accumulating in the atmosphere, which are estimated to have added 2.5 watts per square metre.

Some climate scientists have rushed to dismiss the Dane’s work. Shine is not among them. "The idea that cosmic rays could play a role by influencing clouds is interesting," he says. "There might be something in it. You’d be daft not to pursue it." But that doesn’t make it true, he says.

And, he asks, "why is the assumption being made that there is only one mechanism for global warming?" Hundreds of millions of pounds have been spent in the past decade by scientists trying to disentangle natural and man-made influences on climate. To suggest that global warming must be down to one or the other exclusively is absurd, says Michael Grubb, a long-time analyst of global warming at the Royal Institute of International Affairs in London. "There is just a giant logical non-sequitur here," he says. "It is madness to claim this cosmic ray mechanism, even if it were found to be correct, proves that the greenhouse effect is some kind of myth."

And most investigators of the cosmic ray theory agree. Right now, says Harrison, there is far more hard scientific evidence for the greenhouse effect than for the theory that radiation from space could create clouds.

And Tinsley does not believe his theories could explain recent accelerated global warming. "After about 1970, the further increase should be attributed to something else, e.g. greenhouse gases," he said. "There are good theoretical reasons for expecting greenhouse gas effects to become obvious in the early 21st century." To head off disruptive effects, he said, "it would be prudent to take reasonable measures to reduce the consumption of fossil fuel."

Reply to Fred Pearce

View Contributors