Global Meltdown

Richard Alley's eyes glint as we sit in his office in the University of Pennsylvania discussing how fast global warming could cause sea levels to rise. The scientist sums up the state of knowledge: "We used to think that it would take 10,000 years for melting at the surface of an ice sheet to penetrate down to the bottom. Now we know it doesn't take 10,000 years; it takes 10 seconds."

That quote highlights most vividly why scientists are getting panicky about the sheer speed and violence with which climate change could take hold. They are realising that their old ideas about gradual change - the smooth lines on graphs showing warming and sea level rise and gradually shifting weather patterns - simply do not represent how the world's climate system works.

Dozens of scientists told me the same thing while I was researching my book The Last Generation. Climate change did not happen gradually in the past, and it will not happen that way in the future. Planet Earth does not do gradual change. It does big jumps; it works by tipping points.

The story of research into sea level rise is typical of how perceptions have changed in the past five years. The conventional view - you can still read it in reports from the UN's Intergovernmental Panel on Climate Change - holds that sea levels will start to rise as a pulse of warming works its way gradually from the surface through the 2km- and 3km-thick ice sheets on Greenland and Antarctica, melting them. The ice is thick and the heat will penetrate only slowly. So we have hundreds, probably thousands, of years to make our retreat to higher ground.

Recent research, however, shows that idea is wholly wrong. Glaciologists forgot about crevasses. What is actually happening is that ice is melting at the surface and forming lakes that drain down into the crevasses. In 10 seconds, the water is at the base of the ice sheet, where it lubricates the join between ice and rock. Then the whole ice sheet starts to float downhill towards the ocean.

"These flows completely change our understanding of the dynamics of ice sheet destruction," says Alley. "Even five years ago, we didn't know about this."

This summer, lakes several kilometres across formed on the Greenland ice sheet, and drained away to the depths. Scientists measured how, within hours of the lakes forming, the vast ice sheets physically rose up, as if floating on water, and slid towards the ocean. That is why Greenland glaciers are flowing faster, and there are more icebergs breaking off into the Atlantic Ocean. That is why average sea level rise has increased from 2mm a year in the early 1990s to more than 3mm a year now.

Soon it could be a great deal more. Jim Hansen of Nasa, George Bush's top climate modeller, predicts that sea level rise will be 10 times faster within a few years, as Greenland destabilises. "Building an ice sheet takes a long time," he says. "But destroying it can be explosively rapid."
Alarmist? No. It has happened before, he says. During the final few centuries of the last ice age, the sea level rose 20 metres in 400 years, an average of 20 times faster than now. These were sudden, violent times. And the melting was caused by tiny wobbles in the Earth's orbit that changed the heat balance of the planet by only a fraction as much as our emissions of greenhouse gases are doing today.

There is more evidence of abrupt and violent change, most of it culled from ice cores, lake sediments, tree rings and other natural archives of climate. We now know that the last ice age was not a stable cold era but near-permanent climate change. Towards the end, around 11,000 years ago, average temperatures in parts of the Arctic rose by 16C or more within a decade. Alley believes it happened within a single year, though he says the evidence in the ice cores is not precise enough to prove it.

All this comes as a surprise to us because, in the 10,000 or so years since the end of the last ice age, the climate has been, relatively speaking, stable. We have had warm periods and mini ice ages; but they were little compared with events before.

It is arguable that this rather benign world has been the main reason why our species was able to leave the caves and create the urban, industrial civilisation we enjoy today. Our complex society relies on our being able to plant crops and build cities, knowing that the rains will come and the cities will not be flooded by incoming tides. When that certainty fails, as when Hurricane Katrina hit New Orleans last year, even the most sophisticated society is brought to its knees.

But there is a growing fear among scientists that, thanks to man-made climate change, we are about to return to a world of climatic turbulence, where tipping points are constantly crossed. Their research into the workings of the planet's ecosystems suggests why such sudden changes have happened in the past, and are likely again in future.

One driver of fast change in the past has been abrupt movements of carbon between the atmosphere and natural reservoirs such as the rainforests and the oceans. Hundreds of billions of tonnes of carbon dioxide can burp into the atmosphere, apparently at the flick of a switch.
That is why the Met Office's warning that the Amazon rainforest could die by mid-century, releasing its stored carbon from trees and soils into the air, is so worrying. And why we should take serious note when Peter Cox, professor of climate systems at Exeter University, warns that the world's soils, which have been soaking up carbon for centuries, may be close to a tipping beyond which they will release it all again.

Other threats lurk on the horizon. We know that there are trillions of tonnes of methane, a virulent greenhouse gas, trapped in permafrost and in sediments beneath the ocean bed. There are fears this methane may start leaking out as temperatures warm. It seems this happened 55m years ago, when gradual warming of the atmosphere penetrated to the ocean depths and unlocked the methane, which caused a much greater warming that resulted in the extinction of millions of species.

All this suggests that, in one sense, the climate sceptics are right. They say the future is much less certain than the climate models predict. They have a point. We know less than we think. But the sceptics are wrong in concluding that the models have been exaggerating the threat. Far from it. Evidence emerging in the past five years or so suggests the presence of many previously unknown tipping points that could trigger dangerous climate change.

Can we call a halt? Hansen says we have 10 years to turn things round and escape disaster. James Lovelock, author of the Gaia theory, which considers the Earth a self-regulated living being, reckons we are already past the point of no return. I don't buy that. For one thing, there is no single point of no return. We have probably passed some, but not others.