With Speed and Violence: Why Scientists Fear Tipping Points in Climate Change (8 page)

Sixteen years later, Hansen was the senior U.S. government employee who, seven days before the 2004 presidential election, began a public lecture with the words "I have been told by a high government official that I should not talk about dangerous anthropogenic interference with climate, because we do not know how much humans are changing the earth's climate or how much change is dangerous. Actually, we know quite a lot." And he went on to describe what we know in some detail. Most of his fellow researchers thought that would be the end for Hansen as a government employee. But a year later this outwardly diffident man-who couldn't stop apologizing for keeping me waiting when we met in his large, paper-strewn office-was still at his post. To the astonishment of many of his colleagues. "He is saved by his science; he is just too good to be fired," said one. "Also, he is one of the good guys. He doesn't have enemies. If he needed saving, there are a lot of people who would volunteer for the job."

And now Hansen says the world, or more particularly Greenland, is on a slippery slope to hell. We had better listen.

The world's three great ice sheets-one over Greenland and the other two over Antarctica-contain vast amounts of ice. Leftovers from the last ice age, they are piles of compressed snow almost 2 miles high. Glaciologists divide the sheets into two parts. On the high ground inland, where snowfall is greatest and melting is least, they accumulate ice. But on the edges and on lower ground, where snowfall is usually less and melting is greater, they lose ice. The boundary between the two zones is known as the equilibrium line.

For many centuries these great ice sheets have been in balance, with ice loss at the edges matched by accumulation in the centers, and the equilibrium line remaining roughly stationary. Glaciologists have regarded this balance as rather secure, since such huge volumes of ice can change only very slowly. Glacially. This image of stability and longevity is reassuring. If the ice sheets all melted, or slumped into the ocean, they would make a big splash. They contain enough ice to raise sea levels worldwide by 230 feet. That would drown my house, and probably yours, too. Luckily, as glaciologists have been telling us for years, this won't happen. Not even if there is fast global warming. Large ice sheets, they say, tend to maintain their own climate, keeping the air above cold enough to prevent large-scale melting. And even if warming did take hold at the surface, it could penetrate the tightly packed ice only extremely slowly.

The scariest suggestion, made by the IPCC in 2001, was that beyond a warming of about 5°F, Greenland might gradually start to melt, with a wave of warmth moving down through the ice. Once under way, the process might be unstoppable, because as the ice sheet melted, its surface would lower and become exposed to ever-warmer air. But the melting would take place very slowly, "during the next thousand years or more." Now, that is not a nice legacy to leave to future generations, but a thousand years is forty or so generations away. So maybe it is not something to worry us today.

That used to be the scientific consensus. But Hansen is the spokesperson for a growing body of glaciologists who say that things could happen much faster. Because ice sheets, even the biggest and slowest and most stable-looking, have a secret life involving dramatic and dynamic change. And their apparent stability could one day be their undoing. The story is told best in a single picture. Hansen's "slippery slope" caption accompanied a photograph of a river of water flowing across the Greenland ice sheet and pouring down a hole. The photo has an apocalyptic feel, and in the top right-hand corner a couple of researchers look on from a distance, giving an awesome sense of scale.

What is going on here? The water is not entirely new. Small lakes have always formed on the surface of Greenland ice in the summer sun. And sometimes those lakes empty down flaws in the ice-whether crevasses or vertical shafts, which are known to glaciologists as moulins. But what is new is the discovery that as the surface warms, more and more water is pouring into the interior of the ice sheet. Waterfalls as high as 2 miles are taking surface water to the very base of the ice, where it meets the bedrock. "The summer of 2005 broke all records for melting in Greenland," says Hansen. And such melting threatens to destabilize large parts of the ice sheet on timescales measured in years or decades, not millennia.

Jason Box, of Ohio State University, is a young researcher who knows more about this than most. Every year, he visits Swiss Camp, a research station set up in r 990 on Greenland ice. The name was chosen by the camp's founder, Konrad Steffen, of Zurich, so that he felt more at home. The station was originally sited on the equilibrium line, where the ice melt in summer exactly matches the accumulation of new snow in winter. But the equilibrium line has since moved many miles north, as ever-larger chunks of Greenland find themselves in the zone of predominant melting. These days, Box goes boating in an area close to Swiss Camp dubbed the "Greenland Lake District." "Some of these lakes are three or four miles across and have lasted for a decade or more now," he says. "You wouldn't think it was Greenland at all."

The lakes are more than just symptoms of melting. They are also reservoirs for the destruction of the ice sheet. "These lakes keep growing and growing until they find a crevasse, into which they drain,"Box says. "Down there are extensive river systems, between the ice and the hard rock, that eventually emerge at the glacier snout. There may be great lakes, too."

Another regular visitor to Swiss Camp is the glaciologist Jay Zwally, one of Hansen's colleagues at NASA. He made the alarming discovery that during warm years the half-mile-thick ice lifts off the bedrock and floats on the water-rising half a yard or more at times. And it floats toward the ocean. Ice sheets are never entirely still, of course. But Swiss Camp is already more than a mile west of where it started. And Zwally found that in summer, when the surface is warmer and more water pours down the crevasses, the velocity of the ice sheet's flow increases. Acceleration starts a few days after the melting begins at the surface. It stops when the melting ceases in the autumn.

This discovery is a revelation, glaciologists admit. "These flows completely change our understanding of the dynamics of ice-sheet destruction," says Richard Alley, of Penn State. "We used to think that it would take Io,ooo years for melting at the surface to penetrate down to the bottom of the ice sheet. But if you make a lake on the surface and a crack opens and the water goes down the crack, it doesn't take io,ooo years, it takes ten seconds. That huge lag time is completely eliminated."

As ever, Alley has a good analogy. "The way water gets down to the base of glaciers is rather the way magma gets up to the surface in volcanoesthrough cracks. Cracks change everything. Once a crack is created and filled, the flow enlarges it and the results can be explosive. Like volcanic eruptions. Or the disintegration of ice sheets." The lakes on the surface of Greenland are, he says, the equivalent of the pots of magma beneath volcanoes. "More melting will mean more lakes in more places, more water pouring down crevasses, and more disintegration of the ice." No wonder, in a paper in Science, Zwally called the phenomenon "a mechanism for rapid, large-scale, dynamic responses of ice sheets to climate warming."

Could such processes be close to triggering a runaway destruction of the Greenland ice sheet? It is hard to be sure, but Greenland does have past form, says David Bromwich, Box's colleague at Ohio State. There is good evidence that the ice sheet lost volume around 120,000 years ago, during the warm era between the last ice age and the previous one. "Temperatures then were very similar to those today," he says. "But the Greenland ice sheet was less than half its present size." He believes that the Greenland ice sheet is a relic of the last ice age whose time may finally have run out. "It looks susceptible, and with the drastic warming we have seen since the 198os, the chances must be that it is going to melt, and that water will go to the bottom of the ice sheet and lubricate ice flows."

Greenland melting seems to have set in around 1979, and has been accelerating ever since. The interior, above the rising equilibrium line, may still be accumulating snow. But the loss of ice around the edges has more than doubled in the past decade. The NASA team believes that "dynamic thinning" under the influence of the raging flows of meltwater may be responsible for more than half of the ice loss. In early 2006, it reported the results of a detailed satellite radar study of the ice sheet showing that it was losing 18o million acre-feet more of ice every year than it was accumulating through snowfall. That was double the estimated figure for a decade before. And all this gives real substance to the evidence accumulating from Greenland's glaciers, the ice sheet's outlets to the ocean.

Swiss Camp is in the upper catchment of a glacier known as Jakobshavn Isbrae. It is Greenland's largest, flowing west from the heart of the ice sheet for more than 400 miles into Baffin Bay. It drains 7 percent of Greenland. Jakobshavn has for some decades been the world's most prolific producer of icebergs. From Baffin Bay they journey south down Davis Strait; past Cape Farewell, the southern tip of Greenland; and out into the Atlantic shipping lanes. Jakobshavn was the likely source of the most famous iceberg of all-the one that sank the Titanic in 1912. But it has been in overdrive since 1997, after suddenly doubling the speed of its flow to the sea. It is now also the world's fastest moving glacier, at better than 7 miles a year.

Jason Box has installed a camera overlooking the glacier to keep track. It takes stereo images every four hours throughout the year. As well as flow ing ever faster toward the sea, he says, the glacier is becoming thinner, and in 2003 a tongue of ice 9 miles long that used to extend from its snout into the ocean broke off. "What is most surprising is how quickly this massive volume of ice can respond to warming," says Box. There seems to be a direct correlation between air temperatures in any one year and the discharge of water from glaciers into the ocean. Long time lags, once thought to be a near-universal attribute of ice movement, are vanishing. Jakobshavn, he estimates, could be shedding more than 40 million acre-feet a year, an amount of water close to the flow of the world's longest river, the Nile. Half of that volume is water flowing out to sea from beneath the glacier, and half is calving glaciers.

Other Greenland glaciers are getting up speed, too. The Kangerdlugssuaq glacier, in eastern Greenland, which drains 4 percent of the ice sheet, was flowing into the sea three times faster in the summer of 2005 than when last measured in 1988. At an inch a minute, its movement was visible to the naked eye. Meanwhile, its snout has retreated by three miles in four years. This familiar pattern of faster flow, thinning ice, and rapid retreat of the ice front has also shown at the nearby Helheim glacier, where Ian Howat, of the University of California in Santa Cruz, concludes that "thinning has reached a critical point and begun drastically changing the glacier's dynamics."

Most of these great streams of ice are exiting into the ocean beneath the waterline, in submarine valleys, via giant shelves of floating ice that buttress them. But as the oceans warm, these ice shelves are themselves thinning. It is, says Hansen, a recipe for rapid acceleration of ice loss across Greenland.

The picture, then, is of great flows of ice draining out of Greenland, lubricated by growing volumes of meltwater draining from the surface to the base of the ice sheet and uncorked by melting ice shelves at the coast. All this is new and frightening. "The whole Greenland hydrological system has become more vigorous, more hyperactive," says Box. "It is a very nonlinear response to global warming, with exponential increases in the loss of ice. I've seen it with my own eyes. Even five years ago we didn't know about this." Alley agrees: "Greenland is a different animal from what we thought it was just a few years ago. We are still thinking it might take centuries to go, but if things go wrong, it could just be decades. Everything points in one direction, and it's not a good direction."

"Building an ice sheet takes a long time-many thousands of years," says Hansen. "It is a slow, dry process inherently limited by the snowfall rate. But destroying it, we now realize, is a wet process, spurred by positive feedbacks, and once under way it can be explosively rapid."

8

THE SHELF

Down south, shattering ice uncorks the Antarctic

Over three days in March 2002, there occurred one of the most dramatic alterations to the map of Antarctica since the end of the last ice age. It happened on the shoreline of the Antarctic Peninsula-a tail of mountains 1,200 miles long and more than a mile high pointing from the southern part of the continent toward the tip of South America. A shelf of floating ice larger than Luxembourg and some 650 feet thick, which had been attached to the peninsula for thousands of years, shattered like a huge pane of glass. It broke into hundreds of pieces, each of them a huge iceberg that floated away into the South Atlantic.

There were no casualties, except the self-esteem of Antarctic scientists who believed that after a century of studying the continent's ice, they knew how it behaved. Their subsequent papers revealed their shock. "The catastrophic break-up of the Larsen B ice shelf is remarkable because it reveals an iceberg production mechanism far different from those previously thought to determine the extent of Antarctic ice shelves," wrote Christina Hulbe, a peace activist and glaciologist from Portland State University, in Oregon. Rather than the normal "infrequent shedding of icebergs at the seaward ice front," this time "innumerable icebergs were created simultaneously through the entire breadth of the shelf."

The demise of the Larsen B ice shelf was not in itself a surprise. Both the air and the water around the Antarctic Peninsula had been warming since the r96os. It had become one of the hot spots of global warming. Warm currents had been gradually eating away at the underside of the floating shelf, while warmer air produced pools of melting water on the surface. It was obvious that the sheet was under strain. Some cracks formed across the surface in 1994; a chunk around the edge of the shelf broke off in 1998. But nothing had prepared glaciologists for what was about to happen. During January 2002, the height of the southern summer, temperatures hit a new high and the heavy winter snow on the shelf's surface began to melt. By the end of the month, satellite pictures showed dark streaks across the shelf. Some were ponds, but others were crevasses that had filled with water.