With Speed and Violence: Why Scientists Fear Tipping Points in Climate Change (22 page)

But the crux of the public debate on Broecker's ocean conveyor remains a very simple question: Could global warming shut the conveyor down? Broecker seems rarely to have doubted it. And the claim has in recent years seemed almost to have a life of its own. This struck me most strongly at a conference on "dangerous" climate change held at the Hadley Centre for Climate Prediction, in Exeter in 2005. There I met Michael Schlesinger, of the University of Illinois at Urbana-Champaign. He is a sharp-suited guy sporting a pastiche of 1950s clothes and hairstyle. But if there were serious doubts in Exeter about whether his style sense would ever come back into fashion, there was no doubt that his ideas about climate change had found their moment.

For more than a decade, Schlesinger has been making Broecker's case that a shutdown of the ocean conveyor could be closer than mainstream climate modelers think. Some critics feel that he just doesn't know when to give up and move on. But he has stuck with it, criticizing the IPCC and its models for systematically eliminating a range of quite possible doomsday scenarios from consideration. "The trouble with trying to reach a consensus is that all the interesting ideas get eliminated," he said at the conference. Science by committee ends up throwing away the good stufflike the idea of the conveyor's shutting down. But in Exeter, Schlesinger was back in vogue. He had been invited to present his model findings that a global warming of just 3.6°F would melt the Greenland ice sheet fast enough to swamp the ocean with freshwater and shut down the conveyor. The risk, he said, was "unacceptably large."

Although he had been saying much the same for a decade, he was now considered mainstream enough to be invited across the Atlantic to expound his ideas at a conference organized by the British government. And he was no longer alone. Later in the day, Peter Challenor, of the British National Oceanography Centre, in Southampton, said he had shortened his own odds about the likelihood of a conveyor shutdown from one in thirty to one in three. He guessed that a 3-degree warming of Greenland would do it. Given how fast Greenland is currently warming, that seems a near certainty.

But all this is models. What evidence is there on the ground for the state of the conveyor? The truth is that dangerous change is already afoot in the North Atlantic. And, whatever the skepticism about some of Broecker's grander claims, the conveyor may already be in deep trouble. Since the mid-r96os, says Ruth Curry, of the Woods Hole Oceanographic Institution, the waters of the far North Atlantic off Greenland-where Wadhams's chimneys deliver water to the ocean floor and maintain Broecker's conveyor-have become decidedly fresher.

In fact, much of the change happened back in the i96os, when some 8 billion acre-feet of freshwater gushed out of the Arctic through the Fram Strait. Oceanographers called the event the Great Salinity Anomaly. To this day, nobody is quite sure why it happened. It could have been ice breaking off the great Greenland ice sheet, or sea ice caught up in unusual circulation patterns, or increased flow from the great Siberian rivers like the Ob and the Yenisey. Luckily, most of the freshwater rapidly headed south into the North Atlantic proper. Only 3 billion acre-feet remained. Curry's studies of the phenomenon, published in Science in June 2005, concluded that 7 billion acre-feet would have been enough to "substantially reduce" the conveyor, and double that "could essentially shut it down." So it was a close call.

With the region's water still substantially fresher than it was at the start of the r96os, the conveyor remains on the critical list. Another single slug of freshwater anytime soon could be disastrous. In the coming decades, some combination of increased rainfall, increased runoff from the land surrounding the Arctic, and faster rates of ice melting could turn off the conveyor. And there would be no turning back, because models suggest that it would not easily switch back on. "A shift in the ocean conveyor, once initiated, is essentially irreversible over a time period of many decades to centuries," as Broecker's colleague Peter deMenocal puts it. "It would permanently alter the climatic norms for some of the most densely populated and highly developed regions of the world."

As I prepared to submit this book to the publisher, new research dramatically underlined the risks and fears for the conveyor. Harry Bryden, of the National Oceanography Centre, had strung measuring buoys in a line across the Atlantic, from the Canary Islands to the Bahamas, and found that the flow of water north from the Gulf Stream into the North Atlantic had faltered by 30 percent since the mid- i99os. Less warm water was going north at the surface, and less cold water was coming back south along the ocean floor. This weakening of two critical features of the conveyor was, so far as anyone knew, an unprecedented event.

Probing further, Bryden found that the "deep water" from the Labrador Sea west of Greenland still seemed to be flowing south. But the volume of deep water coming south from the Greenland Sea, the site of Wadhams's chimneys, had collapsed to half its former level. The implication was clear: the disappearing chimneys that Wadhams had watched with such despair were indeed hobbling the ocean circulation. Broecker seemed on the verge of being proved right that the ocean conveyor was at a threshold because of global warming.

None of this demonstrated that Broecker's bleaker predictions of what would happen if the conveyor shut down were about to come truethat "London would experience the winter cold that now grips Irkutsk in Siberia." Something more like the little ice ages was the worst that most climate modelers feared. But there did seem to be a real possibility that many of Broecker's ideas were about to be put rather dramatically to the test.

24

AN ARCTIC FLOWER

Clues to a climate switchback

It must have felt like the springtime of the world. Anybody living on Earth 13,000 years ago could only have felt elation. An ice age of some 8o,ooo years was coming to an end. Temperatures were rising; ice was melting; rivers were in flood; and permafrost was giving way to trees and meadows across Europe and North America. In the Atlantic Ocean, the Gulf Stream was pushing north again, bringing warm tropical water and reestablishing an ocean circulation system that had shut down entirely in the depths of the ice age. Westerly winds blowing across the ocean were picking up the heat and distributing it across Europe and deep into Asia.

Meanwhile, in the tropics, the deserts were in retreat, the rainforests were expanding again from their ice-age refuges, and the Asian monsoon was kicking back in. Most spectacularly, the Sahara was bursting with life, covered in vegetation and huge lakes. This was the dawn of the age of Homo sapiens, who had supplanted the last of the Neanderthals during the long glaciation. If there had been a Charles Keeling around, he would have measured rising atmospheric levels of carbon dioxide and methane that were amplifying the thaw. He might even have invented the term "global warming" to describe it.

Then the unthinkable happened: the whole thing went into reverse again. Almost overnight, the thaw halted and temperatures plunged. Temperatures became as cold as they had been in the depths of the ice age. The forests returning to northern climes were wiped out; the permafrost extended; and ice sheets and glaciers started to regain their former terrain.

The springtime seemed to be over almost before it began. But this reversal was not the first. The previous 5,000 years had been full of them. Some i8,ooo years before the present, there was still a full-on ice age. By 16,ooo years ago, the world was warming strongly. But by 15,000 years ago, it was cold again, with ice sheets reforming. At 14,500 years ago, it became so warm that within 400 years the ice caps melted sufficiently to raise sea levels worldwide by 65 feet. The cold gained the upper hand once more, only to give way to the pronounced warming of 13,000 years ago, which crashed again 12,800 years ago.

Today we can see this extraordinary climatic history recorded in ice cores extracted from the ice of Greenland and Antarctica. Graphs of the temperatures back then look like seismic readings during a big earthquake-or cardiac readouts during a heart attack. They show a climate system in a protracted series of spasms. Looking back, we recognize the death throes of the ice age. But that is with hindsight. At the time, there was little evidence that the climate system had any sense of direction at all. It lurched between its glacial and interglacial modes. The one thing it didn't do was settle for a happy medium.

The last great cold snap of the ice age, 12,800 years ago, is known today as the Younger Dryas era. The dryas is a white Arctic rose with a yellow center that suddenly reappeared in European sedimentary remains, indicating that the old cold reasserted itself. The era is called the Younger Dryas to distinguish it from the Older Dryas, the climate reversal of a thousand years earlier, and the Oldest Dryas, which came before that. The Younger Dryas, like the others, was swift and dramatic. Within about a generation, temperatures fell worldwide-perhaps by as little as 3 to 5°F in the tropics, but by an average of as much as 28 degrees farther north, and, according to ice cores analyzed by George Denton, of the University of Maine, by 54 degrees in winter at Scoresby Sound, in eastern Greenland.

Not only temperatures crashed. Records of Chinese dust and African lakes and tropical trade winds and South American river flows and New Zealand glaciers all reveal dramatic changes happening in step 12,800 years ago. The world was much drier, windier and dustier. But in the Southern Hemisphere, temperatures may have gone in the opposite direction. Marine sediment cores show dramatic warming in the South Atlantic and the Indian Ocean-as do temperature records in most Antarctic ice cores.

The Younger Dryas freeze lasted for fifty or so generations: 1,300 years. One can imagine tribes of Homo sapiens desperately relearning the crafts that got their ancestors through the ice ages. But it may also have triggered innovation. Some believe that dry conditions in the Middle East at the time may have encouraged the first experiments with crop cultivation and the domestication of animals. And then the freeze ended, and temperatures returned to their former levels even faster than they had fallen. Analysts of the Greenland ice-core chronology say publicly that the warming must have happened within a decade. But that is the minimum time frame for the change of which they can be certain, given the resolution of the ice cores. Richard Alley, who was there handling the ice cores, says: "Most of that change looks like it happened in a single year. It could have been less, perhaps even a single season. It was a weird time indeed." Like The Day After Tomorrow, only in reverse.

All this is doubly strange, because the Younger Dryas cooling went against the grain of all the long-term trends for the planet. The orbital changes that had triggered the glaciation had faded by then; astronomical forces were pushing the planet toward the next interglacial era. Of course, the real work was being done by feedbacks like melting ice, the return of greenhouse gases like carbon dioxide and methane into the atmosphere, and the revival of the ocean conveyor. These feedbacks would have turned a smooth progression into a series of jumps. But they would not easily have altered the direction of change. So why the backward flip? What made climate plunge back into the icy abyss when all the forcings and all the feedbacks should have been kicking the world into warmer times?

Chaos theory may help here. Alley says that it is just when conditions are changing fastest that the chances for seemingly random, unexpected, and abrupt change are greatest. The system is stirred up and vulnerable. The drunk is on a rampage. And there is a reasonable chance that some of the abrupt changes will be in the opposite direction to that expected. This is what, in the clever subtitle to his 2001 report on abrupt climate change, Alley called "inevitable surprise." What is equally clear is that at the time, the entire planetary climate system had just two possible states: glacial and interglacial. It knew no third way. And so, during the several thousand years when it was on the cusp between the two, it flickered between them.

On the ground, one element was a sudden switch in Broecker's ocean conveyor. It would be going too far to say that the Younger Dryas proves that the global conveyor is the great climate switch that Broecker claims. But the event makes a compelling case that events in the far North Atlantic can, without help from astronomical or any other forces, sometimes have dramatic and long-lasting effects on global climate.

The unexpected switch of the ocean conveyor was almost certainly triggered by melting ice. In the final millennia of the ice age, as melting made fitful but sometimes dramatic progress, a very large amount of liquid water was produced. Often it did not pour directly into the oceans but formed giant lakes on the ice or on land around the edges. The largest known of these is called Lake Agassiz, after the discoverer of the ice ages. It stretched for more than 6oo miles across a wide area of the American Midwest, from Saskatchewan to Ontario in Canada, and from the Dakotas to Minnesota in the U.S., generally moving with the advancing front of warming.

In the early stages of the deglaciation, the lake drained south, down the Mississippi River into the Gulf of Mexico. But about I2,Soo years ago, it seems, something stopped this and forced the lake to drain east. Perhaps the route south was blocked by land gradually rising after the weight of the ice was removed. Perhaps the lake simply passed over a natural watershed as it moved north with the retreating face of the ice sheet. But at any rate, there was eventually a huge breakout of freshwater from the heart of North America into the basin now occupied by the Great Lakes, and on into the North Atlantic.

The vast inrush of cold freshwater would have drastically cooled and freshened the ocean. High salinity was critical for sustaining the newly revived, and perhaps still precarious, ocean conveyor. So a fresher ocean shut down the conveyor once more. The warm Gulf Stream was no longer drawn north. Temperatures crashed across the North Atlantic region, and probably particularly around Greenland. The entire global climate system would have been shaken, and may have lurched back from its interglacial to its glacial mode.