With Speed and Violence: Why Scientists Fear Tipping Points in Climate Change (9 page)

Water is denser than ice. So, once inside the crevasses, it created pressure that levered them ever wider. There were, in effect, thousands of mechanical wedges pushing ever deeper into the ice shelf. Then, in three climactic days at the start of March, the entire structure gave way. Some 500 billion tons of ice burst into the ocean. In many ways, says Richard Alley, what happened at Larsen B mirrored the processes under way in Greenland. "Water-filled cracks more than a few tens of yards deep can be opened easily by the pressure of water. Ponding of water at the ice surface increases the water pressure wedging cracks open." In their enthusiasm to study ice, glaciologists had forgotten about water.

Larsen B was one of a series of floating shelves formed by ice draining from the mountains of the Antarctic Peninsula. The shelves are the floating front edges of glaciers, and where they meet the ocean, icebergs regularly break off. In recent years, Larsen B had been moving forward by about a yard a day. Despite this constant movement, the ice shelf itself, at more than 650 feet thick, was a surprisingly permanent structure. After its collapse, study of the diatoms in the sediment beneath the former shelf suggested that Larsen B had been there for the entire 12,000 years since the end of the last ice age, when a single ice sheet covered the whole region.

Larsen B wasn't alone; nor has it been alone in disappearing. In all, more than 500 square miles of ice shelves have been lost from around the Antarctic Peninsula in the past half century. The Larsen A ice shelf, the other side of an ice-covered headland called Seal Nunatak, broke up in a storm in 1995. And before that, the Wordie shelf, on the west side of the peninsula, disappeared between 1974 and 1996, triggering a dramatic thinning of the glaciers that fed it. But both were much smaller than Larsen B, and neither disappeared in the catastrophic manner of Larsen B.

"Really we don't think there is much doubt that the collapse of the Larsen B shelf was caused by man-made climate change," says John King, chief climatologist at the British Antarctic Survey (BAS), the inheritor of the great tradition of explorers such as Robert Scott and Ernest Shackleton. From their base at Rothera, on Adelaide Island, BAS researchers have mapped in detail how a pulse of warmer air temperatures has pushed south across the peninsula over the past fifty years, lengthening the summer melt season, sending glaciers into retreat, and destabilizing ice shelves as it goes.

Armed with the evidence of Larsen B, glaciologists are reassessing the stability of dozens of peninsula ice shelves-starting with Larsen C, immediately to the south, which is thinning and widely expected to be the next to go. Eventually, they say, the warming will reach the Ronne ice shelf, a slab of ice the size of Spain at the south of the peninsula. And on the other side of the continent is the Ross ice shelf, the continent's largest. It, too, now seems to be vulnerable, says Hulbe.

Disappearing ice shelves do not contribute to sea level rise because their ice is already floating. Their loss no more raises sea levels than an ice cube melting in a drink causes the glass to overflow. But their disappearance does change what happens inland. Ice shelves buttress the glaciers that feed them. After Larsen B disappeared, it was "as if the cork had been removed from a bottle of champagne," says the French glaciologist Eric Rignot, who works at NASA's Jet Propulsion Laboratory, in California. The glaciers that once discharged their ice onto the Larsen B shelf are now flowing into the sea eight times faster than they did before the shelf collapsed. Similar acceleration has happened after other ice sheet collapses. And that faster discharge of ice from land into the ocean is raising sea levels. With the Ross Sea being the main outlet for several of the largest glaciers on the West Antarctic ice sheet, which contains enough ice to raise sea levels by six yards, the stakes are rising.

9

THE MERCER LEGACY

An Achilles heel at the bottom of the world

John Mercer was an English eccentric and, frankly, somewhat disreputable. The list of charges against him is long. He had a penchant for doing his fieldwork in the nude, and was once convicted for jogging naked near his campus at Ohio State University, in Columbus. He regularly fell out with colleagues, and once abandoned two graduate students, including his acolyte and eventual successor Lonnie Thompson, high in the Andes after the money ran out on a field trip. Thompson thought it was something he'd said, until he realized that "those kinds of things kept happening to John; he was the same with everyone."

Mercer, who died of a brain tumor in 1987, is now a largely forgotten figure outside the glaciology community. But within it he is regarded by many, not least Thompson himself, as a genius. In the late 1940s, he set off alone to explore the ice in distant Patagonia, mapping much of the area, and came to realize that tropical glaciers might hold clues to the history of the world's climate. He is credited with inventing the term "greenhouse effect" during a symposium at Ohio State in the early 196os. But probably his greatest legacy is in Antarctica, where back in the 196os he made a prophetic warning that may one day ensure the revival of his memory.

At a time when everyone else saw Antarctic ice as just about the most dependable glacial feature on the planet, Mercer began to argue that much of it may have entirely disintegrated during the last interglacial era, about 125,000 years ago. And, though it took him a decade to get his warning into print, he feared that it might be about to happen again. In 1978, in Nature, he published a paper declaring: "I contend that a major disastera rapid deglaciation of West Antarctica-may be in progress ... within about 50 years."

The two ice sheets covering Antarctica are vast. The smaller of them, the West Antarctic ice sheet, covers around 1.5 million square miles. It is vulnerable because, unlike its larger eastern neighbor, it does not sit on dry land. Instead, like a giant ship that has foundered in shallows, it is perched precariously on an archipelago of largely submerged mountains. Ocean currents are swirling beneath its giant ice shelves. The sea temperatures today are close to freezing, but the risk is that as they rise, melting will loosen the ice sheet's moorings.

The heart of the West Antarctic ice sheet has some protection from the ocean. On two sides it is buttressed by mountains, and on the other two sides it is held in place by the Ronne and Ross ice shelves. But Mercer warned that if the ice shelves gave way, the entire sheet could lift off and float away: "Climate warming above a critical level would remove all ice shelves, and consequently all ice grounded below sea level, resulting in the deglaciation of most of West Antarctica." Once under way, the disintegration would "probably be rapid, perhaps catastrophically so." Most of the ice sheet would be gone within a century. He reckoned that a warming of 9 degrees would be enough to set the process in train. Parts of the continent have already experienced more than 3.6 degrees of warming. "One warning sign that a dangerous warming is beginning will be the break-up of ice shelves in the Antarctic Peninsula," he said. Like Larsen B.

Another old acolyte of Mercer's is Terry Hughes, of the University of Maine. Back in 1981, he suggested that the West Antarctic ice sheet might have another vulnerability-a "weak underbelly" in Pine Island Bay, a large inlet on the Amundsen Sea, west of the Antarctic Peninsula. This is one of the most remote places on Earth. Head north from Pine Island Bay, and you don't hit land until Alaska. These are dangerous waters-deep, with unusually tall icebergs breaking off the glaciers and being blown fast across the bay by fierce winds. There is a constant danger of getting trapped by the ice if the wind changes. Onshore, the terrain is rugged, and its weather is violent, with intense snowstorms steered inland by the Antarctic Peninsula. Even Antarctic researchers have given Pine Island Bay a wide berth. There are no bases here.

Hughes's "weak underbelly" theory was, like Mercer's warnings a decade before, roundly ignored at the time. When I first wrote about it, a few years later, other glaciologists warned me off, suggesting that it had been discredited. But today, just mentioning Pine Island Bay is enough to send a shudder through the hearts of many glaciologists. Hughes, they now believe, was right on the mark.

The bay is the outlet for two of Antarctica's top five glaciers: Pine Island and Thwaites. Together, they drain about 40 percent of the West Antarctic ice sheet. They were already the fastest-flowing glaciers in Antarctica when, in the i99os, Pine Island began to accelerate sharply, and Thwaites, while traveling at the same speed, doubled its flow by becoming twice as wide. The glaciers were responding to a rapid melting of their own ice shelves. The melting was in turn caused by warmer seawater circling into the bay.

The discovery of the accelerating glaciers has, once again, turned conventional thinking about the dynamics of ice on its head. The old view holds that events on the coast, where a glacier meets the ocean, have little bearing on what happens inland. But at Pine Island Bay, the impacts of coastal melting are swiftly being felt throughout the glaciers' network of tributaries across the ice sheet. In the past decade, the flow of the two glaciers has speeded up, not just at the coast but for 125 miles inland. The NASA glaciologist Eric Rignot reported in 2004 that the two glaciers are dumping more than 200 million acre-feet of ice a year into Pine Island Bay. This dwarfs even the very heavy snowfall, which adds about 130 million acre-feet a year. The net "mass loss" of ice from the Pine Island Bay catchment has tripled in a decade.

Since Rignot's paper was published, the news has become even grimmer. Studies of the Pine Island glacier show that its ice shelf is thinning fast. As it thins, ever more warm seawater penetrates beneath the glacier. The "grounding line," the farthest point downstream where the ice makes contact with solid rock, has been retreating by more than a mile a year. Once under way, the retreat of the grounding line is "theoretically selfperpetuating and irreversible, regardless of climate forcing," says Rignot. The glacier is primed for runaway destruction.

In 2005, British and Texas researchers flew more than 45,000 miles on more than a hundred flights back and forth across the Pine Island and Thwaites glaciers, using ice-penetrating radar to map the rocks beneath an area of ice the size of France and sometimes nearly 2 miles high. They found that inland along its major tributaries the Pine Island glacier sat on great lakes of meltwater. There seemed to be remarkably little to hold back its flow. Meanwhile, the Thwaites glacier, which is a stream of ice flowing through a wider area of ice sheet, could be about to widen again, says David Vaughan, of the BAS, who masterminded the survey.

If the Pine Island and Thwaites glaciers are on a one-way trip to disaster, the implications are global. Together they drain an area containing enough ice to raise sea levels worldwide by 1-2 yards. In all probability, the Pine Island and Thwaites glaciers are already the biggest causes of sea level rise worldwide. Hughes believes their collapse could destabilize the entire West Antarctic ice sheet, and potentially parts of the East Antarctic ice sheet, too. "The well-documented changes happening just within the past decade are a numbing prospect," he told me. "And we have only hints about exactly what is going on."

Days after Vaughan presented the first findings of the survey to a conference in the U.S., I met Richard Alley. He had been in the audience and had been astounded by the findings. "Thwaites just taps right into the vast reservoirs of ice in the middle of the ice sheet, and the question is whether it will drag them along with it," he said. "I think Thwaites could be absolutely critical. If you pull the plug, the ice goes faster and there is thinning. The only question is whether the plug can re-form a bit further back, or whether the ocean will deliver enough heat for it to just blowtorch its way to the center. I don't think we know the answer to that yet." There was, he said, "a possibility that the West Antarctic ice sheet could collapse and raise sea levels by 6 yards in the next century."

The East Antarctic ice sheet is the biggest, highest slab of ice on the planet. In the unlikely event that it all melted, sea levels would rise by 50 yards or more. But it has been in place for some 20 million years. And in 2005, Curt Davis, of the University of Missouri, reported, after analyzing satellite data, that extra snowfall linked to global warming is raising the height of the ice by almost three quarters of an inch a year-enough to shave current rates of sea level rise by i0 percent. All seemed well, then, with the East Antarctic ice sheet.

But there was a slight problem. Davis's study could cover only the flat interior. Satellite instruments are not yet good enough to establish altitude trends near the coasts, where there is sloping terrain. A footnote to his paper mentions that "mass loss in areas near the coast could be even greater than the gains in the interior." Unfortunately, other researchers say that is precisely what may be happening.

Exhibit A in this case is the Totten glacier. It is a biggie-62 miles wide at its mouth, where it calves icebergs into the Indian Ocean. Totten's network of tributary glaciers drains an area containing more ice than the whole of the West Antarctic. And since the early 19gos, says Andy Shepherd, of the Scott Polar Research Institute, in Cambridge, England, that catchment has been losing enough ice to lower its height by more than 10 yards a year. Another giant of the East Antarctic ice sheet, the Cook glacier, is doing the same.

The last bastion of glacial stability suddenly looks much less safe. And Shepherd points out that Totten and Cook have something else in common with Pine Island, Thwaites, and the other troublesome glaciers on the west side-something suggesting that worse could be ahead. Both Totten and Cook have grounding lines in the ocean that are below sea level-more than 300 yards below in the case of Totten. That is, its contact with the continental land mass is so tenacious that the glacier slides 300 yards under water before the ice gives up contact with the rock and begins to float. That sounds like good news: evidence of stability. The problem is that warmer waters appear to be weakening that contact. Should the grounding line start to retreat, we can expect the glacier to begin the familiar process of thinning and accelerating. The retreat would, in other words, remove the cork from a very large bottle.