Last month in Greenland, more than a tenth of the ice sheet’s surface was melting in the unseasonably warm spring sun, smashing 2010’s record for a thaw so early in the year. In the Antarctic, warm water licking at the base of the continent’s western ice sheet is, in effect, dissolving the cork that holds back the flow of glaciers into the sea; ice is now seeping like wine from a toppled bottle.
The planet’s polar ice is melting fast, and recent satellite data, models, and fieldwork have left scientists sobered by the speed of the sea level rise we should expect over the coming decades. Although researchers have long projected that the planet’s biggest ice sheets and glaciers will wilt in the face of rising temperatures, estimates of the rate of that change keep going up. When the Intergovernmental Panel on Climate Change (IPCC) put out its last report in 2013, the consensus was for under a meter (3.3 feet) of sea level rise by 2100. In just the last few years, at least one modeling study suggests we might need to double that.
Eric Rignot at the University of California, Irvine says that study underscores the possible speed of ice sheet melt and collapse. “Once these processes start to kick in,” he says, “they’re very fast.”
The Earth has seen sudden climate change and rapid sea level rise before. At the end of the planet’s last glaciation, starting about 14,000 years ago, sea levels rose by more than 13 feet a century as the huge North American ice sheet melted. But researchers are hesitant about predicting similarly rapid climate shifts in our future given the huge stakes involved: The rapid collapse of today’s polar ice sheets would erase densely populated parts of our coastlines.
“Today, we’re struggling with 3 millimeters [0.1 inch] per year [of sea level rise],” says Robert DeConto at the University of Massachusetts-Amherst, co-author of one of the more sobering new studies. “We’re talking about centimeters per year. That’s really tough. At that point your engineering can’t keep up; you’re down to demolition and rebuilding.”
Antarctica and Greenland hold the overwhelming majority of the world’s ice: Ninety percent of the planet’s freshwater ice is locked up in Antarctica’s ice cap and nine percent in Greenland’s. Today, the ice sheet that’s inarguably melting fastest is Greenland. That giant block of ice, which has the potential to raise global sea levels by 23 feet if it melts in its entirety, is losing some 200 billion tons of ice each year. That rate has doubled from the 1900s to the 2000s.
Illustration of the rapid expansion of ice melt on Greenland over just two days in April 2016. Danish Meteorological Institute
“We are seeing changes in Greenland in all four corners, even in the far north,” says Rignot. Many of the outlet glaciers that flow down fjords into the sea, which were “on the fence” about retreating or advancing over the past decade, are now “starting to fall apart,” he says.
And they’re moving fast. “The flow speeds we talk about today would have been jaw-dropping in the 1990s,” says Ted Scambos of the University of Colorado’s National Snow and Ice Data Center. Greenland’s Jakobshavn Glacier dumped ice into the sea at the astonishing rate of 150 feet per day in the summer of 2012. The most dramatic action in Greenland is simply from surface melting, as temperatures there and across the Arctic have soared in the last four decades. In 2012, Greenland lost a record 562 billion tons of ice as more than 90 percent of its surface melted in the summer sun.
Many questions remain about the physics of Greenland’s ice loss, such as whether meltwater gets soaked up by a ‘sponge’ of snow and ice, or trickles down to lubricate the base of the ice sheet and speed its seaward movement. Most modeling work has been about how Greenland’s melt tracks rising air temperatures; far less is known about how warming waters might eat away at the edges of its ice sheet. Rignot is part of a team now launching the Oceans Melting Greenland project (with the intentionally punny acronym OMG) to investigate that. These uncertainties make Rignot think that estimates of Greenland’s melt — contributing as much as 9 inches of global sea level rise by 2100, according to the 2013 IPCC report — have been far too conservative. Assuming that the Greenland ice sheet’s demise “will be slow is wishful thinking,” Rignot says.
But most scientists say there shouldn’t be too many serious surprises about the physics governing Greenland’s ice loss. Although the ice sheet can be expected to steadily melt in the face of rising temperatures, Greenland’s ice cap shouldn’t rapidly collapse, because most of its ice sits safely on rock far above sea level. “Greenland is more predictable and straightforward,” says DeConto.
For fear of rapid, runaway collapse, the research community turns its eyes south.
Antarctica is, for now, losing ice more slowly than Greenland. The latest data from the GRACE project — twin satellites that measure mass using gravity data — say Antarctica is losing about 92 billion tons of ice per year, with that rate having doubled from 2003 to 2014. But Antarctica is vast — 1.5 times the size of the United States, with ice three miles thick in places — and holds enough ice to raise global sea levels by roughly 200 feet.
The sizeable western half of Antarctica holds some of the fastest-warming areas on the planet.
The larger, eastern half lies mostly above sea level and remains very cold; researchers have typically considered its ice stable, though even that view is beginning to change. The sizeable western half of the Antarctic, by contrast, has its base lying below sea level, and holds some of the fastest warming areas on the planet. “You look at West Antarctica and you think: How come it’s still there?” says Rignot.
Warming ocean water licking at the underside of the floating edges of the Western Antarctic Ice Sheet is eating away at the line where the ice rests on solid rock. Much of the bedrock of the Antarctic slopes downward toward the center of the continent, so as the invading water flows downhill it seeps further and further inland, causing ever-larger chunks of glaciers to flow faster into the sea. This so-called “grounding line” has been eroding inland rapidly, in some parts of West Antarctica at rates of miles per year. In 2014, satellite radar images revealed just how vulnerable five massive glaciers flowing into the Admundsen Sea are from this effect. And a 2015 paper showed that the same thing is happening more slowly to Totten Glacier, one of the biggest glaciers in the east.
Such dramatic processes have been the bane of Antarctic modeling and the reason why scientists have been loathe to put a number on sea level contributions from a melting southern continent. Then in March came a report in Nature that some say represents a step change in our ability to do that. DeConto and David Pollard of Pennsylvania State University put into their ice sheet model two basic phenomena: meltwater trickling down to lubricate glacier flow, and giant walls of ice (created when the ends of glaciers snap off) simply collapsing under their own weight. These new modeling parameters gave DeConto and Pollard a better understanding of past sea level rise events. For the Pliocene era 3 million years ago, for example — when seas were dozens of feet higher than today — older models estimated that a partially melting Antarctic added about 23 feet to global sea level rise. The new model increased Antarctica’s contribution to sea level rise during the Pliocene to 56 feet.
Turning their model to the future, DeConto and Pollard project more than three feet of sea level rise from Antarctica alone by 2100 — assuming growing greenhouse gas emissions that boost the planet’s temperature by about 4 degrees C (7 degrees F). That is far more than the last IPCC estimate in 2013, which projected less than eight inches of sea level rise from a melting Antarctic by 2100, with a possibility for inches more from the dramatic collapse of Antarctic glaciers.
Satellite image showing sediment plumes from meltwater exiting glaciers in Greenland. Jesse Allen/NASA
Even DeConto admits that, under the model used in his paper, the timing and pace of Antarctica’s ice loss is “really uncertain” — it could be a decade or two, or three or four, before these dramatic processes start to kick in, he says. “The paper just shows the potentials, which are really big and really scary,” says DeConto. But Scambos and other observers call DeConto’s numbers “perfectly plausible.”
Researchers could better pin down their models if they could track the rate of sea level rise from polar ice sheet collapse in the past, but this has proven hard to do. When seas rose a whopping 13 feet per century at the end of the last glaciation (the current record-holder for known rates of sea level rise in the past), much of the water came from an ice sheet over North America, where there isn’t one today. “I wouldn’t use that as an analogue for the future,” says paleo-geologist Andrea Dutton of the University of Florida, who wrote a recent review of past records of sea level rise. “But it has important lessons for us nonetheless — that ice sheets can retreat suddenly and in steps instead of gradually.”
For a better analogue of what’s going on today, researchers often look to the last interglacial period, about 120,000 years ago, when temperatures were about a degree warmer than pre-industrial levels and seas were 20 to 30 feet higher than today. Ice cores from Greenland have suggested that much of that water must have come from the Antarctic. To find out just how fast sea levels rose at that time, Dutton is now looking at old corals in Mexico, Florida, and Australia; corals can be used to track sea level, since they grow in shallow waters to capture sunlight. A map of sea level rise around the world, and how it was higher in one place than another, could be used to infer where the water came from. Success isn’t guaranteed; corals are notoriously difficult to date. And whatever they find, notes Scambos, it will still be hard to draw a parallel to the modern world.
“That was a natural warming period in Earth’s history,” Scambos says. “We’re putting our pedal to the metal today; we’re driving the system very hard.”
James Hansen, a climatologist at Columbia University, summarized the evidence for rapid sea level rise in a recent controversial paper, raising some eyebrows at its stark warnings of catastrophe. Though many researchers have taken issue with the dramatic tone and specific details of that paper, its conclusion — that multi-meter sea level rise is possible in the next 50, 100, or 200 years — does not seem so alarmist in the face of other recent work.
“I think a lot of us who work on paleo records are all aware that a lot of change can happen very quickly — I’m always looking at big numbers,” says Dutton, who hasn’t been startled by recent studies like DeConto’s. “It’s always going to be a difficult question to answer. Maybe we need to accept we’re always going to have this uncertainty and just prepare for the worst.”
Correction, May 5, 2016: An earlier version of this story incorrectly stated the amount of freshwater that is locked up in the polar ice caps. The article should have stated that 99 percent of the world’s freshwater ice is contained in the Antarctic and Greenland ice caps.
