For many scientists studying Antarctica, and particularly the vulnerable West Antarctic ice sheet, a major new study significantly increasing expectations for sea-level rise is the culmination of a large body of prior research - combined with alarming recent observations.
The study is based on an improved understanding of past warm eras in Earth's history that featured much higher seas. By creating advanced computer simulations of how Antarctica's ice melts and flows - ones that can accurately capture sea level during these eras - the current study was also able to project considerable sea-level rise in the relatively near future.
In effect, it found that we're about to start repeating the past.
The research stated that for a very high-carbon-emissions scenario, melting ice from Antarctica alone could cause seas to rise 1.14 meters (3.74 feet), give or take 36 centimeters, by 2100 -- and vastly more, as much as 15 meters (over 50 feet), by 2500. The world has recently taken steps to try to avoid such a high-emissions pathway -- steps whose effectiveness remains to be seen - but even in a more moderate emissions scenario, the study found the Antarctic contribution could be 58 centimeters (nearly two feet), give or take 28 centimeters, and close to six meters (nearly 20 feet) by 2500.
That would be in addition to contributions from melting mountain glaciers, swelling seas caused by warming itself, and losses from Greenland. The smaller ice sheet in Greenland is melting even faster than Antarctica and contains up to six meters (20 feet) of potential sea-level rise, including about two meters in vulnerable areas where marine-based ice is in contact with the ocean.
So how did scientists reach this conclusion? Through a series of key steps that weave together modern observations and a better understanding of the ancient Earth.
Alarm about West Antarctica
The first major sign that consensus projections about sea-level rise might be too conservative came in 2014, when two groups of scientists published research suggesting that the marine-based glaciers of West Antarctica had begun an irreversible retreat, as a result of warm water reaching the bases of glaciers such as the enormous Thwaites (which is larger than Pennsylvania, and over a mile thick in places) and melting them from below.
Thwaites and other West Antarctic glaciers not only are perched deep below the sea surface but also rest on downward-sloping seabeds, meaning that as they move downhill, they will present taller and thicker fronts to the ocean, and ice could flow out of them still faster. Researchers have dubbed this condition a "Marine Ice Sheet Instability," and other recent research has suggested that similar alignments exist in the far larger ice sheet of East Antarctica, as well.
In light of these observations, the community of West Antarctic scientists has recently called for an urgent international research mission to the remote and inaccessible region to gather new data about Thwaites glacier, in particular. The reason is that there is still a lack of understanding of much about Thwaites, including the precise topography of the seafloor beneath it. Better mapping of this terrain could validate, or potentially refute, some of the results of the most recent computer modeling study, where a major retreat of Thwaites does indeed drive a key part of the result.
The new study, said Robin Bell, an Antarctic expert at Columbia University's Lamont-Doherty Earth Observatory, "points a laser pointer at where we need to go" to do more research.
"Right now we do not know if the topography under the Thwaites Glacier is a smooth ramp sloping back to some of the thickest ice in West Antarctica or if it is a lumpy terrain," said Bell. "We just do not know, and it matters for models like this."
The troubling message from the Earth's past
At the same time, eyeing places like Thwaites and fearing the potential for far greater sea level rise than currently forecast by groups like the United Nations's Intergovernmental Panel on Climate Change, scientists have been closely examining the ancient past for clues about what the Earth, and its ice sheets, are actually capable of.
For instance, the last interglacial period (sometimes called the Eemian), between 130,000 and 115,000 years ago, featured a sea level high stand that is believed to have been 6 to 9 meters above current levels. The Eemian has drawn major attention lately because events at that time, including claims of enormous waves moving very large boulders in the Bahamas, have fueled the prominent climate researcher James Hansen's particularly dire ideas about polar melt, rising seas, the shutdown of ocean circulations, and worsening storms.
The problem has been that computer models that simulate Antarctica in different climates have, until lately, been unable to reproduce these past sea level high stands. "We were not [able] to get more than a meter of sea level rise out of the ice sheet in the last interglacial," said the University of Massachusetts-Amherst's Rob DeConto, who authored the new study in Nature with David Pollard of Penn State University.
But what about the way ice moves and changes was missing from models?
The key message from Greenland
Even as the scientists were struggling to get the equations to work, real world events were hinting at a possible solution.
Take, for instance, Jakobshavn glacier of Greenland, which may be the fastest retreating major glacier in the world, and one that holds the potential for 0.6 meters (nearly 2 feet) of sea level rise, according to the University of California-Irvine glaciologist Eric Rignot. Jakobshavn's flow speed increased greatly after the loss of its floating ice shelf, an extension of the glacier out over open water in the fjord where it lies, in the early 2000s.
Ice shelves provide stability because they tend to freeze onto landscape features, such as islands, the sides of fjords, or the seafloor - and thus, provide back-pressure on glaciers, holding them in place. However, at Jakobshavn, or at the Larsen B ice shelf on the Antarctic peninsula - part of which dramatically disintegrated in 2002 - ice shelves are vulnerable to collapse. Warm waters can eat into them from below, and at the same time, surface melting can lead to pools of water that fill crevasses on these shelves and weaken them, widening surface cracks.
After losing its ice shelf, Jakobshavn now terminates in a tall, sheer ice cliff extending above the surface of the water, with the vast bulk of the glacier below the water. There is no stabilizing shelf any longer. Another Greenland glacier, Helheim, is in a similar state (see image above). And Antarctica's Crane glacier, which was previously held back by the Larsen B ice shelf, is as well.
But this presents a problem. What these glaciers have in common is that they have quite tall cliffs above the water, approaching 100 meters in height. In each case, below the water is much more ice - potentially nine times as much. But that ice is partly held in place by the ocean itself. It's the part above the water that matters, because there are reasons to believe that ice, not being very strong as a material, can't maintain a cliff above around 100 meters in height without collapse.
"With the fantastic observations of Jakobshavn, Helheim and other Greenland cliffs, the whole picture started coming together," said Richard Alley, a glaciologist at Penn State University who has published with DeConto and Pollard, of the revelations regarding ice cliff collapse.
Indeed, one key implication of the research is that, as more major glaciers lose ice shelves, we could be moving into a new world where the dynamics of ice cliffs matter more and more. Another major Greenland glacier, Zachariae Isstrom in the northeast, has also now lost the vast majority of its ice shelf and features a 75-meter-high cliff, scientists reported last year. As the glacier retreats further, "the height of the calving cliff will increase from its current 75 m to enhance the risk of ice fracture," they noted.
West Antarctica's glaciers could create much bigger cliffs
Greenland may seem vast, but it's actually far smaller than Antarctica and contains much less ice. The reason this matters is that in Antarctica, and most immediately West Antarctica, there is the possibility to create ice cliffs much taller than 100 meters high (above sea level), because we are talking about marine-based glaciers with far more than 1,000 meters (or 1 kilometer) of thickness overall in their bulkiest parts.
"For this effect to be important, the ice would have to be about a kilometer thick," said Knut Christianson, a glaciologist at the University of Washington in Seattle. "In some places, it gets to be significantly deeper than that. The interior of West Antarctica, some areas are more than 2 kilometers below sea level, more than 3 kilometers thick."
Thus, in the new ice sheet model, two new processes are included -- "hydrofracture," in which ice shelves collapse due to meltwater on their surfaces, and "cliff collapse," in which too-tall ice cliffs also fail and crumble. Meanwhile, the model also runs so-called "ensembles," in which large numbers of simulations with slightly altered physics are tested out -- but only model runs capable of accurately simulating sea level rise for past eras like the Eemian are ultimately included.
Add it all together - the past, and the alarming present -- and sure enough, scientific understanding produces the possibility of major ice retreat from Antarctica, in this century and beyond. "Finally coming up with a combination of physics based on basic observations, sound pretty fundamental physics I think, made all the difference in being able to recreate ancient sea levels," said DeConto.
And it's important to emphasize that there are reasons to think that the real world could possibly outstrip even this study.
"Their model is not a worst case, as they do place a limit on the rate of retreat in the model that does not come directly from the physics," said Alley. In an interview, DeConto explained that the model does not allow Antarctica's ice to retreat as quickly as Jakobshavn's ice currently is retreating, for instance.
And there's another potentially missing factor. Large ice loss from Antarctica would pour enormous amounts of freshwater into the Southern Ocean, in the form of both water but also vast icebergs that will begin to melt. A much noted, but controversial, climate change catastrophe study by former NASA researcher James Hansen used the concept of large meltwater injections like this to outline a feedback mechanism that leads to still more ice loss - meltwater stratifies the polar ocean with cold water on top and warmer water below. The warmer water then causes more melting of ice sheets.
The new study finds that the ocean would be gaining, in some extreme scenarios, more than a sverdrup of fresh water as melting really advances, which translates into a flow of about 84 billion tons daily. "The amount of meltwater going into the Southern Ocean in these scenarios is more than ocean modelers even think about when they think about doing a crazy freshwater forcing experiment," said DeConto.
In this scenario, not only would seas rise rapidly -- but there could be other major effects. No wonder, then, that the new research, by squaring the past with the present, raises deep and troubling new questions.