The world’s thickest glaciers may “calve” icebergs up to three times faster than previously predicted. However, such rapid retreat may not be sustained over long periods.

Instead, the large bodies of ice – which include Antarctica’s Thwaites glacier – may collapse in bursts, with intervening periods of stability.

The behaviour of tall ice cliffs is one of the biggest sources of uncertainty in predicting future sea level rise. Several of the world’s largest glaciers in Greenland and Antarctica extend out to sea, with their bases resting on the sea floor. Their coastal edges can form ice cliffs looming as high as 100 metres above the ocean’s surface.

For years, researchers have feared that such marine-based glaciers might be vulnerable to runaway cliff collapse. The idea here is that each time an ice cliff collapses, it exposes a new, taller – and so potentially even less stable – ice cliff further towards the glacier’s interior. The loss of ice due to runaway collapse of these ever-taller cliffs could , according to an Intergovernmental Panel on Climate Change report. But we don’t yet have enough empirical data to know how likely it is that runaway cliff collapse will occur, so previous studies have relied on ideal mathematical models to envision the process instead.

at the Massachusetts Institute of Technology-Woods Hole Oceanographic Institution Joint Program and her colleagues used satellite data to provide the first direct, large-scale observations of how the tallest ice cliffs actually behave, offering a firmer basis for estimating the danger of runaway collapse – and the rapid sea level rise that could result.

According to the study, the risk of runaway collapse depends on two main factors: how steeply the sea floor beneath the glacial ice deepens, and how quickly the ice flows seawards. Imagine an up-ended jar of thick honey spilling onto a plate: it won’t remain in a tall, stable shape, but will thin outwards at the edges. In a similar way, if the viscous flow of ice is fast enough, it will thin out the edge of a glacier, reducing cliff height. Even if the glacier is thicker further towards its interior, sufficiently rapid ice flow appears able to compensate, lowering those ice cliffs and preventing a runaway effect.

This idea has been suggested by idealised mathematical models, but had never been systematically measured in real glaciers. “None of the glaciers we observed underwent unstable retreat, even though they did retreat much faster than our current models predict,” says Needell.

Free newsletter

Sign up to The Earth Edition

Sign up to newsletter

That retreat could speed up even further in the future. Needell notes that most glaciers in the study are “buttressed” by sea-floor features. For instance, a series of underwater ridges currently help stabilise Thwaites from beneath.

“Over the next decades, it’s going to jump from one pinning ridge to another,” almost like “tripping” downhill, says at the University of St Andrews in the UK. Eventually, Thwaites will reach a point where those ridges disappear. “The bed then begins to get real deep,” says Benn. At that point, which the glacier may reach within the current century, “the whole structural integrity of the ice becomes unviable”.

The ‘doomsday’ glacier’s giant ice shelf is about to break away

The floating ice shelf of world’s widest glacier – Thwaites glacier in Antarctica – is detaching, with worrying implications for global sea-level rise

That is when runaway collapse becomes a real possibility. However, those conditions may also speed up the flow of ice, thus thinning out any looming cliffs. The key question, says Needell, is whether thinning can outcompete rapid ice loss, restabilising the glacier. Her study suggests that it might – but the question remains open.

at the British Antarctic Survey points out that we haven’t yet seen this kind of runaway cliff collapse. One study suggests that it may have happened in the vicinity of Thwaites around 12,000 years ago. “The problem is, nobody was around to take observations at the time,” he says. “It’s a possibility we have to keep our eye on.”

Bubbles in a 3-million-year-old ice core have allowed researchers to measure gases in the atmosphere during the late Pliocene for the first time.

This epoch, when global temperatures are thought to have been around 1°C warmer than today and sea levels were as much as 25 meters higher, is often taken as a cautionary tale for our own time. But the levels of carbon dioxide and methane in the bubbles are much lower than today’s, which might mean Earth’s climate is more sensitive than we thought to small atmospheric changes.

In some parts of Antarctica, snow accumulates each year on top of the last year’s snow and gets squeezed into layers of ice containing air bubbles, giving us a continuous record of the past atmosphere. Last year, the Beyond EPICA group extracted the oldest continuous ice record so far,

But scientists have also extracted even older ice in places where younger ice has been eroded by the wind and older “blue” ice is close to the surface.

of Oregon State University and her colleagues used this ancient ice from Allan Hills to offer the first direct measurement of carbon dioxide and methane in the Pliocene’s atmosphere. The team found unexpectedly low levels of both gases around 3 million years ago: 250 parts per million for carbon dioxide, and 507 parts per billion for methane. During the subsequent cooling period, the team found a small decrease in carbon dioxide and no noticeable changes in methane.

The results are significantly lower than previous estimates based on indirect measurements, which had indicated carbon dioxide levels at 400 parts per million, closer to today. There are no indirect ways to measure past methane, which today hovers just under .

Free newsletter

Sign up to The Earth Edition

Sign up to newsletter

“We definitely were a bit surprised,” says Marks-Peterson. If correct, the findings may suggest that even small changes to greenhouse gas levels could trigger major shifts in climate. “Maybe the Earth system is even more sensitive to changes in CO2 than we have understood,” she says. “That’s a little bit of a scary thought and something that I would say that our record can’t answer yet.”

“Ultimately, any new data that suggests Pliocene CO2 levels were lower than previously expected means future climate change might be worse than previously expected,” says at the University of Illinois Urbana-Champaign, who was not involved in the study.

But more work is needed before we can use the data to inform climate models, cautions at Victoria University of Wellington in New Zealand. “It’s way too early to throw the baby out with the bathwater,” he says. Naish wants to see the record carried back further in time, when the Pliocene was even warmer.

, of the European Center for Research and Education in Environmental Geosciences, says he is confident that the study’s low CO2 values are correct. However, he points out that distortion of ancient ice makes it impossible to know whether the low values represent a cold period, a warm period, or an accurate average. “It’s representative of something. We need to know what it’s representative of. Because it doesn’t come with its own little temperature estimate, and it certainly doesn’t come with a global temperature estimate,” he explains.

The team hopes to improve the reliability of their analysis by comparing their results with those of the continuous ice core from the Beyond EPICA group. “That will help us better understand our own record,” says Marks-Peterson. “We are all on pins and needles waiting to hear what they find.”