For scientists studying Thwaites glacier on the West Antarctic Ice Sheet, life is as remote as it gets. On a still, cloudy day, it’s white in every direction, completely silent, and you can only get a sense of where you are from looking at undulations of the otherwise featureless surface.
“There’s nothing else around for thousands of kilometres,” says Dr Peter Davis, a researcher with the British Antarctic Survey (BAS), who has spent five seasons in the field in Antarctica. “No aircraft, no cars – nothing.”
Just getting to Thwaites can take a month. From the UK, it’s a commercial flight to New Zealand, then three more flights, the last one from an Antarctic camp called WAIS Divide to what’s known as the ‘deep field’ atop the glacier.
Researchers must plan months and years ahead, packing larger items of scientific equipment into shipping containers, and carrying with them only the smaller, more fragile instruments and what they need to survive – including tents and camp stoves.
An expedition like this could mean three months away from home. But what scientists are learning here is vital to the fate of our planet.
As a spectacle of nature, Thwaites is immense – a vast ice flow, like a frozen river, that drains an area the size of Britain as it slides towards the ocean.
However, the so-called ‘Doomsday Glacier’ is also unstable – losing ice faster than it accumulates – meaning the whole thing will eventually melt.
Researchers want to get a better handle on when that’s going to happen, and what it means for sea-level rise.
What a melted Thwaites would mean for coastal cities
For the last 80 years, Thwaites has been losing more water through melting than it’s been gaining in snow.
“It’s all about the balance and the balance has shifted,” says Davis’s colleague at BAS, Dr Kaitlin Naughten.
Today, Thwaites is responsible for about 4 per cent of all sea-level rise. At 120km (75 miles) wide and 2km (1.2 miles) thick, it’s such a monster that its complete meltdown will unleash enough water to put half a metre (1.6ft) on global sea levels all on its own – though, of course, other polar ice caps are defrosting at the same time.
Half a metre of sea-level rise would submerge large parts of Asia’s coastal cities including Manila and Bangkok, as well as sizeable chunks of the Netherlands and the east of England. It’s also half of the sea-level rise needed to begin flooding Manhattan.
With 230 million people around the world living on land less than 1m (3.3ft) above high tide lines, the flooding from Thwaites would be enough to displace vast populations (Bangkok alone has 11.4 million residents).
And whilst we expect mass migrations in the coming centuries, Thwaites suddenly lurching towards collapse could give coastal regions little time to adapt.
For the last few decades, field scientists like Davis and modelling specialists like Naughten have been collaborating to make sense of a host of different factors that could influence the Doomsday glacier’s demise.
Meanwhile, other researchers have proposed epic engineering schemes to protect it from warming ocean currents or to freeze it in its tracks as it slips into the Amundsen Sea.
But can we save Thwaites in time? Here’s what scientists know about how long we have before the Doomsday glacier collapses – and some of the bizarre plans to delay its tipping point.
Melting ice from Thwaites and West Antarctica
The science of Antarctica’s ice sheets and glaciers is relatively new. Even in the early 1990s, when the first Intergovernmental Panel on Climate Change reports were released, there were suggestions Antarctica could gain snow and ice over the coming century.
But this thinking soon shifted when scientists started looking at the ice sheets from a different angle – above.
“That’s when the main satellites that we use went up,” says University of Edinburgh’s Prof Noel Gourmelen, one of a team of polar scientists who have been monitoring the mass balance of the world’s ice sheets since 1992.
From the team’s analysis of satellite observations, we now know that West Antarctica has been losing an average of 82 billion tonnes of ice each year for the last three decades.
Over this time, the rate at which Thwaites is losing ice has doubled. And we’ve watched in dismay as Antarctic ice shelves – floating platforms of ice that extend from glaciers and hem them in at the front – have crumbled into the sea, their disappearance speeding up the melting.
The ice shelves buttressing Thwaites are already cracking and thinning, and that’s one of the reasons scientists are so worried about it. But Thwaites itself is also on dodgy territory.
Most of the west side of the Antarctic ice sheet is partly submerged, making it more vulnerable to melting than the east side.
“The ice sheet is so thick and so heavy that, over time, it has pushed the bedrock underneath,” explains Prof Mathieu Morlighem, who studies glaciers at Dartmouth College, in the US.
Thwaites, in fact, teeters on the edge of a deep seabed basin, like a lid on the rim of a saucepan. Warm ocean water can get under this ‘lid’, despite it being secured by ridges in the seabed near where the grounded ice meets the open sea.
And, with the rapid retreat of the front of the glacier, these holds on the rim will inevitably be lost. When that happens, Thwaites will sink backwards, deeper into the basin, where it will shrink faster than ever.
We don't know when Thwaites will collapse
In 2014, Morlighem’s work with colleagues at the University of California, Irvine, in the US, confirmed that Thwaites had lost 14km (8.7 miles) from its front in a single decade. At the same time, another study concluded that ‘early stage collapse’ had already begun.
However, there remains some uncertainty around the exact timeframes for this collapse – which could play out more slowly than the wording might suggest – and the factors that could hasten it.
There’s lots we still don’t know when it comes to Thwaites. A big uncertainty is how icebergs will form from a glacier that thick once its supporting ice shelves break up.
Icebergs ‘calve’ (break off) periodically from the front of glaciers, but some scientists think Thwaites is so big that each ice cliff exposed by this process would be too tall to stay upright for long. In other words, the whole ice flow could go down like a row of dominoes.
In this scenario, sea-level rise from Antarctica has been predicted by some scientists to hit a metre (3.3ft) by 2100 – the level at which Manhattan begins to flood – although others think it might be closer to 0.5m (1.6ft).
According to Morlighem, however, the dominoes scenario is unlikely. His modelling instead shows the ice sheet thinning rapidly following the loss of its shelves, meaning the glacier face wouldn’t reach the critical height for rapid calving.
Read more:
- Mysterious 'surge' under Earth crust could reshape world map, study claims
- Scientists have found what sounds eerily like Atlantis off Australia's coast
- Can humans live under the sea? This live experiment plans to find out
What lies beneath the glacier
There’s also a mystery hidden in deep lakes below the glacier’s base. These lakes are buried under kilometres of ice, but they occasionally drain out, a process that satellites can spot via elevation changes at the surface.
Gourmelen’s team recently suggested that when this happened in 2013, it led to faster melting under Thwaites’ ice shelves.
As he explains, the Amundsen Sea is warmer near the bottom. So, when lighter, more buoyant, fresh water draining from the lakes escapes under Thwaites’ partly submerged front, it rises as a plume under the ice shelf, pumping the warmer seawater upwards.
“The turbulence and the heat essentially increase the ability of the ocean to melt ice,” says Gourmelen. Whether another event like this could catastrophically destabilise the entire glacier may depend on exactly where the lake water drains out.
Thwaites also has other features scientists haven’t seen until recently, and which could be important to melting timelines.
Davis, for example, spent the 2019–20 Antarctic summer season getting a first look at the underside of the ice, where it transitions from grounded glacier to floating ice shelf – a critical region called the ‘grounding zone’ that is exposed to the open sea.
His team worked with a couple of hefty pieces of equipment that had to be dragged in by tractors and sledges. One: a hot water drill for boring through nearly 600m (0.4 miles) of ice. And two: a remotely-operated underwater vehicle dubbed ‘Icefin’ that they sent in through the borehole.
Before this, Davis says, they had “no idea” what the underside of the ice looked like or how to accurately represent it in BAS’s models. Smooth? Crevassed? They didn’t know.
But with Icefin’s cameras, they were able to see that, far from being smooth, it was an upside-down landscape of deep crevasses, large flat terraces, and smaller scallops or dimples, like those on a golf ball.
A gloomy forecast of rapid ice sheet melting
They now know how and where Thwaites might melt – but it doesn’t change the Doomsday glacier’s gloomy forecast.
Indeed, Naughten’s work presents a future for West Antarctica that looks bleak whichever way you paint it.
According to her 2023 models, carbon emissions already in the atmosphere have now committed us to a path of rapid ice sheet melting and sea-level rise that will differ very little, regardless of how much we curb emissions in the next century.
“Because there’s such a long response time of the ice sheet, even after the world stops warming, sea-level rise will continue for centuries,” she explains, noting that with other impacts of climate change, like flooding caused by intense rainfall and drought, putting an end to warming would have an almost immediate stabilising effect.
But what does this mean in terms of how fast Thwaites will collapse? Naughten says she “can’t put exact numbers on it,” but that the melting is going to speed up.
Potential geoengineering solutions to slow the melting
So, everything we know suggests that Thwaites is going down, even if we don’t know exactly how, or when.
But whilst some grimly accept Thwaites’ demise in the next couple of centuries, others are unwilling to write it off just yet.
Prof Brent Minchew, a geophysicist at Massachusetts Institute of Technology, in the US, just launched the Arête Glacier Initiative, a non-profit organisation that funds research to resolve “deep uncertainties” in existing models and forecasts, whilst at the same time developing interventions to prevent total meltdown in the near future.
Marrying traditional physics methods with new machine learning models, what they learn will feed directly into the design of interventions intended to, for example, increase the drag to slow or even stop the ice flow.
His suggestion? Freezing the glacier to its bed. This scheme would use ‘thermosyphons’, devices that could draw heat away from the base to stop the melting and sliding via a series of pipes dropped through 2km (1.2 miles) deep drill holes. (Thermosyphons have been used previously to stabilise permafrost around oil pipelines.)
However, Arête itself is more “agnostic” about solutions, Minchew notes, being set up to fund an array of potential interventions.
Glacial interventions – also known as geoengineering schemes – have been viewed with scepticism by some scientists due to the logistical challenges of working in the Antarctic, as well as the costs.
One such scheme, championed by glaciologists Prof John Moore and Dr Michael Wolovick, is a giant seabed ‘curtain’ designed to block the warmer waters driving melting deep down, near Thwaites’ grounding zone.
They think that a curtain, attached to the seabed, with the ability to bend thanks to overlapping panels would allow it to slide under breakaway icebergs.
The authors estimate the cost for an 80km (48 mile)-long curtain at around $40-80bn plus maintenance (approx. £30-65bn), which they emphasise is “much lower” than the cost of adapting the globe’s coastlines.
Finances aside, it’s the choice between these two impossible problems – stabilising an ice block the size of a country or orchestrating mass migrations to deal with sea-level rise – that has Minchew convinced he’s choosing the right path.
“If you look at both of these challenges in isolation… dealing with sea-level rise and populations at coastlines around the world is closer to impossible,” he says.
“So I think we actually have to know what the intervention ideas are, and if it does turn out to be virtually impossible, then we need to know that. Let’s not just assume that’s the case, because the lives of tens of millions of people are going to be affected by sea-level rise.”
In late 2023, two workshops on ‘glacial climate interventions’ led by proponents including Minchew, Moore and Wolovick drew around 50 scientists, engineers and students to Stanford University and the University of Chicago.
Morlighem was one of the Chicago participants; he went with an open mind, despite being unconvinced such interventions would work. “I try to be neutral,” he says.
“It feels like something that will be investigated by societies going forward and I’d rather be involved in the research and make sure things are done properly – rather than implementing something that could be potentially harmful.”
A sense that Thwaites' days are now numbered
Naughten is a little more sceptical, not just because of the practicalities of large-scale construction projects in Antarctica, but because Thwaites is “only one source of sea-level rise”.
To zoom out for a moment and absorb the bigger picture, the entire Antarctic and Greenland ice sheets are also currently melting. The Greenland ice sheet is responsible for two-thirds of the sea-level rise we’ve seen since 1992.
“So how much will it improve things versus how much it costs?” says Naughten, who’s also concerned about the cost of other climate change adaptations, including dealing with drought and wildfire.
Whatever the promise of glacial intervention, there’s a strong sense that Thwaites’ days are now numbered. Even Minchew admits that interventions can’t provide a final solution – they’re about “giving the world the chance” to adapt at a steadier pace.
For those working in the field, there’s still a bittersweet opportunity to witness Thwaites in all its splendour. “It’s an enormous privilege to go to these places,” Davis says.
“I can’t possibly come to work every day feeling doom and frustration... It’s not that I don’t feel that, but you have to disconnect.”
About our experts
Dr Peter Davis is a researcher and physical oceanographer with the British Antarctic Survey, who has spent five seasons in the field in Antarctica. He is published in various scientific journals including Geophysical Research Letters, Nature and Journal of Geophysical Research: Oceans.
Dr Kaitlin Naughten is a researcher and ocean modeller with the British Antarctic Survey. She has been published in the likes of Antarctica and the Earth System, EGUsphere and The Cryosphere.
Prof Noel Gourmelen is a polar scientist and personal chair of Earth observation at the University of Edinburgh, in the UK. His work is published in Nature Communications, Nature and EGUsphere, to name a few scientific journals.
Prof Mathieu Morlighem is an Earth sciences professor who studies glaciers at Dartmouth College, in the US. He has been published in scientific journals such as Proceedings of the National Academy of Sciences, Geophysical Research Letters and Nature Geoscience.
Prof Brent Minchew is an associate professor and chair of geophysics at Massachusetts Institute of Technology, in the US. He is published in the likes of IEEE Transactions on Geoscience and Remote Sensing, Geophysical Research letters and Bulletin.
Read more: