The Totten hots up
A major Antarctic glacier is becoming dangerously unstable
December 2016. The Twin Otter aircraft drops into the polar void, skis skidding over wind-carved corrugations of hard snow, and pulls up to a shuddering halt.
The pilots have landed as close as they dare to the crevasse fields of the East Antarctic coast. Their location is due south of Albany, Western Australia, and there’s nothing between here and there but ice, water, and the storied latitudes of the roaring forties, the furious fifties and the screaming sixties.
This day, the elements are uncharacteristically obliging. The wind is still, the sunshine punishingly bright as it ricochets off the white. The pilots and a team of expeditioners from the Australian Antarctic Division – a glaciologist, an engineer, a field officer – climb out onto the colossal Totten glacier.
They’re adrift upon a frozen river, an imperceptibly oozing stream. A glacier is a solid cascade of ice, the definitive immovable object submitting to the irresistible force of gravity. Totten glacier is one of Earth’s mightiest, and the single largest in East Antarctica.
Outlet glaciers such as Totten are the valves regulating the flow of polar ice into the oceans. They are the machinery behind seesawing global sea levels over deep time. Totten holds back a frozen catchment the size of Spain. Within it is enough ice to raise sea levels by 3.5 metres, if the glacier opened to let it drain away.
Throughout the geologically brief climate sweet spot that human civilisation has enjoyed, spared the orbital meanderings that have spun our planet in and out of ice ages, Totten’s flow has hummed obligingly in neutral. Put simply, it has maintained its “glacial mass balance”, whereby snowfall coming in and ice flowing out equals a negligible shift in sea levels.
Now it appears Totten has slipped into gear. The grounded ice is thinning and has been for decades. Vast regions are identified as “fundamentally unstable”. Its “grounding line” – the junction of ice, bedrock and ocean where the glacial river spills off the land and starts to float – has retreated up to 3 kilometres in 17 years. Is this a consequence of human climate change, or the stirring of natural variability that has shifted it through deep history? Might it be some confluence of both? It’s more than idle curiosity that has brought the team all this way to find out.
The expedition, led by glaciologist Ben Galton-Fenzi, is the first to get this close to the grounding line of the glacier, where the force of Totten’s flow rips open yawning fractures in the ice. They want to pop the hood and try to figure out how it ticks. On one side, inland, is the gargantuan East Antarctic ice sheet, where snowfall welded over aeons is piled 4 kilometres deep on bedrock. On the other, stretching to the coast, is the Totten ice shelf, where the glacier floats until it fractures and sets sail as icebergs.
“The ice mass … has passed from ice dome to sheet ice to glacier ice to shelf ice … to the diminutions of the bergs, cycle by cycle, like the gears of an ice orrery,” explained American historian Stephen J Pyne in his book The Ice. All that may change in this inexorable process is the speed.
Pyne cautioned that to enter Antarctica was “to be drawn into a slow maelstrom of ice”. Which is precisely where Galton-Fenzi wants to go. His mission is to install instruments that will measure Totten’s innermost workings. He and his fellow travellers assemble a sturdy structure 6 metres high, augering holes in the ice to secure its legs. Onto its frame they fix a GPS unit that will log the rate at which the ice is moving and any tiny shift in surface elevation, as well as an instrument that can see clear through to the ocean or bedrock to measure the ice depth within millimetres. The topography deep below – the interface of ice, rock and sediment – determines how Totten might slide.
Updates on the glacier’s vital statistics will be transmitted every 90 minutes. Scientists will compare the glacier’s pace and condition with nearby ocean temperatures. When Galton-Fenzi returns next December to collect the databanks holding the mother lode of observations, he expects that snowfall will have engulfed the structure.
This is the sixth, and final, of these structures that the team leaves on Totten. It’s a hard slog in the Antarctic high summer, their boots churning a bog of snow. GoPro footage shows them stripping off layers of too-warm kit. They leave a frenzy of footprints that still show up in the pictures from space.
But their efforts will capture a treasure trove of observations the satellites could never pick up, and well outside the reach of long-haul expeditioners who have tried to fathom Totten for the past half-century.
Galton-Fenzi’s project will, ultimately, “allow us to provide a risk assessment of the likelihood of the collapse of the Totten over the next few hundred years”.
This intelligence comes not a moment too soon. Yet maybe decades too late.
Neal Young is one of only a handful of people who have ever traversed the deep reaches of Totten. He was a 23-year-old mathematician when recruited to the Australian Antarctic Division’s (AAD) expanding glaciology program in 1970. Humans had been to the moon and back, yet understood little about our “cryosphere”, the frozen parts of Earth. Fast-evolving radar and satellite technologies, and the background noise of concern around greenhouse gas–induced climate warming, were powering a resolve to change that.
Two years earlier, in 1968, a US-based glaciologist, John Mercer, whose eccentricity perhaps undercut his prescience (legend has it he preferred doing his field work nude), published a paper spelling out how the structure of glacial systems could make the polar ice caps vulnerable. His focus was the West Antarctic ice sheet. He described how ice shelves – where glaciers flow off the bedrock and float on the ocean – were corking up deep reservoirs of ice. If these shelves disintegrated under a slight warming, the whole sheet could flow into the ocean, with catastrophic implications for sea levels. The theory, published in an obscure journal, didn’t get much traction.
This was a messy, inglorious era for climate science. The prospect of global warming from burning fossil fuels had been flagged some 75 years earlier in 1896, and rising levels of CO2 in the atmosphere had been tracked for a decade, but some scientists were still predicting a period of global cooling.
There was, Young recalls when we meet by the docks in Hobart, the hub of the AAD, “a lot going on”.
In the spring of 1971, after spending a winter at Casey station, Young and fellow expeditioners rode in a cavalcade of Mad Max–esque vehicles to the small Antarctic ice cap of Law Dome, 125 kilometres distant. They pegged out survey lines across glacial streams running off the summit. The plan was to create a network of surveys that would eventually follow these flow lines all the way to Russia’s Vostok station, 1500 kilometres inland, and skirting along the coast. If they could capture observations on the speed and patterns of ice flow off the sheet, and the rate of snowfall coming in, they could determine whether the ice was in “mass balance” or out of whack.
That season they didn’t get far, but in the summer of 1976–77 Young joined a team of Soviet glaciologists on an epic field expedition, returning to complete the network in ’77–78. They clocked up nine months and 5000 kilometres of traversing behind a monster ice-adapted Russian tank.
In temperatures reaching minus 60 degrees Celsius, they dug holes and planted poles across Totten and neighbouring glaciers, and logged the poles’ locations using Doppler satellite receivers, revolutionary technology at the time. “You need to sit at each spot for a couple of days,” Young explains. “So though the measurements were accurate, progress was slow.” When they went back the following summer, they found these markers and logged their new positions. The comparison provided some of the first clues about the motion of the ice sheet deep in the interior.
En route, Young picked up a Muscovite accent and plotted the last leg to a remote ice cap known as Dome C using only a theodolite, his watch and the stars. He keeps a photograph he took of the “little black specks on the horizon”, the huts of an itinerantly occupied field station marking the end of the journey. “And I still have a little bottle of cognac”, a gift from the previous tenants.
After all that, it’s perhaps anticlimactic to observe that his most significant discovery came 25 years later at his desk in Hobart, while studying satellite data. Young is widely recognised by his peers as the person who sounded the alarm on the stirrings within Totten glacier.
After his adventures with the Russians, Young returned to Hobart to direct the survey program he had started out of Casey. He trained and dispatched hardy bands of surveyors into the remote field. Their focus was upstream of Totten’s grounding line – they didn’t want to lose expeditioners or kit down a crevasse. “We were trying to come up with flow laws that specify how the ice sheet behaves.” As well as measuring the glacier’s speed they examined its stretch and strain.
By this time – the 1980s – scientists were theorising that global warming would destabilise polar ice and raise sea levels enough to affect populated coastal regions, but they disagreed about why, when and how much. The dynamics of ice “turned out to be, like most things geophysical, a complicated snarl of influences”, observed American science historian Spencer R Weart.
Nonetheless a review of glaciology’s state of play in 2002 observed that the Greenland ice sheet, the second-largest body of ice in the world after Antarctica’s, was losing mass by near-coastal thinning. Also, satellite images were revealing dramatic ice sheet retreats in West Antarctica. Because much of that sheet was grounded below sea level, there was growing concern about its exposure to warming seas. “Perhaps the most important finding of the past 20 years has been the rapidity with which substantial changes can occur on polar ice sheets,” wrote glaciologists Eric Rignot and Robert H Thomas.
The Antarctic Peninsula – the tentacle that kicks up towards South America – was also cooking. Temperatures had spiked by 2.5 °C over 50 years. Big changes were evident, none more profound than the moment in 2002 when the massive Larsen B ice shelf fractured and vanished. No one had anticipated such sudden disintegration. It shattered any notion that science had a grip on the dynamics at play within the ice, and sent dazed modellers back to their drawing boards.
Nevertheless, the behemoth East Antarctic ice sheet – bigger, higher and colder than the one in the west – was widely assumed to be inviolable, safely anchored on elevated bedrock.
In January 2003, NASA launched a satellite that was able to measure Antarctic ice elevations. Neal Young pulled down the readings it was sending back about Totten. “I compared observations of the surface with what we got from our field surveys [the last of which was in 1987],” Young recalls. “And I went, ‘Oh!’ It hits you in the eye.” In some areas, Totten’s elevation had shrunk by 10 metres in 16 years.
He and others suspected that, as in West Antarctica, warm water from the ocean was getting under the Totten ice shelf, increasing the rate of melting at its base. As the ice shelf changed, the flow on the lower reaches of the glacier sped up. The ice that was piled up behind the grounding line thinned as it joined the chase.
It’s the loss of grounded ice that is the worry. For all the drama of an ice shelf collapse, it doesn’t matter in terms of sea level because it has come from the floating part of the glacier, so it is already in the global seas bucket. What matters is what happens next, when grounded ice exploits the weakness in the dyke.
When Young blew the whistle on Totten’s dramatic decline, there was still dispute around whether the rise in sea levels observed across the 20th century was due to thermal expansion (water fattening up as it warms) or ice melting. But as scientists crunched the new satellite data, they discovered that the planet’s coasts had changed profoundly, that sea-level rise was accelerating and that, while ice sheets had likely only been a minor contributor so far, they were the greatest looming concern.
They desperately needed to know what was happening under the ice.
Splinters, the bar at Casey, is a scene so beardy and retro veneer-and-plastic you’d swear you were deep in urban hipster territory. In December 2009, I found myself there, perched alongside some Arctic ring-ins – a life-size cardboard polar bear and a pair of Canadian pilots stranded by lousy weather. They were working for the Investigating the Cryospheric Evolution of the Central Antarctic Plate project, with the apposite acronym ICECAP. Then in its second year, ICECAP would evolve into a revelatory collaboration of Australian, American, British and French glaciologists.
Every Antarctic summer the ICECAP team climbs aboard its airborne laboratory, a modified version of the workhorse DC-3, and navigates across sweeping grids of unexplored East Antarctica. Bit by bit – using ice-penetrating radar, laser sounder, magnetometer and gravimeter – they have assembled a detailed map of the landscape beneath the ice.
The pilots introduced me to Duncan Young, a Kiwi-born, Texas-based radio-glaciologist. He fetched a ream of readings, unfurled it across one of the tables set for Christmas dinner, pinning the ends with beer glasses, and began to explain the fine lines that traced buried mountain ranges and what he was reading between those lines.
Was this landscape configured in a way that would make the East Antarctic ice sheet unstable? The smothered coastline of bedrock, a fragment of the supercontinent Gondwana, could have the capacity to influence the world above. Its valleys and ranges and canyons, the interface of ice and rock and seawater, all determine how the glacier behaves.
“The problem with ice,” Young said, “is that it wants to flow. Part of what we’re trying to do is to get enough surveys around the Totten to figure out how sensitive it might be to feedbacks.”
In the ’60s and ’70s, ground expeditions and airborne surveys had identified a subglacial basin, the Aurora, nestled behind Totten. ICECAP revealed its true dimensions, mapping a huge rock bowl full of ice sitting below sea level. ICECAP also discovered lowlands connecting it to the coast. This caused a sensation, and scuppered all assumptions that the eastern ice sheet was sitting up out of harm’s way.
In 2011, Duncan Young and his colleagues published a landmark paper mapping mountain ranges and coastal fjords like those in New Zealand or Norway. Sections of the rock under Totten dipped more than 1 kilometre below today’s sea level, making it especially vulnerable if the surrounding ocean found a way in. That publication set off a continuing avalanche of papers drawing on the ICECAP data, and ushered in a new era in understanding East Antarctica, says AAD glaciologist Tas van Ommen, one of the architects of ICECAP.
“And Neal [Young] was spot on [about] the fact that there was an area where the surface was going down about 2 metres a year, a large change in the Totten.
“But what we have discovered since is that it’s really dynamic. When Neal was looking, [Totten] was going down a couple of metres a year. Since then it has gone up a bit. Overall it is losing ice, but in a really complicated way where it speeds up, slows down, thins a bit, thickens a bit.”
In 2014, two teams of scientists scrutinising the West Antarctic ice sheet announced its “unstoppable” disintegration, the hastening retreat of its glaciers, and that 3-metre sea-level rises in coming centuries are likely unavoidable. John Mercer (the naked glaciologist) was posthumously cited as prophetic.
The next shock was a 2015 paper co-authored by van Ommen, exposing two deep troughs in the seabed beneath the Totten ice shelf. This showed it could be vulnerable to similar changes to those playing out in the west. Warm seawater could travel along these to penetrate deep under the shelf. It wasn’t proof that the warm water was actually there, but it offered a mechanism that could explain Totten’s thinning. And with so much potential extra sea level looming behind, it sounded a warning: “coastal processes in this area could have global consequences”.
Glaciology moved into overdrive. Another group scooped up every satellite image they could find, and discovered that between 1989 and 2015, except for one slow year, more ice was flowing out to sea through Totten than was accumulating over the inland basin.
The next scientific scoop on Totten will be a study by ICECAP’s lead investigator, glaciologist Jason Roberts, revealing that when the glacier speeds up and slows down it correlates with whether the ocean temperatures off the front are warm or cold. “While it’s only circumstantial, when you put it together, you start to build up a picture in which the Totten glacier is really quite sensitive to the amount of ocean heat,” says van Ommen.
But what if the bigger worry was not deep under the ice, but hiding in plain sight? This theory argues that melting of the surface ice may, “in the long term, be the real killer”, says van Ommen. The scenario is “hydrofracture”, where meltwater pooled on the surface slips down crevasses, cracking the shelf like a splitter through a log. This was almost certainly responsible for the destruction of the Larsen B ice shelf in 2002. It could destroy ice shelves and hasten glacial flow much faster than melting at the base.
The idea got a big kick with a paper published in March 2016, in which American modellers Rob DeConto and David Pollard concluded hydrofracture and ice shelf failure in Antarctica could add 1 metre of sea level rise by 2100, and more than 15 metres by 2500 if greenhouse emissions continue unabated.
“The concern that [the authors] raised is that if the temperature goes up even a small amount, a degree or so, do you actually change the amount of meltwater by huge amounts?” says van Ommen. “This hydrofracture process might start to become quite important to rapidly removing ice shelves.”
There’s dispute about DeConto and Pollard’s modelling, the detail if not the concept. What worries van Ommen and others is that it does provide an explanation for the high-water marks etched in Earth’s paleo record. “If you look at the last time the planet was a degree or so warmer than now, just before the last Ice Age, about 120,000 years ago, sea level was 6 to 9 metres higher.” In other eras it has been way higher, rising in the warmth of the middle Pliocene, around 3 million years ago, by as much as 30 metres.
Where does that water come from? Estimated losses from Greenland and West Antarctica plus heat expansion of the ocean only account for around 6 metres. It is now widely accepted that the missing water “is going to be in East Antarctica”, says van Ommen.
On New Year’s Day, 2015, the research icebreaker Aurora Australis sailed where no vessel has sailed before, right to the sheer white cliffs where Totten’s ice shelf fractures and falls into the Southern Ocean.
“Looking at [satellite] ice imagery before we headed into the region, I thought we were stuffed,” recalls Steve Rintoul, CSIRO oceanographer and voyage leader. Having rammed through 60 nautical miles of sea ice, the Aurora Australis became stuck.
Then Rintoul saw small cracks opening up in the sea ice. “The wind blew from an unusual direction, the south-west. That released the pressure, and all the ice shifted north a little and opened up this channel that we could follow directly to the front of the Totten glacier.” Happy New Year.
Up close, the 80-metre cliffs were magnificent and terrifying. Here the glacial stream progresses at top speed, more than 2 kilometres per year at the front. “Because the glacier is very fast-moving, it’s very corrugated and crevassed and broken up, so when it reaches the snout it is peeling off in continuous generations of icebergs and shards of ice.”
Rintoul and his team couldn’t linger. “You have to sneak in and sneak out again before the door slams shut and the ice shifts back, shutting off that little narrow pathway to the glacier.” After consulting the oracles of weather and satellite images, the captain gave the scientists 48 hours.
They trailed instruments over the side, fishing for temperature, salinity and oxygen, watching the sonar ping back the shape of the sea floor. They worked through the night, which, at this time of year, is a luminous mess of sunrise and sunset. Below where they were working, the sea floor dropped abruptly from about 500 metres to more than 1 kilometre. Rintoul fed an instrument into the water. “The temperatures were at about freezing point, not very interesting – cold, and cold, and cold. And then suddenly … there’s this jump of about a degree and a half within a few tens of metres, which in this part of the world is a big jump.”
They’d found a layer of warm water coursing along the bottom of this canyon, sliding under the glacier. Two years later, it’s in the journals – direct evidence, and validation, of ideas and models evolved over decades.
The Totten region is a hot spot, and Rintoul says “the destabilising factor is warm water getting in under the ice shelves”.
That’s not the whole story. There are other factors, Rintoul says. There are natural rhythms to glacial systems, and these are barely understood. Ben Galton-Fenzi stresses that warm water is very likely a driver of Totten’s system in its natural state, oiling its flow as it has through deep history. The question is whether the water is getting warmer, and why. Understanding and modelling the natural systems is critical in determining the implications of human interference.
“Climate change is warming the Southern Ocean,” says Rintoul. “Warmer waters and changes in circulation that bring more heat to the glaciers will drive thinning, [which] will lead to more ice flowing off Antarctica and an increase in sea levels.
“The models tell us the decisions we make in the next few decades will commit us to a future of multiple metres of sea-level rise – or not … This issue of commitment is really critical.”
Last year, Western Australian geophysicist Alan Aitken trawled through the ICECAP data, analysing the wear and tear left on the bedrock as Totten moved back and forth over deep time. In periods of rapid retreat a glacier doesn’t leave much of a scar in its wake, but if it gets stuck grinding away in one spot, it scours deep into the rock.
Aitken discovered that Totten has a couple of comfort points along a well-worn track. One of them is about 150 kilometres inside today’s grounding line, and the next one is way back, more than 350 kilometres inland. If the glacier retreats to the first line, the ice lost to the ocean would add almost 1 metre to sea levels. If it gets over the ridge and slides to the backline, then that’s another 2 metres.
“We understand how the system works,” says Aitken. “What we don’t understand is what temperatures and conditions it takes to make it really go, and the timescale of operations. We haven’t seen temperatures like the ones we are going to get for 3.5 million years. Past performance isn’t a guide to future probability. We are writing a whole new set of rules in my view.” And just because it hasn’t changed in a long time “doesn’t mean it can’t change incredibly quickly”.
The current modelling shows that Totten’s system plays a moderate part in the first 4 or 5 metres of sea-level rise coming out of Antarctica – maybe 50 centimetres of a 5-metre rise. Beyond that, from 5 to 20 metres, Aitken says that Totten’s contributions become bigger and bigger, but across timescales of thousands of years. “It’s the kick in the tail – the fat end of the wedge.”
This work, says van Ommen – a co-author of the Aitken paper – gives powerful incentive to pin down what the thresholds are, and how they fit with the ambition laid out in the Paris Agreement to cut emissions enough to hold warming to 2 °C. Looking at the paleo record, he suspects even that’s too much.
If we do discover too late that we’ve overshot the emissions boundaries required to prevent large East Antarctic losses, we’d be relying on “some way of sucking CO2 out of the atmosphere later this century,” says van Ommen. “That would be an expensive and difficult exercise, but if people knew it would save 2, 3 or 6 metres of sea-level rise, that might be the information, the motivation, they need.”
The challenge for science is peeling back the caveats of uncertainty. Records from Antarctica are still so sparse, and so short. “In nearly everything we measure, the amount of year-to-year variability is so large it is really difficult for us to put our finger on it and say it is really anthropogenic warming causing it.”
Perhaps the changes occurring in Totten will be pinned down to natural cycles, but van Ommen says that may still be cold comfort. “We might just be a little unlucky, that this thing has always advanced and retreated, and we’ve come along and kicked it when it is in a retreating mode, when it is at its weakest, driving it back into the deeper interior.”
In January this year, the Aurora Australis tried once more to push through the summer sea ice and reach Totten. The ship couldn’t find a way through, but before she turned for the home run to Hobart the scientists on board measured temperatures as close as they could to the coast and found a stream of warm water deep down near the floor. The 2015 results were no fluke.
They left Totten in peace, if not in slumber, for the long, dark Antarctic winter night.