Warnings in the water

Lifeforms on the verge of destruction

IT IS A muggy, unseasonably warm day in March and I am standing at the back of a building in the Australian Antarctic Division’s (AAD) headquarters on the outskirts of Hobart. I am here with Dr So Kawaguchi and Rob King. An ecologist and one of the world’s foremost experts on Antarctic krill, Kawaguchi leads AAD’s krill research projects. King is a biologist and marine research facility specialist, and for the past eight years has been the biology lead in the design process for the division’s new icebreaker, the RSV Nuyina. The two of them are about to show me one of the AAD’s most prized ­possessions: living Antarctic krill.

The two men lead me into the lab, one wall of which holds a rack of 1.5-metre-­high plastic tubes. They are lit from behind and filled with bubbling water and bright-­green or dirty-­orange algae. As we pass, Kawaguchi points to one that contains algae bred from samples collected in a saline lake near Australia’s Davis Station in the early 1980s and sustained as a continuous culture ever since.

At the far end of the room, King opens a door and ushers me into a second, smaller space that contains three round tanks a metre or so across and about the same height, their white shapes wrapped in black insulation. A collection of hoses connect the tanks to pipes on the wall behind them; the sound of water moving through them and the thrum of the pumps suffuses the space. The illumination is low, almost greenish; when I ask about the lighting, King tells me it replicates the daily light cycle in Antarctic waters.

We approach the middle and largest tank. The water – refrigerated to just half a degree above zero – is so cold you can feel the chill rising off the surface. Yet beneath its surface swarm thousands of krill. These are sub-­adults: probably a year or two in age, and slightly shorter than my little finger. They look a bit like prawns, except longer and leaner. Save for a translucent dusting of red along their bodies, which fades from a rusty hue at the front to a brighter red along the segmented abdomen and tail, they are mostly transparent, their most prominent features their large, jet-­black eyes and the digestive gland behind it. On those that have fed recently the latter is bright green, distended with algae that’s being digested.

Despite their small size they do not dart or flicker in the manner of schooling fish. Instead, they move like tiny boats or torpedoes, the rippling motion of their back legs – their pleopods – pushing them through the water with surprising speed. Most of them also seem to be moving in formation, following the same path around the tank, swimming steadily against the current created by the pump, while those not swimming in formation twist and circle, feeding on floating algae.

In the next tank are a smaller number of adult krill. At about five centimetres in length, they are longer and larger than the sub-­adults, as well as less brightly coloured, the red bled through their bodies like selenium in glass. In contrast to the krill in the first tank only a few are swimming in formation, but nonetheless they move quickly and purposefully through the water.

Beside me Kawaguchi and King gaze into the tanks, their affection and fascination for their charges palpable. And as the minutes pass, I find I understand: there is something hypnotic about these tiny creatures, their beauty and their pressing numbers, the way they dance beneath the water.

Yet the krill are not here because they are wondrous. They are here because there is no easy way to study them in the hostile and icy waters of the Antarctic. And it is imperative we understand them as best we can, because their world is changing with frightening implications not just for the krill, but for the entire Antarctic ecosystem.


ANTARCTIC KRILL ARE crustaceans, but despite their appearance they are not a variety of prawn or shrimp. Instead, along with the eighty-­four other species of krill, they form their own order, the Euphausiacea, which is separated from prawns and shrimp and their relatives, lobsters and crabs, by 100 million years of evolution.

They are also distinguished from other crustaceans by the sheer scale of their population. Although estimates vary, there is little question Antarctic krill are the most abundant wild animals on the planet, with a total biomass of somewhere between 300 and 500 million tonnes. Only two other species – humans and cattle – comes close (again, estimates vary, but the nearly eight billion humans on Earth weigh approximately 600 million tonnes, and the total biomass of every cow on earth is slightly higher again). Indeed, the biomass of Antarctic krill far outsrips the biomass of every living wild mammal, bird and reptile combined (approximately ninety million tonnes). If these figures are so large as to feel meaningless, the number of individual krill – somewhere in the region of 800 trillion, although this also varies from year to year and season to season – is even more so.

While this vast population is distributed throughout the Southern Ocean and at all depths – krill have been sighted 4,500 metres down on the sea floor – as much as half of it concentrates into huge swarms of billions of individuals that gather near the surface. The largest of these swarms, which can extend for kilometres and contain as many as 60,000 krill per cubic metre, are so immense they are visible from space and so vast and complex they are effectively a form of super-­organism in their own right. Video taken within these swarms shows the massing krill as clouds of innumerable individuals that swirl and dance, like moths around a lamp, the aggregation of bodies so dense it reduces the light from above to a dull red glow.

The krill in the tanks at the AAD were caught off the coast of East Antarctica, somewhere between Australia’s Davis and Casey stations, but 70 per cent of the world’s krill population is concentrated in the region below the South-­West Atlantic, an expanse of ocean that takes in the ice-­bound Weddell Sea and stretches from the Drake Passage in the west to Queen Maud Land in the east. These waters, and in particular those off the Antarctic Peninsula and around the string of islands scattered from its tip to the South Georgia and the South Sandwich islands, seem to be attractive to the krill because their geography suits the krill’s life cycle, which is closely synchronised to the annual ebb and flow of the sea ice and the solar cycle.

As the ice begins to retreat in the spring months, it exposes the ocean to the sun’s radiation and so triggers vast blooms of phytoplankton in the water’s surface layers.

These blooms attract the krill, which arrive to gorge themselves on the plankton. Meanwhile the bodies of the females begin to swell, distended by up to 10,000 tiny eggs. As the eggs mature the female krill migrate away from the coast, massing in the transition zones where the sea floor drops away into the deep ocean – which is where the males fertilise the eggs.

Once the eggs are fertilised, the females expel them to drift slowly downwards until they are somewhere between 700 to 1,000 metres below the surface. Here they hatch and krill larvae emerge. These larvae look nothing like their adult counterparts: instead they are mouthless and resemble some sort of four-­legged flea. Now able to move under their own power, the larvae begin to ascend once more; as they rise to the surface over the next few weeks they pass through several more stages of development, gradually acquiring more legs and, finally, a mouth.

However their return to the sunlit zone brings new challenges. During their ascent the larval krill have been sustained by the nutrients contained in their egg. But by the time they reach the upper layers of the ocean that energy supply has been exhausted. This means it is critical they find food as soon as possible. Yet the larvae have arrived back near the surface just as the Antarctic is hastening into winter, bringing to an end the bountiful supply of phytoplankton that sustained the adult krill only a few months earlier.

The krill’s solution to this problem is ingenious. Although it is customary to associate Antarctica with stark, white cleanliness, the underside of the sea ice is home to a thriving ecosystem of ice-­algae and other phytoplankton and microbes, all drawing sustenance from the light leaking through from above. Hidden beneath the rapidly expanding sea ice, the larval krill feed on this film of algae, moving across the underside of the ice in a zigzag motion as they rake food from the ice with specially adapted bristles on the tips of their thoracic legs. ‘They’re like the combine harvesters of the ocean,’ says King, before pointing out that despite their focus on algae they’re also quite capable of gobbling up any small animal foolish enough to get in their way. ‘If combine harvesters did the same that would make it very dangerous walking around a farm during harvest,’ he laughs.

The krill pass the winter in this strange, inverted world, protected from predators by the cracks and runnels in the ice’s underside. But, as the summer returns and the ice begins to break up, they are forced to abandon this protection and move back into open water. It is here that they begin to congregate into swarms, gathering in larger and larger numbers.

This swarming is fundamental to krill behaviour, although the precise reasons behind it are a matter of debate. What is certain is that it’s at least partly about safety in numbers: as the aggregations grow the odds of any particular krill being eaten decline. But the advantage of gathering into groups in this way also extends beyond a simple numbers game: by swimming in formation the krill confuse the visual systems of their predators, who find it difficult to isolate and attack particular krill amid so many bodies moving in unison.


JUVENILE KRILL DO not become sexually mature until their second summer – but as the sea ice expands outwards again at the end of this second summer, both they and the adult krill vanish beneath it once more. King admits the question of exactly where the animals go remains a mystery: ‘Where do 400 or 500 million tonnes of krill go each winter?’ he asks. ‘Do they sit under the sea ice and eat the tiny amount of plankton they can find there? Or do they dive down to the sea floor and eat the detritus from the summer blooms and the faecal pellets that have drifted down? Or do they attack the benthic community for whatever food they can get?’

Both by virtue of their ubiquity and their sheer weight of numbers, krill are the cornerstone of the Antarctic ecosystem. Penguins, seals and baleen whales, such as humpbacks and blue whales, all depend upon them for survival, as do the fish and squid that sustain predators such as sperm whales. This is a very short food chain: the microscopic phytoplankton transform the energy of the sun and the nutrients they absorb into chemical energy, which the krill in turn consume and transform into energy able to be consumed and transformed once more by penguins, seals and whales. The efficiency of this unusually truncated food web is visible in the bodies of animals at its top, which are startlingly dense with stored energy: as much as 30 per cent of the body mass of some penguins can be fat, while in Weddell seals this percentage can be even higher.

Nowhere is this efficiency more evident than in the bodies and life cycles of the great whales. Blue whales are the largest animals that have ever lived, growing to up to thirty metres in length and reaching a weight of more than 180 tonnes – up to half of which can be fat (more, in fact, than any other mammal). A female blue whale feeds for half the year but is able to produce up to 180 litres of milk a day: this has nine times the fat content of human milk and provides such concentrated sustenance that her calf can gain as much as ninety kilograms a day and reach sexual maturity at just five years of age. The milk of humpbacks is even richer – with a fat content of up to 50 per cent it is closer to cream in consistency – and females produce as much as 350 litres a day.

This efficiency also lies behind much of the horrifying massacre of Antarctic mammals and birds during the nineteenth and twentieth centuries. When James Cook landed in the Kerguelen Islands in 1776, he found seals ‘so fearless we killed as ma[n]y as we chose to make oil for our lamps and other uses’. Within a decade, sealers had begun to take advantage of this fearlessness in the Kerguelens and elsewhere, resulting in a bloodbath that saw at least seven million fur seals die in the first three decades of the nineteenth century. Although much of this first wave of brutality was focused on fur seals, elephant seals were also killed in their millions for their blubber, which was boiled down to make oil. Even penguins were not spared: in the 1880s and 1890s vast numbers of king penguins were butchered on Macquarie Island, their bodies rendered down for oil in specially designed digesters.

Yet nothing compares to the carnage inflicted upon whales and their societies. Although the difficulties of whaling in the remote and hostile southern latitudes meant industrial whaling did not really commence until the early twentieth century, that changed with the establishment of the first whale-­processing station on South Georgia in 1904. Initially, attention was concentrated on humpbacks, which were relatively slow and easy to catch. But as diesel-­powered ships began to replace sail, and explosive harpoons took over from mere metal, whalers began to pursue larger and faster whales. This saw the slaughter of almost 30,000 blue whales a year by the end of the 1920s; as the population of blues began to collapse, whalers transferred their attentions to fin whales, leading to catches in excess of 25,000 a year by the 1950s, at which point they moved onto sei whales and then, finally, minke whales. In all, more than two million whales were killed in the southern hemisphere in the twentieth century.


WHALE OIL WAS originally used in lamps and lighthouses – it was prized for burning clean and bright, although cheaper grades tended to produce a revolting odour. Yet by the twentieth century, whale oil’s most important use was in food. In 1929 chemists working for Britain’s Lever Brothers and the Dutch company Margarine Unie discovered how to harden whale oil into fat – and thus manufacture margarine. A merger of the two companies created the company we now know as Unilever, who used their market power to control the oil market so successfully that by 1935 they were purchasing 84 per cent of the world’s whale oil. (This was used interchangeably with other oils such as palm and groundnut oils, as it occasionally left a ‘fishy taste’.)

Nor were the effects of this grotesque destruction limited to whales and their societies. Although pre-­whaling whale populations consumed huge amounts of krill – close to half a billion tonnes a year – the collapse in whale numbers did not lead to an explosion in krill. In fact the opposite occurred: the krill population actually fell by as much as 80 per cent. This happened because whales are not simply passive consumers. Instead, in a reminder of the intricate webs of ecological connection and energy exchange that underpin ecosystems, the iron-­rich whale excrement provides vital nutrients for the phytoplankton upon which the krill depend; without that phytoplankton numbers crashed, and so did krill populations.

The environmental significance of krill extends beyond their role as food. They also play a vital part in the processes that regulate the Earth’s climate. Like all plants, algae and phytoplankton absorb carbon dioxide. As the krill consume the algae they absorb it in their turn, either fixing it in their bodies or expelling it in faecal pellets that sink downwards, locking it away in deeper waters. This process absorbs as much as twelve billion tonnes of carbon dioxide a year, approximately two thirds of the amount absorbed annually by the world’s forests.

Yet while it is easy to only think of the krill as agglomerations and statistics, they are also creatures in their own right. When I ask Kawaguchi whether he ever wonders what the world must be like for a krill, his face lights up. ‘Often!’ he says, laughing.

For Kawaguchi this is a question with a practical dimension, because one of the benchmarks of how well they are doing in the aquarium is whether or not they display normal schooling behaviours. Developing tanks that encourage schooling has been surprisingly challenging: not only will the krill not school in a black tank, they won’t school in a clear tank. They only school when kept in a white tank with no direct light.

Kawaguchi believes this is because it resembles the diffuse light and white boundaries of the world beneath the ice. But it is also because in clear-­sided tanks the glass becomes reflective from within: ‘The krill are very visual animals and use their eyesight to form schools; because they’re always watching each other, being able to see their reflection confuses them, so they just retreat to the corner and hide.’

For Kawaguchi these challenges underline the degree to which the krill’s sociability underpins its success as a species. ‘One hypothesis is that the krill form schools because it’s energy efficient, and swimming and turning in unison is hydrodynamically effective. But they also seem to be able to sense and respond to the movement of their neighbours in the swarm through changes in the water. What that shows is they’re really social animals, and it’s that sociability that has allowed them to become the key species in the Southern Ocean.’

This sociability is equally evident in the krill’s tendency to become stressed when separated from each other: if isolated from their schoolmates their hearts beat faster, just as they do when exposed to threats such as whales. But the complexity of the swarming behaviour underpinned by the krill’s sociability also means scientists are really studying two things: the behaviour of the individual krill and their aggregate behaviour, which exhibits a form of collective behaviour, rather like a beehive or an ant colony.

‘The swarms are a whole other animal,’ says King. ‘And when you’re trying to understand how they’re interacting with their ecosystem, or how they’ll respond to fishing, then understanding how krill behave as that super-­organism is just as important as understanding the behaviour of the individual krill that underpins it.’ This is especially obvious in interactions with predators such as whales. ‘When a whale attacks a swarm you get a whale-­sized hole in it – not because the whale has eaten all the individual krill, but because the krill are using tail flicks to attempt to escape, and those tail flicks warn neighbouring krill, who also attempt to escape and so on, and the process of those tail flicks radiating outwards means the super-­organism changes shape to avoid the whale.’

These massive concentrations of krill do not just make them attractive to predators such as whales and penguins. Over the past fifty years a growing fleet of ships has begun to target the krill directly. Most of the krill caught by these ships is processed into feed for aquaculture or used to meet the sky­rocketing demand for krill oil as a dietary supplement.

For some years now, the body charged with managing the krill fishery, the Commission for the Conservation of Antarctic Marine Living Resources (CCAMLR), has set the allowable catch in the South-­West Atlantic sector at 5.6 million tonnes, or 1–2 per cent of the krill’s total biomass, with further limits imposed in specific areas in order to prevent overfishing of particular regions. Until recently catches have remained well below these levels, but in the past few years they have been growing rapidly: in 2020 the reported catch was just over 450,000 tonnes, an increase of more than 60,000 tonnes, or 15 per cent on 2019, and more than double the catch in 2010. And this upward trend looks likely to continue: South Korea and Russia are both committed to expanding their krill industries, while China, which has already more than doubled its catch since 2019, recently commissioned the world’s largest krill trawler, and Norwegian company Aker BioMarine, which in 2020 accounted for nearly half the total krill catch, is also expanding its fleet.

Concerns about increasing pressures from fishing have been sharpened by research suggesting krill populations in the South-­West Atlantic Sector have declined by 70–80 per cent since the 1970s, probably due to changes in sea ice. Kawaguchi, who works closely with CCAMLR, is wary of such figures: while he acknowledges krill populations may have declined and the range of krill habitat may have shifted southward across that period, he argues the data is not robust enough to be so definitive. Instead he believes the real threat of fishing is not to the krill but to the animal communities that depend upon them. ‘When they’re raising young, land-based predators such as penguins and seals are restricted to shorter journeys of a day or so. That means their breeding colonies are positioned very close to where the krill aggregates. The problem is that those regions of reliable production are also very attractive to fishing vessels, and if they keep returning to them there is the potential they will deprive the penguins and seals of the food they need to rear their young, which could cause a significant and irreversible effect on that particular population or ecosystem.’

Kawaguchi believes these potential threats can be managed effectively by ensuring fishing is avoided in vulnerable areas, or at times when penguin and seal populations are particularly reliant upon an abundance of krill, and points to CCAMLR’s continuing efforts to assess these risks in order to develop a framework capable of protecting not just the krill but the entire Antarctic ecosystem.


YET IN MANY ways the real challenge the krill face is not fishing but global heating. Because the krill’s life cycle is so closely synchronised to the solar cycle, krill are extremely vulnerable to changes in the timing of the sea ice’s retreat and return. Without abundant supplies of phytoplankton at the right time, female krill grow at a slower rate, produce fewer eggs and reduce the success of the spawning season. Likewise the larval krill must find sea ice, and therefore food, within ten days of their return from deeper waters, or starve.

Until 2015 the sea ice seemed to be resisting the warming that has affected the rest of the Antarctic. Yet since 2015 the total area covered by sea ice around Antarctica has declined rapidly. In 2017 and 2018 both the winter maximum and summer minimum coverage were the lowest on record, and although levels recovered slightly in subsequent years, the ice started to retreat a month earlier than usual in 2021, so that by February this year sea ice had declined to even lower levels than in 2017, setting a new record minimum and dropping below two million square kilometres for the first time. ‘We do now have to consider whether or not the system is starting to change,’ said one scientist.

This has the potential to have drastic effects upon krill populations. One 2016 study found that the combined effects of warming water and changes in sea ice coverage could lead to a reduction in krill habitat of as much as 80 per cent. More recently research from the University of Tasmania’s Institute for Marine and Antarctic Studies has shown that without significant reductions in emissions the krill’s habitat will contract southwards and deteriorate significantly. There is also evidence that the population of salps – free-­floating, gelatinous animals sometimes known as sea grapes, or sea snot, which also feed on phytoplankton and form huge aggregations like krill – are increasing.

These changes may already be affecting animals that rely on krill. Humpback whales spend the summer feeding in the Antarctic, gorging themselves on krill before migrating thousands of kilometres northwards into warmer waters to calve and rear their young. Despite the huge distance they travel the whales do not eat during their migration north and south – instead they survive on the fat they build up during summer, sometimes losing up to half their body weight across the course of their journey. The demands of this process, which often pushes the whales to the edge of their physical endurance, means it is vital they be able to locate abundant krill easily and reliably during the summer.

It is significant, therefore, that in recent years whales arriving on Australia’s east coast have been in poorer condition, with lower levels of body fat and higher concentrations of organic pollutants in their blubber. This was particularly pronounced in 2017, when low levels of sea ice and the break up of the Larsen C Ice Shelf on the Antarctic Peninsula resulted in wide areas of open water where there would usually be ice, and in 2011 when an unusually strong La Niña caused sea ice to retreat further than normal.

The effect of this year’s record low levels of sea ice on the whales will not be visible until this winter. But observations of whales in Hervey Bay last year, when the sea ice began to retreat a month earlier than normal, suggest the whales arrived late and left the resting ground early. Susan Bengtson Nash from Griffith University’s School of Environment and Science heads up the Humpback Whale Sentinel Program, which co-­ordinates long-­term, standardised biomonitoring of humpback whales migrating along Australia’s east and west coasts to New Caledonia, Ecuador and Brazil. ‘Last year was certainly a very different year,’ she says. ‘Even before we went into the field, we had the tourist operators in Hervey Bay asking where the whales were, because they hadn’t arrived. And although they did finally turn up, they then left early as well. And the reports from New Caledonia and Brazil were similar: the whales arrived late and then disappeared very quickly.’ Disturbingly these difficult years also seem to be associated with sharp upticks in the number of whales stranding themselves: in 2021, 230 humpbacks stranded themselves in Brazil, almost double the previous record set in 2017. ‘The question now is whether we’re going to see these events occur with greater frequency,’ says Bengtson Nash. ‘And if we do, will they still be interspersed with bumper years that somehow counter the poorer years?’

Nor are rising temperatures the only challenge facing the krill. As ocean waters absorb more carbon dioxide from the atmosphere, seawater is acidifying. This is especially true in polar regions, where the cooler water absorbs carbon dioxide more efficiently. For animals with calcium carbonate shells or internal structures – such as molluscs, crustaceans and corals – this can have serious effects on growth and development, or in extreme cases result in their shells simply dissolving.

Research suggests adult krill are mostly unaffected by the sorts of increases in ocean pH expected in coming centuries. But experiments on embryonic krill exposed to the levels of acidity likely to be experienced in deeper waters by 2100 showed they failed to develop, prompting fears of a tipping point beyond which krill populations could almost entirely collapse. According to Kawaguchi and his colleagues this could have ‘catastrophic consequences for dependent marine mammals and birds of the Southern Ocean’.

These effects would extend beyond the Antarctic and its ecosystems: given the krill’s significance to the Earth’s carbon cycle, a decline in their population would create a feedback effect, further reducing the ocean’s ability to absorb carbon dioxide – and amplifying the effects of existing heating.

‘The problem is that it isn’t just warmer water, or changes in sea ice, or acidification or even organic pollutants,’ says King. ‘It’s the synergistic effect of all of them at once.’

Kawaguchi agrees. ‘The krill are experiencing an environment that they’ve never experienced before, and that’s very difficult to comprehend.’ But he believes that when change arrives, it may arrive abruptly. ‘These ecosystems are operating in a very fine balance. That means that once things change, there might well be a major shift in the structure of the system.’


THAT SHIFT IN the structure of the system may already be underway. In mid-March of this year Antarctica experienced an unprecedented heatwave, with some locations registering temperatures more than 40 degrees above normal, and the continent as a whole a staggering 8.6 degrees warmer than normal. Although part of a broader trend – over the past fifty years temperatures on the Antarctic Peninsula have risen by almost three degrees, almost triple the global average – this was far outside the bounds of anything experienced before.

‘Antarctic climatology has been rewritten,’ said one scientist, who described the temperature anomalies as ‘unthinkable’. Another said it ‘upended our expectations about the Antarctic climate system’. A week later came reports East Antarctica’s Conger Ice Shelf had abruptly collapsed, an almost 1,200-­square-kilometre structure completely disintegrating without warning in a day or two. The cause of the collapse is not yet clear, but it seems likely the heatwave in combination with the effects of warming water beneath the shelf pushed it past a tipping point and its integrity simply gave way.

The Conger is not the first ice shelf to break up in recent years: over the past two-­and-­a-­half decades large sections of the Larsen Ice Shelf – a vast structure that once extended out into the Weddell Sea from the Antarctic Peninsula’s eastern flank – have collapsed. The northernmost section, Larsen A, broke up in 1995; this was followed by the much larger Larsen B in early 2002, a process that saw almost two thirds of the ice sheet – 3,300 square kilometres of ice some 220 metres thick – break away into the sea. Finally, in January 2022, much of what remained of Larsen B broke up in a few days, releasing yet more ice into the ocean.

Meanwhile the even bigger section known as Larsen C has been growing increasingly unstable. The fourth-­largest ice shelf remaining in Antarctica, Larsen C covered some 50,000 square kilometres – at least until 2016, when scientists observed a huge crack, more than 100 kilometres long, along its seaward side. Within months the entire section, which accounted for about a tenth of Larsen C’s total area, had broken away to form an immense iceberg 175 kilometres long, fifty kilometres wide and 200 metres thick.

At least for now, what remains of Larsen C seems to be held in place by two distinct ice rises, or areas where the ice comes into contact with the sea floor. But unprecedentedly early melting in 2019 and 2020 that resulted in large bodies of meltwater on the shelf has given rise to concerns the entire structure might suddenly destabilise in the same way as Larsen B or the Conger.

The behaviour of Antarctica’s ice is both extremely simple and incredibly complex. Its ice sheets, almost five kilometres thick in some places and containing 70 per cent of the world’s fresh water, are constantly replenished by falling snow; as the snow builds up it gradually compacts into ice, which then flows downwards towards the ocean in the great rivers of ice we call glaciers. Where these meet the water they gradually melt from beneath, resulting in floating ice shelfs like Larsen and the Conger, which project outwards over the ocean for tens or even hundreds of kilometres, before finally beginning to break up.

The speed at which glaciers flow depends upon a host of variables, ranging from the slope beneath them to the presence of obstacles and barriers such as ridges and valleys beneath the ice. Yet in Antarctica one of the most significant constraints on the movement of the glaciers – and therefore the rate of ice loss – is the existence of ice shelfs, which act as barriers, slowing the flow of ice behind them. As they break up the glaciers behind them flow faster, increasing ice loss and pushing sea levels rapidly higher: following the collapse of Larsen B, some of the glaciers behind it sped up sixfold.

The effects of this process are already visible. As ice shelfs retreat and glaciers melt, Antarctica is losing ice at almost five times the rate it was in the 1990s, with ice loss rising from approximately fifty billion tonnes a year to more than 200 billion tonnes a year. And this process is accelerating: in the five years between 2012 and 2017 alone the rate of loss tripled.

At Pine Island Bay in West Antarctica, where the Thwaites Glacier flows into the Amundsen Sea, there is growing evidence this process may herald a much larger transformation. Sometimes dubbed the Doomsday Glacier, Thwaites is a vast river of ice, 120 kilometres wide and three times the size of Tasmania. It is so huge that if it were to collapse entirely, it would raise sea levels by sixty-­five centimetres.

Even without the hastening effects of global heating, Thwaites is highly dynamic, flowing oceanward at an astonishing two kilometres a year. Until very recently its eastern side has been constrained by a massive ice shelf, which obstructs that end of its flow, and is in turn restrained by a pinning point where the sea floor rises high enough to hold the ice shelf in place. Yet changes in ocean circulation are causing warmer water to move in beneath the ice shelf; where this warmer water comes into contact with the bottom of the glacier the ice melts, causing Thwaites’ grounding line – the point where the ice on its underside parts company with the sea floor and begins to float – to retreat. This is hollowing the glacier out from below, weakening its structure.

These same processes are at work at many locations around Antarctica. And it only takes a quick glance at a topographical map to understand why they are so concerning. Although we are used to thinking of the continent as roughly circular, or teardrop-­shaped, such representations are deceptive, and much of what looks like land is nothing of the sort. Instead it is ice, kilometres thick, the base of which lies hundreds or even thousands of metres below sea level. This is especially true of West Antarctica, where the ice sheet sits in a vast basin bounded on one side by the Transantarctic Mountains and on the other by a scattering of what would otherwise be islands.

Thwaites sits on the seaward edge of this basin. Behind it the ground drops away rapidly, meaning that as the warm water eats away at the underside of the glacier it flows downhill, allowing it to spread further and further beneath the surface of the ice and destabilising the structure above.

The result of this is not yet clear. Some scientists speculate it might result in a phenomenon called marine ice-­cliff instability in which the glacier begins to crumble under its own weight, leading to a runaway collapse of the entire structure. What we do know is that as the ocean around Antarctica warms, the rate at which Thwaites’ grounding line is retreating is increasing. On the glacier’s eastern side, the rate of retreat has doubled from approximately 600 metres a year between 1992 and 2011 to 1.2 kilometres a year between 2012 and 2017. Even on the western side, where the rate of retreat remains relatively steady at between 600 and 800 metres a year, there are signs of dramatic change. In 2019 NASA researchers discovered a vast cavity, two thirds the size of Manhattan and approximately 300 metres high, below the glacier. Most of the fourteen billion tonnes of ice that would once have filled its space is believed to have melted in just three years.

Meanwhile the ice shelf that holds the glacier’s eastern section in check is also on the verge of collapse, its structure fatally weakened by a rapidly spreading network of fissures. In 2021 scientists predicted complete collapse was likely to occur within five years. Like the break up of Larsen B, that process is likely to be extremely rapid. American glaciologist Erin Pettit has compared the shelf to a windshield with a single crack in it. ‘You’re, like, I should get a new windshield. And one day, bang – there are a million other cracks there.’

Because the collapse of the West Antarctic would raise sea levels by over three metres, understanding its dynamics is vital. Yet observations of the behaviour of Antarctica’s ice only go back a few decades. As a result scientists have turned to the deep past, and in particular the period of warming that took place just under 130,000 years ago. In 2019, Chris Turney, Professor of Earth Science and Pro-Vice-­Chancellor Research at the University of Technology Sydney, led a team that collected blue ice deposits in the Patriot Hills, on the eastern edge of the West Antarctic Ice Sheet. Blue ice occurs in regions where fierce katabatic winds remove upper layers of snow and ice, allowing ancient ice to rise to the surface. The samples Turney’s team found revealed a break in the ice sheet record that coincided with the last interglacial, suggesting an almost complete melting of large parts of West Antarctica had occurred. ‘It doesn’t look like anything, there’s not a hole in it,’ Turney says. ‘It’s just that when you date it you realise there’s this massive gap in the record.’

More importantly, though, the presence of tiny shards of volcanic glass from the ice immediately before the gap allowed the team to pinpoint the date and speed at which the melting occurred. And this showed the collapse of the ice sheet occurred rapidly and at under 2 degrees of warming.

Zoë Thomas is a paleoclimatologist at the University of New South Wales and one of the researchers on the Patriot Hills study. She says the study has concerning implications for today. ‘The melting occurred well before temperatures peaked, so it wasn’t a situation where the ice melted gradually as the temperatures rose; it was a very fast response and it happened very early.’

It is likely this process of collapse is already underway. In 2021 some of the same scientists looked at the sediment deposited on the seafloor in what is known as Antarctica’s ‘iceberg alley’ since the end of the last glacial period 11,700 years ago. Examining these sediments, they identified eight periods in which the amount of sediment left by icebergs increased, each of which corresponded with a period of rapid sea-­level rise.

This demonstrated that these periods were caused by increased ice loss from the ice sheets. But it also revealed that these phases of rapid ice loss did not begin gradually. Instead the ice sheets seemed to flip from relative stability to rapid mass loss in less than a decade and, once triggered, continued to lose mass for hundreds of years. ‘The nature of tipping points is that once the thresholds are passed, you have all these positive feedbacks that accelerate the melt, and that means things happen really, really fast,’ says Thomas, who also sees ‘concerning parallels’ between the past and the present. ‘The data suggests we’re already past a tipping point. If we’re lucky it might be a small one that only results in sustained mass loss for a few decades, but that’s kind of irrelevant because any mass loss causes sea-level rise and any sea-­level rise is dangerous.’ Dr Michael E Weber, from the Institute of Geosciences at the University of Bonn and lead author on the paper, is even blunter, saying the findings are ‘consistent with a growing body of evidence suggesting the acceleration of Antarctic ice mass loss in recent decades may mark the beginning of a self-­sustaining and irreversible period of ice-­sheet retreat and substantial global sea-­level rise.’

Trying to get to grips with the idea of this kind of rapid and unstoppable change is not easy, but perhaps one way to grapple with its immensity can be found in the work of the MELT project (melting at Thwaites grounding zone and its control on sea-­level rise), which in the first weeks of 2020 created a 700-­metre-­deep shaft through the glacier and lowered a submersible into the water beneath.

A section of the resulting footage is on YouTube. In the top half of the image the bottom of the ice shelf moves past overhead, its dirty grey-­green flecked with pale flecks and fissures oddly reminiscent of videos of the Earth moving beneath an orbiting spacecraft. Between the camera and the ice a glassy layer in the water marks the threshold between the fresh meltwater above and the saltier seawater below; ice crystals and shards float along it, together with the drifting form of an anemone, its delicate tendrils resembling a strange, alien craft escaped from a science-­fiction film. As the submersible rises through the shimmer of the halocline, the bottom of the ice shelf comes suddenly, sharply into focus, its surface ridged and scalloped. The tiny forms of fish or other swimming things flick by out of range before some sort of crustacean darts past, its body bleached and overexposed in the light from the submersible. There are several seconds more of the upside-­down moonscape of this ice shelf’s underside, a final glimpse of another crustacean, and then it’s done.

Like anything from the beginning of 2020, this footage takes on a curious charge from its proximity to the first weeks of the pandemic, its innocence of what was to come. Yet it’s also charged with something else, something more. It is hauntingly beautiful, even without the piano music that accompanies it. And like the Blue Marble photo of Earth taken by the crew of Apollo 17, it is also affecting and profound, a window not just into the extraordinary ingenuity of the scientists but a glimpse of planetary change in action. In this place we see the inflection point of the climate crisis made suddenly, inescapably tangible, a transition zone not just between ice and water but between past and future and the deep time of the ancient, slow-­moving ice and the hastening fears of our brief human lives. And in that collapsing of temporalities it is possible to glimpse the rushing collision of the human and the planetary, and also something of the way that inflection reveals the connection between heedless consumption by the wealthiest among us and the lives of those who are already being displaced, and the many more who will be displaced by rising waters in the years to come.


THE ICE IS the memory of the world. It contains minute particles of soot and pollen and bubbles of gas that testify to changes in the atmosphere over millions of years. These allow us to trace not just the great cycles and irruptions of the planet, but also the history of human expansion and transformation. The ice sheets hold the signature of nuclear tests carried out in the 1950s and 1960s, the smelting of metals and other industrial processes – even the signatures of colonisation and early agriculture. In 2020 scientists published data showing soot in cores extracted from James Ross Island that corresponded with the arrival of the Māori in New Zealand just over 700 years before and that likely resulted from the widespread burning undertaken to convert the landscape to human use. Some scientists even believe that the global pause of the pandemic’s first year will be written into ice cores extracted hundreds of thousands of years from now.

But what these cores also record is a period of unusual climactic stability, extending from about 11,700 years ago until the beginning of this century. Known as the Holocene, this window of relative calm provided the conditions for the rise of the immensely complex and technologically innovative civilisation that now encircles our planet – the conditions that enabled development of agriculture, that opened up oceanic and continental trade, that allowed for the accumulation of wealth and power and the rise of globalised society.

This stability would not have been possible without the Antarctic and its ice. The Circumpolar Current and the systems of heat and saline exchange driven by Antarctica’s sea ice help drive the conveyor belt of the world’s currents while the continent and its sea ice help reflect the sun’s rays back into space to keep the planet’s temperature in equilibrium.

The ice also shows this period of relative stability is coming to an end. Ice cores allow us to map the levels of atmospheric carbon dioxide against global temperatures, revealing the two follow in lockstep – temperatures echoing fluctuations in carbon dioxide. That is, until you reach the past few decades, when the level of carbon dioxide begins to rise precipitously. Depicted on a graph the scale of this process is stark: over the past million years levels of atmospheric carbon dioxide have oscillated between about 180 parts per million and 300 parts per million, but since the beginning of the twentieth century they have risen faster and faster. In 1900 they were under 300 parts per million; by the late 1980s they were 350 parts per million. A mere twenty-­five years later, in 2013, they passed 400 parts per million. In 2021 they were 419 parts per million – and still rising.

Yet while fluctuations in carbon dioxide drive changes in temperature, the two processes are not simultaneous. The planetary cycles that control our atmosphere and oceans are vast and slow, meaning that changes to the climate we experience today are the result of emissions released a generation or more ago. Likewise, the cumulative effect of the emissions released since then will not be felt for another generation or more.

We inhabit the brief window between these rapidly rising emissions and the temperature rise that will follow. And it will come: we have pumped so much carbon dioxide into the atmosphere that even if we completely ceased greenhouse gas emissions tomorrow temperatures would continue to rise for decades.

This matters in a very direct way: the last time levels of atmospheric carbon dioxide were this high was four million years ago, during the Pliocene. Global temperatures were three degrees warmer than they are today and the collapse of the Greenland, West Antarctic and parts of the East Antarctic ice sheets meant sea levels were more than twenty metres higher. A rise of even a fraction of that amount would devastate contemporary coastlines and displace billions.

Yet it also matters because it tells us that the future is already here. The question is no longer whether the ice will melt and sea levels will rise, but how fast and how far. The choices of the past mean there is no way back. Change is coming whatever we do. The question now is how do we respond to that change?

And what does this mean for the krill? How will they survive in such a radically altered world? If their numbers fall, what will that mean for the species that depend upon them?

Neither King nor Kawaguchi believe the krill are likely to disappear entirely. But that does not mean they are not deeply concerned. ‘Their life cycle is so linked to the dynamics of the sea ice, you have to ask where they’re going to hide from predators or find food without it,’ says Kawaguchi.

King agrees. ‘We know they’re robust and durable animals because they’ve survived previous extinction events when 70–90 per cent of marine species disappeared. So it might just be an unprofitable time for them. But I also see potential for massive disruptions in the amount of krill and the predictability of their presence and numbers year to year. And that’s a serious problem for the species that rely upon them, because they can’t just swim another thousand kilometres if the krill aren’t where they’re supposed to be: they’ll be very unwell or dead before they get home if they try.’

He pauses. ‘The problem is what we’re doing, the change that’s taking place, it’s just happening so fast. That’s a real challenge.’

These questions are also a reminder of the degree to which our own future depends upon the fate of the Antarctic. What will the people of the deep future find stored in the ice? Will they see extinction and collapse? Or something less convulsive?

Get the latest essay, memoir, reportage, fiction, poetry and more.

Subscribe to Griffith Review or purchase single editions here.

Griffith Review