Earlier this month, a colossal iceberg dislodged from Antarctica. At about twice the size of Rakiura/Stewart Island, the long and narrow slab known as A-76 is the largest iceberg in the world. The spectacular satellite footage was global news.
A-76 broke off the Ronne Ice Shelf, the continent's second-largest floating ice platform, which stretches across a massive bay in the Weddell Sea to the east of the Antarctic peninsula. When scientists at the US National Ice Centre made the official announcement, they were quick to point out that iceberg calving is a natural process and the forces that cleaved the berg from the face of the ice shelf could not be directly attributed to climate change.
In any case, even if the 181km-long and 26km-wide berg now drifts north into warmer waters and eventually melts completely, it won't make any difference to global sea levels – just as a melting ice cube doesn't change the water level in a glass.
But the news sharpened attention on Antarctica's accelerating melting and the question of how much its icy cap will add to future sea levels. An even more urgent question is whether we're approaching a temperature threshold that would trigger irreversible disintegration of some Antarctic glaciers and commit us to multigenerational sea-level rise over coming decades and centuries.
"There's a concern that as we approach 1.5C or 2C global warming, that may be a threshold for the complete loss of ice shelves," says Tim Naish, a climate scientist at Victoria University of Wellington's Antarctic Research Centre.
Ice shelves, such the Ronne and the Ross, are floating extensions of the kilometres-deep ice sheet that wraps around the Antarctic continent. Because they float, they don't contribute to sea-level rise.
"But once they're gone, the glaciers feeding them are unleashed," Naish says. "Because the ice shelves block up these huge bays on the coast, they are a bit like a cork in a bottle. They hold back and stabilise the ice sheet and slow it down from flowing faster into the ocean. There's a tipping point at 1.5C-2C whereby ice shelves can no longer survive."
In 2018, the Intergovernmental Panel on Climate Change (IPCC) published a special report focused on the world's oceans and cryosphere (an all-encompassing term for those portions of Earth's surface where water is in solid form), and it echoed this warning.
"Acceleration of ice flow and retreat in Antarctica, which has the potential to lead to sea-level rise of several metres within a few centuries, is observed in the Amundsen Sea Embayment of West Antarctica and in Wilkes Land, East Antarctica. These changes may be the onset of an irreversible ice-sheet instability."
The IPCC was equally clear about the consequences accelerating ice loss from Antarctica – and Greenland's ice sheet – already has for global sea levels.
"Global mean sea level is rising, with acceleration in recent decades due to increasing rates of ice loss from the Greenland and Antarctic ice sheets. The rate of global mean sea-level rise for 2006-2015 of 3.6mm a year is unprecedented over the past century, and about 2.5 times the rate for 1901-1990 (1.4mm a year). Mass loss from the Antarctic ice sheet over the period 2007-2016 tripled relative to 1997-2006."
Three main sources drive rising seas. Naish says about a third currently comes from the loss of mountain glaciers in mid to low latitudes. Another third comes from a process known as thermal expansion – in which warmer water takes up more volume as the oceans warm – and the final third comes from melting polar ice sheets.
"The acceleration is not in the thermal expansion," he says. "It's in the loss of mountain glaciers and the Greenland and Antarctic ice sheets. At the moment, Greenland is contributing significantly more mass than Antarctica. The concerning thing is that if the trend we're seeing in West Antarctica continues, then Antarctica would become the dominant contributor to sea levels by the end of the century."
Scream a rainbow
In 2011, Nancy Bertler had what she describes as a heartbreaking Antarctic experience. The Victoria University of Wellington glaciologist had already spent many summers working in the world's coldest and driest icy desert, drilling deep into the ice on the flanks of Mt Erebus and other remote sites along the coastline in search of climate clues from the past. Each field season, she spent weeks living in a tent and thought she'd seen all that Antarctica's extreme weather has to offer. But then a rainbow appeared – a phenomenon possible only in unusually warm air that can hold more moisture.
In the years since, Pegasus Field, the southernmost of the airstrips delivering people and cargo to Scott Base and McMurdo Station, had to be replaced because of surface melting in summer.
Antarctica is clearly changing, but to predict how exactly the vast continent and its different icescapes will respond to a warming world is a hugely complex exercise. Despite ever-improving models of ice-sheet dynamics, Antarctica's contribution remains the largest uncertainty in sea-level-rise projections.
In the broadest sense, scientists agree that the larger and thicker East Antarctic ice sheet is more resilient to warming than the icy blanket over West Antarctica, which is already losing ice faster than other parts of the continent and sits in a deep basin, 2.5km below sea level in some places. But some parts of West Antarctica are more sensitive than others – and with a third of the continent's total ice sheet grounded below sea level, East Antarctica is by no means a sleeping giant.
A few days before A-76 detached, the scientific journal Nature published two modelling studies, each positing different projections of how much Antarctica's melting might contribute to future sea levels.
One study, led by Robert DeConto, a climate-modelling scientist at the University of Massachusetts Amherst, predicts that Antarctica's ice loss will continue at a pace similar to today's only if we manage to keep warming below 2C – essentially, the target of the Paris Agreement. In a scenario more consistent with the pledges countries have made under the agreement so far, which would bring us closer to 3°C in warming, we would see an "abrupt jump in the pace of Antarctic ice loss after around 2060".
Even if we achieve the 1.5-2C Paris target, Antarctica will contribute 6-11cm to sea-level rise by 2100. But on the current trajectory, unless ambitious action to rein in warming begins by 2060, nothing, not even geo-engineering, will be able to stop 17-21cm of sea-level rise coming from Antarctic ice melt alone by the end of this century. West Antarctica's Thwaites Glacier – already dubbed the "doomsday glacier" – and other ice streams in the same embayment would retreat beyond a point of no return, destabilising the entire West Antarctic ice sheet.
The other study, led by King's College London climate scientist Tamsin Edwards, brings together climate modellers from 15 countries, including New Zealand, and paints a different vision of Antarctica's future, with no marked tipping points. Unlike the DeConto study, which is based on a single model, it involves multiple models and projects ice loss from all land-based ice – polar and mountain glaciers.
It produces a statistical average of the model-generated projections to show that limiting global warming to 1.5C would halve the land-ice contribution to sea-level rise this century. Mass loss from Greenland would decrease by 70 per cent – but for Antarctica, the link between ice loss and different emissions scenarios is less clear, partly because some models predict losses from West Antarctica but increased snowfall, and therefore ice gains, in other parts of the continent.
For Richard Levy, a climate scientist at GNS Science and a co-director of the NZ SeaRise project, one of the fundamental differences between the studies is that some models in the multi-model study have not been tested against data about past climate change in Antarctica.
Antarctica is one of our best archives of information about past climates. The records go back millions of years, written into the ice and the sea floor beneath the continent's ice shelves and off its coast. Levy, Bertler and Naish are all part of the Melting Ice and Rising Seas project, which won the 2019 Prime Minister's Science Prize. Collectively, the team analysed ice and sediment cores to glean information about how Antarctica changed during warmer times in the past.
Going back some three million years to a geological period known as the Pliocene – the last time atmospheric carbon dioxide concentrations were above 400 parts per million (ppm) and it was about 2C warmer than pre-industrial times – they found that up to a third of Antarctica's ice sheet melted, lifting seas by as much as 20m above present levels.
Bertler's focus is on ice and the tiny samples of ancient air it holds. By drilling deep into the ice, she can travel back through past ice ages and interglacial periods. Her most recent project uncovered surprising results. Her team extracted an ice core at Roosevelt Island, a rocky outcrop that sits 200m below sea level with a 700m icy cap on top, which is embedded in the Ross Ice Shelf.
"It's a perfect place to look at how the ice shelf has behaved – it's almost like an independent weather station at the right spot."
Bertler expected to travel back to the previous glacial maximum, 20,000 to 25,000 years ago, when the ice sheet was larger and this island would have been overrun by thicker ice.
"But what we found was a record that goes back well over 80,000 and perhaps as much as 130,000 years, to the time of the last interglacial when temperatures were 1-2C warmer than today. It appears that during this time period, we still had ice on Roosevelt Island."
Her team is running further analysis of the ice-core samples to verify the oldest age, but Bertler says the most logical explanation is that the Ross Ice Shelf was still there, at least in part. If confirmed, the results raise more questions than answers, especially since other research shows that sea levels were 6-9m higher at the time, suggesting not only that West Antarctica lost a lot of ice – even though the ice shelf itself didn't disintegrate – but also that the remaining ice must have come from the larger East Antarctic ice sheet.
The findings sit more closely with the multi-model projections, and in that sense, the DeConto modelling study represents "an end member" in a range of possible outcomes we must avoid, says Naish. The model includes generally agreed concepts specific to Antarctica: the higher vulnerability of marine-grounded parts of Antarctica's ice sheet and the propagation of fractures caused by surface melting. But it also features the more controversial idea of marine ice-cliff instability, which predicts that once an ice shelf is lost, land-based glaciers with a cliff face higher than about 100m would become mechanically unstable and collapse upon themselves.
"The big difference between the two papers is that the one that uses an ensemble of models inherently averages everything," says Naish. "I still think we are underestimating the potential contribution to sea-level rise from Antarctica by using this multi-model average approach."
Point of no return
Among scientists studying polar ice sheets and climates of the past, there is "fairly good agreement" that some West Antarctic glaciers, including the Thwaites and Pine Island glaciers, have reached a tipping point already, Levy says. Others could follow if ice shelves lose their buttressing effect.
"The consistent feature of the models is that when they force them with 1.5-2C of average global warming … that is a climate in which the ice shelves just can't hang around. It's that threshold above which most of these processes take off and you can't stop them or reverse them."
This concern prompted the international cryosphere research community to form a coalition in April, known as 50x30, to press governments, including New Zealand's, to commit to cutting emissions in half by 2030 to prevent cascading and irreversible damage.
During the project's virtual launch, Christiana Figueres, who was head of negotiations at the 2015 UN climate summit in Paris, issued a stark warning: "Science has become much more granular since the adoption of the Paris Agreement. Today, we know that we're not just approaching but we are entering into at least three tipping points: coral reefs, summer ice in the Arctic and the melt of the Western Antarctic ice sheet. We are therefore at a perilous moment in natural history, when equal systems that have supported our human society for 12,000 years are about to turn against us."
According to a 2018 IPCC report on what it would take to keep warming at 1.5C, the world would have to cut emissions by 45-60 per cent by 2030. Although many countries have now pledged carbon neutrality by 2050 and adopted the Paris goal of staying close to 1.5C, a February UN report made it clear that almost no country is taking the steps necessary to achieve 50 per cent reductions by 2030.
Victoria's Antarctic Research Centre is a founding member of 50x30 because, as Naish says, overshooting the Paris goal would "result in many metres of sea-level rise, with major impacts across multiple generations", even if emissions eventually come down to reach carbon neutrality by 2050.
The New Zealand Climate Change Commission is expected to release its final advice to the Government at the end of this month. In its draft advice package, it proposed carbon budgets that would reduce emissions by 36 per cent by 2035, which "will not avoid an overshoot of the Paris goal".
"We have to assume that we're dicing with this tipping point," Naish says. "If we can achieve the Paris target with limited or no overshoot, not going above 2C and having to pull it back again, then we have a much better chance of reducing that risk."
Trouble at the tideline
Wellington can expect a 1.2m sea-level rise by 2150 in projections by the NZ SeaRise project as this country faces the uneven consequences of global warming.
Sea-level rise is usually expressed as a global average. The latest Intergovernmental Panel on Climate Change projections are for sea-level rise between 29cm and 1.1m above a late-20th-century baseline by 2100, depending on which route we follow to reduce emissions. But the seas don't rise universally across the planet.
On a global scale, gravitational forces mean that ice lost in one hemisphere has a bigger effect on sea levels in the opposite half of the globe.
"That's why New Zealand gets 10 per cent more in sea-level rise," says Richard Levy, a climate scientist at GNS Science. "Greenland is contributing more and the melting mass is drawn towards the ocean, away from the ice sheet."
On a regional scale, sea level is modulated by wind and ocean currents. And locally, the actual change in sea levels at any particular stretch of coastline is a combination of rising seas and changes in land elevation, either through earthquakes or subsidence. Figuring this out on a scale that is useful for planning is the task of the NZ SeaRise project of which Levy is a co-director.
"Even if we achieve Paris, we'll still have to prepare for what's coming," he says. "Even under a strong-mitigation scenario that keeps warming under control, the national coastal hazards guidance projects sea-level rise in Wellington will be about 30cm more than 2005 by about 2070. But the SeaRise project, by bringing in the vertical land movement, shows that the same amount of sea-level rise could be as early as 2050 or even 2035."
If projected further into the future, Wellington can expect 1.2m by 2150 because the lower North Island is subsiding. In contrast, Pikowai in the Bay of Plenty should see only 20cm of sea-level rise by then because the land is uplifting there.
A 30cm rise, on top of the roughly 20cm that has already happened over the past century, will make coastal stretches around Wellington Harbour inaccessible and turn a 100-year flood into an annual event.
"That's the fundamental adaptation story. At a minimum, this is what we're going to have to adapt to and start planning for now. At the same time, we're going to have to mitigate or else those numbers are going to get much bigger."
Despite the challenge, Levy says he is feeling more optimistic than he has for some time. "People are starting to act in a way that shows we're taking this somewhat seriously."
Nevertheless, the SeaRise project has also calculated future sea levels for higher-emissions scenarios, so that critical infrastructure currently in planning stages – for example, Auckland's airport expansion – can be stress tested at that upper level, based on accurate and locally relevant projections.