- 31 August 2010 by Chris Mooney
- Magazine issue 2775. Subscribe and save
- For similar stories, visit the Climate Change Topic Guide
The ice may not retreat as much as feared this year, but what remains may be more rotten than robust
LAST September, David Barber was on board the Canadian icebreaker CCGS Amundsen (pictured), heading into the Beaufort Sea, north of Alaska. He was part of a team investigating ice conditions in autumn, the time when Arctic sea ice shrinks to its smallest extent before starting to grow again as winter sets in.
Barber, an environmental scientist at the University of Manitoba in Winnipeg, Canada, went to sleep one night at midnight, just before the ship was due to reach a region of very thick sea ice. The Amundsen is only capable of breaking solid ice about a metre thick, so according to the ice forecasts for ships, the region should have been impassable.
Yet when Barber woke up early the next morning, the ship was still cruising along almost as fast as usual. Either someone had made a mistake and the ship was headed for catastrophe, or there was something very wrong with the ice, he thought, as he rushed to the bridge in his pyjamas.
On the surface, the situation in the Arctic looks dramatic enough. In September 2007, the total extent of sea with surface ice shrank further than ever recorded before – to nearly 40 per cent below the long-term average. This low has yet to be surpassed. But the extent of sea ice is not all that matters, as Barber found. Look deeper and there are even more dramatic changes. This is something everyone should be concerned about because the transformation of the Arctic will affect us all.
The record low in 2007 cannot be blamed on global warming alone; weather played a big role too. That year saw a build-up of high pressure over the Beaufort Sea and a trough of low pressure over northern Siberia – a weather pattern called the Arctic dipole anomaly. It brings warm, southerly winds that increase melting. The winds also drive sea ice away from the Siberian coast and out of the Arctic Ocean towards the Atlantic, where it melts.
In 2008 and 2009, the dipole anomaly did not dominate and the extent of ice did not shrink as much during summer. This rebound led to much talk of a recovery in Arctic ice.
This June, the dipole anomaly returned and the ice extent for the month was the lowest ever. In July, however, the dipole pattern broke up and the rate of ice loss slowed. “Whether or not we set a new record depends very much on the weather patterns,” says Mark Serreze of the US National Snow and Ice Data Center based in Boulder, Colorado, which monitors the extent of sea ice – a particular way of measuring its area.
While much attention is likely to be paid to whether or not a new record is reached in the next month, there is more to sea ice than area alone. New sea ice can grow up to 2 metres thick during the winter. If it survives the summer melt, it can grow even thicker over the three to six years it might last before being swept past Greenland and out into the Atlantic Ocean, or succumbing to the summer melt. In places, this multi-year ice can pile up forming “pressure ridges” as much as 50 metres deep. But its average thickness is now less than 3 metres according to ICESat, the only satellite capable of measuring ice height and thus thickness (Geophysical Research Letters, vol 36, L15501).
There is no long-term record of the total volume of ice because we have only patchy data; ICESat was launched in 2003 and failed earlier this year. The nearest thing we have are estimates from PIOMAS, developed by Jinlun Zhang and his colleagues at the University of Washington’s Polar Science Center in Seattle. Actual satellite measurements of sea ice concentration since 1978 are fed into a computer model of the growth, melting and motion of sea ice to produce an estimate of ice volume. PIOMAS’s results correspond well with independent measurements by submarines and by ICESat.
According to PIOMAS estimates supplied to New Scientist by Zhang, the average volume of Arctic ice between July and September has fallen from 21,000 cubic kilometres in 1979 to 8000 cubic kilometres in 2009. That is a 55 per cent fall compared with the 1979 to 2000 average. “The loss of ice volume is faster than the loss of ice extent,” says Zhang. His model suggests that not only has the total volume of Arctic ice continued to decline since 2007, but that the rate of loss is accelerating (see “Going, going…”).
How can ice volume have kept falling when extent increased again after 2007? Because less and less ice is surviving to see its first birthday. “First-year ice is now the dominant ice type in the Arctic, whereas a few years ago multi-year ice was dominant,” says Barber.
Young ice is thinner than multi-year ice, and thus more likely to break into smaller pieces that melt more quickly, and more likely to be swept out of the Arctic and into warmer seas. That is precisely what happened in 2007, when persistent winds blew a thinner ice pack through the Fram Strait between Greenland and the island of Spitsbergen, leading to the dramatic ice loss. “The same wind 30 years ago when the ice was thicker would not have done as much damage,” says Bruno Tremblay, a climate researcher at McGill University in Montreal, Canada.
And while the area of young ice increased in 2008 and 2009, the amount of multi-year ice continued to fall. “There wasn’t a recovery at all,” Barber says.
Even the nature of the remaining sea ice might be changing. When Barber rushed up to the bridge that morning in September 2009, the first officer told him that while it looked like there was ice, it was no barrier to the ship at all. The reason: the ice was rotten.
It consisted of multi-year ice that had become riddled with surface thaw holes and had broken into pieces. Over winter, a 5-centimetre layer of new ice had formed over the dispersed floes. If a person tried standing on it they would fall right through, so it was no obstacle to the Amundsen. It is not clear how widespread these conditions are because satellites cannot distinguish between rotten and more solid ice (Geophysical Research Letters, vol 36, p L24501). The rotten ice is less of a barrier to waves as well as ships, meaning waves can penetrate further into ice packs and break up more ice.
What it all means is that, much like the Amundsen, we are now cruising effortlessly into a world that may soon feature an essentially ice-free Arctic during at least part of the year. “Thirty years from now, maybe even 20 years from now, if you were to look at the Arctic from space you would see a blue ocean [in summer],” says Serreze.
The implications of such changes for wildlife and the human inhabitants of the region, for the global climate and for geopolitics are profound. The Arctic would be traversable by ship. It would be far more open to oil and gas exploration, and mineral extraction. Its dark ocean waters, mostly devoid of ice, would absorb still more sunlight, further warming the overlying atmosphere during an increasingly lengthy ice-free season, reshaping weather throughout the region and well beyond it.
Worryingly, the melting of the Arctic sea ice is proceeding considerably more quickly than most climate models have predicted. Among the suite of models submitted for the 2007 report of the Intergovernmental Panel on Climate Change (IPCC), only two out of 23 yielded results for Arctic sea ice that were consistent with observations, says Cecilia Bitz of the University of Washington in Seattle.
According to the 2007 models, the Arctic will not become ice-free in summer until some time after 2050. However, researchers like Barber and Serreze think this landmark occurrence will come much earlier. Barber has predicted that it will occur sometime between 2013 and 2030.
If most models aren’t capturing the full extent of changes in the Arctic, it is probably because the modelled feedbacks are too weak, says Bitz. In other words, they may not be sensitive enough to processes that, once they get going, self-amplify in a continuing loop.
Every model includes the “ice albedo feedback”, in which the melting of ice that reflects most of the sun’s heat exposes dark water that absorbs most heat. That leads to more melting and so on – a positive feedback. But there could be many others.
Consider, for instance, the role of Arctic storms. They break up ice with their winds and waves, making it more prone to melting – and the more open water there is, the more powerful waves can become. These larger waves – which were not included in any models – then penetrate further into the ice pack, breaking it up into smaller and smaller pieces, says Barber. From the bridge of the Amundsen as it sat moored in the ice last year, Barber himself watched as a large swell broke a chunk of ice the size of Manhattan into a number of pieces roughly 100 metres across.
Storms also bring snow, which in autumn and winter actually slows the growth of sea ice by insulating it from cold winds, as well as reducing heat loss from the sea below. So if climate change leads to more snow in autumn and winter, this will be yet another factor contributing to the loss of sea ice.
Bitz thinks the 2007 low was a wake-up call for climate modellers, compelling them to look more closely at how their programs handle sea ice. She expects that when the next set of models is submitted to the IPCC for its 2013 report, their outputs will be much more in line with observations. “The modelling centres are short of resources for giving focus to a particular part of the model,” she says. “But when a big story comes out like 2007, they redirect, and that will pay off.”
The implications of the loss of Arctic sea ice in the summer are hard to overstate. Most attention has focused on charismatic megafauna like polar bears and walruses, but they are just the icons of a broader ecosystem that is already being dramatically disrupted. The sea ice is as important as the trees to a rainforest, Barber says.
The loss of sea ice will also have many other impacts. For instance, the increase in the size of waves has already begun to cause serious coastal erosion in places like Alaska, with the effect magnified by warmer waters and rising sea level. The impact of the waves eventually melts the permafrost of which the coastline is composed. “Some of those coastlines are made of very fine silt,” says Tremblay. “The land just washes away.”
A warmer Arctic will also affect weather in the mid-latitudes – indeed, it has already begun. Take the Great Plains of the US. According to Michael MacCracken of the Climate Institute in Washington DC, this region’s weather is very much determined by clashes between cold air masses coming down from the Arctic and warm air masses from the Gulf of Mexico. As the Arctic blasts are less cold than they used to be, the Gulf’s warm air tends to push further northwards. The result is a northward shift of weather patterns, and more extreme storms and heavy precipitation events in regions not used to them.
Finally, there are the economic and industrial implications. “The engineering challenges get simpler,” says Barber, “for drilling, for putting drill ships in place, for having icebreakers, to make tankers carry oil across the pole – all those kinds of challenges associated with industrial development.” Such challenges will diminish, or even vanish entirely. The Amundsen’s surprisingly easy voyage through the Beaufort Sea in September 2009 could be a herald of things to come.
Chris Mooney is a host of the Point of Inquiry