In The Field

On board the Amundsen: how sea ice could affect permafrost

Spending time on board a research ship inevitably blurs the separation line between reporting and doing science, even though most of the time the ‘science’ part of it is limited to dragging a sled-load of gear out onto the ice or, at best, pulling up a sampling net from a dive hole.

But today I have perhaps contributed to science a little bit more than usual, if only by passing on information. Yesterday evening I handed Dave Barber, the chief scientist of the Circumpolar Flaw Lead study, a paper coming out on Friday in the journal Geophysical Research Letters (which my ever-alert colleagues had sent me). The paper describes how, in a model, Arctic sea ice loss leads to strongly accelerated permafrost thawing on land. Dave promised he would read it.DSC00626.jpg

When I entered his office this morning, he was brimming with enthusiasm. “Very interesting stuff,” he said. “We should really start looking for what’s going on in the Canadian permafrost. A model study like this could direct an observational study. Actually, I think I’d like to do this myself.”

The paper, written by a team led by David Lawrence of the National Center for Atmospheric Research (NCAR) in Boulder, Colorado, suggests that if massive sea ice loss, such as occurred last summer, will happen ever more often, Arctic land warming could triple in the near future. This would have an effect on the rate of permafrost melting up to 1,500 kilometres inland on the continents surrounding the Arctic Ocean.

Thawing permafrost can destabilize houses, warp roads, and buckle trees. As Arctic soils hold around one-third of all the carbon stored in soils worldwide, their thawing may also add substantially to the release of greenhouse gases into the atmosphere. As Lawrence emailed me: “If sea ice extent continues to retreat rapidly, then we can probably expect an increase in the rate of warming over land and that this increased rate is likely to contribute to some near-term permafrost degradation, especially where permafrost is currently warm and vulnerable.”

In the last century, the Arctic has warmed by around 3.5 degrees Celsius. And since the 1980s, when the warming trend kicked into gear, the multi-year sea ice pack has decreased by more than 10% per decade. This is equivalent to some 70,000 square kilometres – the area of Lake Superior – per year.

But it was only last summer that scientists observed what many say was a catastrophic loss of ice in the Arctic Ocean. By September, when the ice is at its seasonal minimum, it had shrunk to some 4.3 million square kilometres – a reduction of 65% relative to the 1987-2003 average.

Throughout October to December last year, the CCGS Amundsen (the Canadian icebreaker I’m currently on) sailed in open waters in the eastern Beaufort Sea, a region that should normally freeze in fall. Unusually frequent and strong storms had once and again mixed the upper layer of the ocean, bringing warmer water to the surface and preventing it from freezing.

“Our observations are very much of line with this paper,” says Barber. “We don’t do terrestrial work, but from a sea ice view accelerated melting is certainly what we’re seeing. Our data, and everything we experienced over this last year, agree with what the model study predicts might happen in the future.”

He doesn’t expect a recovery of any kind this year. Our current study areas, the Amundsen Gulf – the western gate to the Northwest passage – is mostly ice-free already, and the unusually sunny and warm weather during the last 10 days has added to the rapid melting of the remaining ice in the region.

The large patches of dark open ocean allow for increased absorption of sunlight, which at this time of year shines 24 hours a day. If there was more ice, the white surface would reflect back to the atmosphere a much larger portion of sunlight. This so-called ice-ocean albedo effect is the dominant of several climatic feedbacks in the Arctic. “The warming effects here have a tendency to accumulate, which is why we will likely see a repercussion of last year’s event in 2008,” says Barber.

“There’s an awful lot of heat accumulating in the upper part of the sea,” he adds. “It may take until late autumn before the ocean gets rid of this excess heat, and it’s intuitively obvious that its horizontal transfer will heat the terrestrial environment.”

Models such as as the Community Climate System Model used by Lawrence and his team capture many, but not all, feedbacks that contribute to Arctic warming. What the study does not account for, says Barber, is that the open ocean is a magnet for storms feeding on the released heat. Models can’t usually represent storms because they happen on too small a scale. But although small-scale, storms add quite some complexity to the marine-terrestrial connection in that they mix the ocean, keeping it warmer than usual, and bring precipitation. More snowfall on land will insulate the ground from warm oceanic air and might help prevent soils from rapid thawing.

Posted on behalf of Quirin Schiermeier


Comments are closed.