I’m (possibly/probably) a loser

The National Snow & Ice Data Centre in the USA has declared that this year’s minimum Arctic sea ice extent has been reached – 4.52 million square kilometres on September 12th, only 390,000 km2 more than the record 2007 minimum (and 2.24 m km2 below the 1979-2000 average minimum). It looks as though their line may bump along the bottom of the graph for a while, so there may be some potential for that number to reduce a little. Their figure for Sept 12 is 9.4 percent above the 2007 minimum, so unless there’s some unprecedented melt over the next couple of weeks, I am prepared to accept that I have lost my bet with Malcolm (see comments here and at Poneke! here), which was based on NSIDC numbers. My cheque book is at the ready.

The situation with my other bet, with William “Stoat” Connelly, is a little less clear. It’s based on the minimum area (not extent) at Cryosphere Today. The 2007 minimum on CT’s metric was 2.92 m km2, and the lowest figure I’ve seen so far (a few days ago), was 3.004 m km2 (using this iPhone formatted graph). It’s bounced up a little since, but if we look back to the same time last year we can see that the graph stayed pretty flat until October (notice how the anomaly increases as we get further into autumn as the late start to freeze has an effect). I reckon that William is likely to win, but it’s still not certain.

[Update] CT has just posted their preliminary minimum for the year (though I can’t see a figure), with a page allowing comparison with earlier minima. Here’s this year’s minimum in context – way below the long term trend:

For some interesting commentary on the difference between satellite estimates and the reality of cruising through the ice, have a look at the most recent posts at the Swedish polar blog I linked to a few weeks ago. They’re finding the satellite data a not entirely useful guide to navigation, but only Google knows what this means…

Each time, we have been forced to take circuitous routes and it has taken 12-24 h to snirkla us out from these sjok.

Your snirkla may differ. (Magnus, help!) It’s certainly an interesting blog, well worth a peruse (there’s more about methane too).

So what can we infer from this year’s melt season? The first thing is that whether you use NSIDC or CT figures (+9.4% and +4.1% 2008 on 2007 respectively), it’s difficult to argue that there has been any substantial “recovery” of the Arctic sea ice. The 2008 minimum is the second lowest in the record, well ahead of 2005 in third place. A large part of 2007’s decline has been blamed on exceptional weather in the Arctic last summer. This year, by comparison, was cooler – but the ice still got very close to a record.

There’s also substantial evidence of dramatic changes in the ice stuck to the islands north of Canada. NASA’s Earth Observatory has an excellent in-depth look at the break-up of the ice shelves along the coast of Ellesmere Island (also at BBC, New Scientist):

On July 22, 2008, a new wave of ice shelf disintegration began and, by late August, these ice shelves had lost a total of 214 square kilometers (83 square miles). A group of researchers led by Derek Mueller at Trent University, and Luke Copland at the University of Ottawa, announced the changes in early September 2008. The Ward Hunt Ice Shelf lost a total of 42 square kilometers (16 square miles). The Serson Ice Shelf lost 122 square kilometers (47 square miles)—60 percent of its previous area. The Markham Ice Shelf, with a total area of 50 square kilometers (19 square miles) completely broke away from the Ellesmere coast.

And here’s what it looks like from space:

These NASA pictures show the ice shelf loss, but they also show the character of the offshore ice – fractured and drifting, even in July.

What odds for 2009, William and Malcolm?

Even if I lose both bets – and I expect to – the margin of my loss is small by either NSIDC or CT’s reckoning. To me, this summer does not suggest that we’re seeing some sort of recovery back towards the long term trend (which, lest we forget, is already showing a steady decline). It’s clear that the loss of ice has increased over the last three to four years, and that the nature of the ice (the thickness, amount of multi-year ice, distribution of same) has significantly changed. There will undoubtedly be a lot more news to come on what actually happened this summer (the change in sea ice volume figures will be very interesting), and if we see a warmer Arctic spring and summer in 2009, a new record could well be on the way.

Meanwhile, it’s worth reflecting on this passage from the ISSS-08 blog about life in the Siberian seas:

Friday late afternoon, in the middle of the East-Siberian pitchblack night, our ship passed Jeannette island at around 76 ° 44N and 157 ° 54E, northeast of the New Siberian Islands. Igor, our chief scientist, told us that this island was named after an earlier polar expedition at the end of the 19th century. This expedition was led by De Long, a lieutanant in the U.S. navy, and was aiming to reach the North Pole. They started in the Bering Strait (between Alaska and Siberia) to be frozen in around Wrangell island in September 1879.

The ship, named Jeannette, was transported by the ice for two years before it got crushed and sunk. Only one of the four life boats made it to the Siberian mainland, where they could spread the terrible news. The most interesting part here is that a couple of years later, pieces of the mast of the Jeannette were found on the eastcoast of Greenland (!). These findings were one of the first pieces of evidence of the transpolar ice transport, and an important reason for the famous Fridtjof Nansen to set sail with his ship Fram to explore these unknown polar ice operational. According to Igor, the Jeannette is lying only a couple of miles away from our sampling station.

Sic transit… etc.

[And for the big picture on glaciers, you can download the World Glacier Monitoring Service’s new 2008 report here. It’s 25.9MB of icy goodness. Great pictures for the terminally morained.]