Unofficially, for the moment, but Cryosphere Today, which doesn’t normally change its appearance much, has made the effort to point out that yesterday, August 9th, saw the lowest ever recorded Summer sea ice area. The link is here.

Bill Chapman also points out that there is still a way to go before the end of the melt season; the date at which Summer ice normally reaches the minimum varies, but in recent years hasn’t been before the end of August, and is ‘normally’ at the end of the first week in September. It could be as late as mid-September, especially this year. So we have between two and five weeks, with the latter being more likely, before the final Summer sea ice minimum for 2007 is reached.

As the old man projected this back in April, in itself it’s not a great surprise, but even I confess to a certain shudder when I look at the graphics on the various pages.

There’s a good graphic on the MMAB sea ice page, and excellent images on the PolarView/Damocles/IWICOS/IOMASA website. The latter don’t seem to like direct links, so you’ll need to hit the ‘overview’ link on the page. The image is then able to be enlarged substantially, showing the various grids quite clearly.

I’d draw your attention to a detail there which is also unusual, even in the context of recent sea ice declines; a close look at the Northern edges of Greenland and Ellesmere Island show that there is substantial thinning along much of the coastline. Whilst it is worrying to note the minimum in itself, this is a possible precursor to something which I haven’t seen discussed anywhere; the possibility of the Summer ice pack detaching itself completely from the land. This might have been something to expect in the coming years, but if it does happen that Kap Morris Jessup or Cape Columbia see open sea this year, it suggest that the calls for a review of the ‘standard’ expectations of future sea ice is now beyond urgent.

How low will it go? How much more melt are we going to see? If the NSIDC July graph is anything to go by, an extent anomaly (sorry for changing metrics mid-post) of -20% or more is plausible. In a year. On top of five consecutive years of record or near-record minima. And take a look at those graphs, like the hemispheric anomaly on CT; where is the variability? Where is the ‘natural’ ebb and flow of numbers caused by changing synoptics and interannual variations? This decline is, in climate terms, near-linear. So, here is the hypothesis. The Arctic Summer sea ice decline reached a tipping-point in 2002-3. Comments welcomed, as ever.

Be loved.

EDIT: RealClimate are commenting on this, too. Peter Hearnden has also linked to a satellite image, in the comments, if anyone is in doubt that the graphics are credible.. thanks, Peter.

Further edit: on the Rc site, mark Serreze has pointed us to the NSIDC’s re-launched weekly update site. Not this deals with sea ice extent, and monthly averages, a different system to CT’s. The picture is still grim. Link here.