In conclusion, Mr. Connor [the Indie journo] has “jumped the gun.” I am firmly convinced that at least part of what we are seeing in the Arctic is due to human influences. However, sensationalist articles like Mr. Connor’s only serve to further polarize what is already a very polarized issue... I feel “ambushed” by Mr. Connor’s article.
So... whats up? Firstly, RP Sr seems to like the UI seaice data, I don't know why. As SB points out in a comment here they were wrong a while ago when they disagreed with NSIDC. To introduce yet another source, I'm going to use the NCEP product for this post (because its easily available; probably the answer to my RP question above is that the UI product is easily available for his purposes), though (as I pointed out here) I actually prefer the Bootstrap to the NASA Team product, at least for the Antarctic. also (for fans of important-but-boring detail) I'm going to use total area for my trends, not extent. Its all different ways of processing the same [[SSMI/SSMR]] dataset, anyway.
NSIDC has a press release out Sea Ice Decline Intensifies Summer Arctic sea ice falls far below average for fourth year, winter ice sees sharp decline, spring melt starts earlier which has some nice pics in it and some useful discussion by scientists (I'm doubtful about "The persistence of near-record low extents leads the group to conclude that Arctic sea ice is likely on an accelerating, long-term decline" early in the piece, I suspect that was put in by the PR folks. Its not really supported by the graph just next to it, which shows 2005 as low, but not really far below the trend line. Having said that, if their fig 2 is correct, the ice does seem to be unusually late recovering this year). Why they felt the need to put it out today, rather than waiting a few more days until the end of september, I don't know. Anyway, sea ice doesn't change that much over a few days so their pic is probably going to be correct at the end of the month. RP says that NSIDC haven't made sept available: but the NCEP GRIB is available in realtime from ftp://polar.ncep.noaa.gov/pub/cdas/ for anyone with the patience to convert it (you want wgrib, if you want to convert it, which you don't want to do...).
So *my* contribution to this is the piccy here, which shows sea ice area trends (% per decade) from the NSIDC (thank you Bob) dataset from 1979-2004 (black) and -2005 (blue) up to august. Because in all this the fact that the trends are quite seasonal is being missed a bit. Note that my trend for september is about 10%; NSIDC say about 7-8% depending exactly on which year you stop. The difference is probably due to me using area rather than extent (a quick test says that switching to extent reduces my trend by about a factor of 1.4).
Oh... and the usual caveats apply. I've checked my pic a bit but I did this all in an hour at home tonight, so if anything looks madly wrong to you, you may well be right.
[Update: and the *other* thing I was going to say which I forgot was... this is all pretty well in line with the model results, as far as I can remember them. I'm thinking of Recent and future changes in Arctic sea ice simulated by the HadCM3 AOGCM by Gregory, J. M.; Stott, P. A.; Cresswell, D. J.; Rayner, N. A.; Gordon, C.; Sexton, D. M. H. So the attribution of the trend to GW seems reasonable.]