Thanks Wipneus. That seems entirely plausible.

So I notice (from limited observations over the past week or so) that the prelimimary JAXA number seems to be systematically revised upwards by around 40k one day later. Not a big deal, just a little curious why it is such a consistent offset. Or is my sample just too small?

Neven, thanks for the insight (and html fix). I see that now on their page. I also see they have a large disclaimer there stating the new results they are posting are experimental and need to be used with caution. So maybe wise not to put too much weight on the absolute values this year. http://www.iup.uni-bremen.de:8084/ssmis/index.html

Oops. That last paragraph in the above posting was my own comment, not a quote from the article. Basic html usage failure. Is there a way to edit comments that I'm missing? (Note to self - use Preview button next time before conferring immortality on masterpiece.) Also note that the quote is from 2011, but the principle should still apply.

Re: Bremen vs NSIDC, I came across this comment: http://www.livescience.com/16083-arctic-sea-ice-2011.html [Bremen] use a satellite sensor that can detect ice cover at a higher resolution than that used by NSIDC. The two groups probably came up with different results because this year ice was more dispersed in the water, and the Bremen group was able to pick up on details, leading to more variability between the two sets of measurements, Meier said. If all else were equal, once might infer that the higher resolution sensors used for the Bremen data would make this a more accurate result. Would be interested to hear opinions on this.

This data series also appears to be at least equalling, and possibly beating, its all time low today: http://arctic-roos.org/observations/satellite-data/sea-ice/observation_images/ssmi1_ice_area.png

How do the Bremen numbers compare with NSIDC? To my squinting eyeball, it looks like they have already equalled the all-time low extent today in that data series: http://www.iup.uni-bremen.de:8084/ssmis/extent_n_running_mean_previous.png

Reading a big more on the winter extent in other blog posts and comments here, it appears that Cryosat 2 covers a smaller area than PIOMAS, so it is unsurprising that PIOMAS shows a larger winter extent.

Newbie here, great blog and insightful posters. Just a couple of observations on the new data compared with PIOMAS. For the past winter, PIOMAS has 22000km^3, Cryosat 2 has 14000km^3. That is quite a big delta. For the current summer, PIOMAS is now 5800km^3 and Cryosat 2 is 7000km^3. Since this summer minimum is still not achieved, and it is not clear what date the Cryosat 2 summer number is from, it is not possible to compare directly with the PIOMAS value, but I'd guess the Cryosat number is at least a few days older than the latest PIOMAS number in the database, which would suggest they are in remarkably good agreement, as are the extremely high summer decline rates. So overall this looks like strong validation for the worst-case scenarios implied by PIOMAS that I suspect many had secretly hoped were too bad to be true.