Friday, August 22, 2014

For Arctic ice watchers - a satellite view from NASA

From NASA:
"While this year is not heading toward a record low minimum extent in the Arctic, sea ice is well below normal and continues an overall pattern of decreasing sea ice during summer in the Arctic,” said sea ice scientist Walt Meier, based at NASA’s Goddard Space Flight Center in Greenbelt, Maryland.

While I'm here, might as well add a couple of live linked charts (click for larger view, or the links below) and a link to Neven's Arctic Sea Ice blog.

Arctic ROOS

Unsurprisingly there's no recovery, but no record this year either, from the look of things.

Arctic ROOS is a bit strange today, so I've added another (Sou: 22 Aug 14)

AMSR2 U Bremen


  1. "Rebound of Arctic sea ice!" Although I'm sure WUWT or similar has already made that point.

    1. He was making that point back in March when he misinterpreted some NSIDC(?) predictions. Not so much today, though...

  2. Arctic ROOS has changed since I put it up yesterday. (The charts are linked directly to the source.)

    Has there really been a steep drop or is there something wrong with the chart? I'd say the latter. Here is a link to AMSR2 from the University of Bremen:

    1. I've now added AMSR2 from the University of Bremen to the main article above, for comparison.

  3. Sou, I suspect the explanation comes down to the thresholds for what is / is not deemed to be ice covered. Arctic Roos uses a 15% threshold - for each pixel of their imaging, 15% - 100% ice = ice, <15% = open water. a pixel of 80% ice has the same value as a pixel of 20%.

    It is possible that during the apparent "hiatus"(!) in melting over the last month or so a large area of ice was slipping down through 40% to 30% to 20% without much change at the areas margins. The amount of ice could have halved with no difference in the area reported. But once this area drops below 15% it appears like its suddenly all melted.

    I'm not saying this is what happened, only that its one scenario that can cause the appearence of a near-instantaneous drop. It has certainly be responsible for some of the sharp movements in the past. This is why I pay a lot more attention to volume measures than to area or extent - the arbitrary thresholds are useful for their various purposes, but not necessarily optimal for interpreting what is actually going on.

    1. The ROOS area graph shows the same step-function behaviour. Other area and extent graphs don't. My money is on there being a bug.

    2. Actually some do, numerobis - for example Cryosphere today, which also uses a 15% cutoff for its area graphs shows the same rollover (though not as big as Arctic-Roos). OTOH, Bremen Uni (15% cutoff) and DMI (30% cutoff) don't.

      A bug is possible, but I think it is a mix of real data (a lot of ice dropping below the 15% mark) together with some differences in the algorithms used to interpret the satellite pics - for example, melt water sitting on top of ice floes poses particular problems, and the different groups factor this into their results differently.

    3. If it was a threshold issue, then why would ROOS area have the same behaviour? "Bug" is perhaps too strong a word; all the data sets are estimates from partially observed data.

      Anyway, this is neither here nor there; all the data sets are concordant that even in an average year these days you still end up melting almost all the ice.


Instead of commenting as "Anonymous", please comment using "Name/URL" and your name, initials or pseudonym or whatever. You can leave the "URL" box blank. This isn't mandatory. You can also sign in using your Google ID, Wordpress ID etc as indicated. NOTE: Some Wordpress users are having trouble signing in. If that's you, try signing in using Name/URL. Details here.

Click here to read the HotWhopper comment policy.