Lodger: That did occur to me as a possibility, but if that was so, wouldn't the data be limited to spring and mid to late summer, with empty patches during the minimum and maximum? Instead, we only get data during the refreeze, with none during the melt, when I'd think... Oh. Hm. Not sure if this is correct, but it occurs to me that the problem might be melt pools on the surface of the ice. After the minimum and on up to the maximum, the pole is shrouded in darkness, and there shouldn't be any melt pools on the surface of the ice--so if the satellite detects open water, it's definitely a crack in the ice itself, and not water pooling on the surface. During the melt season, though, you would get multiple pools forming, which would (depending on the ice thickness and its surface geometry) give multiple different values for the level of the water, thus preventing an accurate calculation of the ice freeboard. I suppose one might get around that by devising an algorithm that only looked at fissure-shaped patches of water, although I don't know enough about how the ice behaves to know if that would really work.

Finally, a graph of Cryosat-2 data! I was beginning to think they'd never release it. I'm puzzled by the lack of spring and summer data, though; does anyone know the reason for that? Is there something about sunlight reflecting off of the ice that makes taking readings difficult (more microwave light emitted by the ice and water themselves, perhaps)?

Thanks very much for the info, Wayne! That's very interesting; I wonder if that ~-11 ˚C temperature would change with different ice thicknesses? That is, would warmer air temperatures be needed above thicker multiyear ice vs. first year ice before the fragility you mention started to become apparent? On another note, that video is breathtaking; I'm particularly interested by all the beautiful, silvery air bubbles (I presume?) visible on the underside of the ice. I highly doubt that they have any real importance in terms of the behavior of the ice, but I do wonder where they come from; surely, no gases released from further down in the water column would make it that high without being dissolved. Air trapped under the ice by wave motion, maybe, or perhaps a much gentler version of the famous diet Coke + Mentos "reaction," with the underside of the ice providing a highly textured substrate that makes it easier for gases to emerge from solution.

Wayne, would you mind giving me a link for that "-11 ˚C air temperature required for freezing" bit that you mention? I've done some (admittedly, cursory) scrounging around, and I can't seem to find a reference that mentions that value.