Frankd 1977, In most of these cases, there's really not much freezing going on in these "freeze backs". Instead, the freeze back day is showing the more accurate representation of the situation on the ground. Periodically, a storm system with optically thick clouds will pass over the sea ice. The microwave sensor is not as good at seeing through these clouds as the more typical thinner clouds. The effect of these cloudy events is to reduce the concentrations under them by 5-15% as they pass by. When the clouds move out of the domain or clear up, the sensors get a more accurate picture of the underlying ice. My guess is that cloudiness has not varied as much this year as in recent years.

Off topic, but fun: Check out the last 5 frames or so of the thickness animation at: http://www7320.nrlssc.navy.mil/hycomARC/navo/arcticictn_nowcast_anim30d.gif Watch Fram Strait. A little blob of thick ice gets ejected into the strait from north of Greenland. This happens at the same time, or just after the western boundary of the thickest ice gets hammered from strong westerly winds. Whether this is an illusion, or not, I don't know. But it looks like the momentum from that western wind impact is being transferred through the thick pack ice and causing that little thick blob to be ejected from the main pack. Like someone pounding on the bottom of the ketchup bottle....