That image normally looks something like this:

The navy is showing a large area of very rapid movement. A long open channel in sync with the orbital period of their satellite could explain it. The definition of “extent” for this graph is 30% concentration or greater. It could be that a large area just went from 30% to 29% due to ice movement. Other concentration graphs show a downturn as well, though not so big.

Source: http://www.iup.uni-bremen.de:8084/amsr/ice_ext_s.png

So, one of three things is happening:

1) The planet is experiencing a sudden and most definitely unprecedented instantaneous period of warming which is affecting both hemisphere’s equally. Pretty unlikely, I’d say - although if that is what’s happening then it’s unlikely to be due to anything we humans are doing.

2) Something is messed up with a satellite. I don’t know enough about the flight-paths of the satellites involved, but I’d guess they are in an orbit which takes them over both poles, so a messed up satellite could give this effect.

3) Someone has gone wrong with the pixel counting software which is used to estimate the sea ice cover, either: a) there is a long-standing bug which has just appeared due to an edge case, or b) they’ve applied an update which has affected the counting algorithm).

My money is on 3b, although 2 is a possibility. This doesn’t necessarily mean it’s undercounting now - it’s possible it was over-counting before. But either way, it makes it look as though “it’s worse that we thought”.

Hmm. There isn’t an important Climate Conference coming up shortly, is there?