I have always thought that following sea ice extent using daily SII values from the NSIDC was not a good idea. Even following weekly, monthly or seasonal extremes didn’t seem to simplify the ‘noise’ that you got with line graph or bar chart. That’s why in my opinion a graph of moving 365 day average sea ice extents is ideal at looking at the long term trend in the last 30 years.
The Arctic has most people must now realise has seen a steady decline since 1990, although in the last year the 365 day average has recovered somewhat, although the best fit curve looks a little ominous.
In the Antarctic, sea ice extents where increasing slightly but steadily from 1990. Then towards the end of 2016 when they took an abrupt nosedive from over an average of 12.8 to less than 10.8 million square kilometres. This coincided more or less with the last El Niño event in the Pacific and which must have been one of the main contributory factors that led to this event. I’ll have to overlay the ENSO events on these two graphs when I get a minute.