I have highlighted a bit of your post which I think I disagree with. Yes, I am being picky.
It is interesting that the worst years are 2005, 2006, 2007 and 2016. Of course, these are for the whole year and in 2012 we didn't have too much days of breaking record, but they happened exactly when they count: at the end of the melting season!
The 2012 end of season melt could be regarded as least important in 2 ways:-
- early season melt (as in 2016) maximises positive feedback from insolation and from that ocean warming to inhibit winter sea ice growth,
- by late August / September the sun is heading South and insolation is in rapid decline,
- some of us still think maybe it is only when winter sea ice reduces sufficiently will we see an ice-free summer (though the Jury is still out on that one - 2012 does NOT support that speculation).
I wonder what the graph if it included freezing months only would look like. Is there any correlation between winter sea ice volume maxima and summer minima ? (I bet someone on ASIF has done it).
Thank for your comment, but I am not sure that I understand it.
2012 started a little low in January and February, but in March it had an important refreeze. So, the daily records on the first five months were not too many, as you can see on the first graph, that compares 2012 with 2007, 2010 and 2011. Even in April, 2012 was above the 1981-2010 average. The melt in June was important, but the real difference appeared after the Great Arctic Cyclone, on August. So, most of the 2012 daily records happened on the second semester, as you can see on the second graph.
2016 was completely the other way. We have daily records almost the whole year, except for July, September and the first half of October. So, I was expecting a minimum record at 2016 lower than what finally happened. I don’t agree with some conclusions that NSIDC makes public, specifically, that 2016 was the fifth lower year. From my point of view, 2016 should be cataloged as the second or at least the third worst year on record, because that it is what it was, in area, daily extent and volume. But well, that is another story.
2016 had 182 days being the daily lowest (against 1979-2015) and 312 days being among the three lowest on record.
2012 had 125 days being the daily lowest (against 1979-2011) and 211 days being among the three lowest on record.
That is, according to my calculations. I compare the days on the year as NSIDC do on a Charctic graph. That is, I compared day 60 on 2011 with day 60 on 2012. Being 2012 a leap year, the 60th day will be Feb 29, while on 2011 would be March 1st. These is what we visually see on Charctic, but the numbers could change a little, if I erase the feb 29´s and I compare the dates matching for the day on the month.
Regarding your last question, I believe that it is not a rule that the year that has a winter minimum, will have a summer minimum. But surely, even that 2012 does not match that rule, it should help to start with a winter minimum. So yes, I agree that there is a bigger possibility of having an ice-free Arctic, on years like 2015-2017, that the melt season starts with low ice. Specially 2017, that we are starting with the lowest volume on record, according to PIOMAS.