Pardon the pun, but watching the ease of ice transport in some of the preceding animations is truly chilling. Whilst there are still Freezing Degree Days getting clocked up in the High Arctic, it does seem a case of "too little, too late".
The fact that, purely in terms of extent, the 2017 melt season appears to have somewhat stalled at the moment, simply serves as a reminder of what transpired during 2012. If one ranks the March, April and May NSIDC monthly averages, 2012 currently occupies 14th, 20th and 13th lowest positions respectively. However, the horror show only really began in June of that year. Given that the present extent is nearly 1 million sq kms lower than that recorded for the equivalent date in 2012 should certainly give cause for apprehension.
One of the (many) ways of graphically illustrating what has been happening to Arctic sea ice is to display the number of days in each year that either has been, or indeed still is, amongst the lowest three recorded for the date.
Possibly the best example - certainly the best I can think of - is to show how 2007 has fared. According to the inhabitants of flat-land, that year was supposed to represent the nadir for Arctic sea ice. A major strand of the "logic" - and I use that word very advisedly - behind this hypothesis, was that the September minimum had increased in each of the two following years. Using NSIDC average monthly values, the September figures for Arctic sea ice extent were...
2007 = 4.32 millions sq kms
2008 = 4.73 millions sq kms
2009 = 5.39 millions sq kms
So, there is no arguing with the fact that 2009 > 2008 > 2007. The format of that inequality, allied with the fact that there had been anomalously high temperatures recorded in the Arctic during the 40's, had been used to peddle the myth that 2008 and 2009 heralded the fact that the turning point had been passed in an ~ 60 year cycle. This immediately gained uncritical acceptance in flat-land, as the average climate change sceptic wouldn't know what genuine scepticism was if it jumped up and bit them on the arse.
However, the following inequalities between NSIDC average September extents somehow seemed to have been overlooked...
1983 > 1982 > 1981
1992 > 1991 > 1990
2001 > 2000 > 1999
(Incidentally, the increase between 1990 and 1992 was quarter of a million sq kms greater than the much-vaunted 2007 - 2009 "recovery".)
{Of course, things have moved on since them, and their current meme is that the Arctic was effectively ice free nearly 100 years ago. There were so many obvious flaws in the original 60-year cycle meme, that even committee members in the House and the Senate would have had trouble keeping a straight face.}
Anyway, getting back to 2007, according to the JAXA/IJIS/ADS database, by the 31st December that year, it had registered the following daily figures...
153 days were lowest for that date
104 days were second lowest, and
105 days were third lowest. (That adds up to 362, and the other 3 days were 4th lowest.)
However, as at today, the equivalent numbers for 2007 read as follows...
Lowest: 4 days
2nd lowest: 70 days
3rd lowest: 46 days
The demise of 2007 from its position of preeminence can be seen in the attached chart. The first column shows how 2007 stood at year-end, and the subsequent columns show the decline at the end of each of the following years. The final column tracks in near-real-time as 2017 data arrives. So far this year, there have only been 8 "lowest 3" instances remaining from 2007. Each of them has been surpassed in 2017, with 2 dropping to 3rd lowest, and a further 6 dropping to 4th. As the 2017 melt season gets into gear, it will be "interesting" to see how much further 2007 will sink.
As we frequently say in Glasgow...
sic transit gloria mundi