Arctic sea ice stages a late spring fight back

I notice that the latest Arctic sea ice extent for the 14th of May was only the fifth lowest for that particular day of the year, which means it’s staged a late spring fight back from a disappointingly low maximum in early March. Having said that, although it’s higher than last year at the same date, it’s still lower than the record low year of 2012.

Figure 1

I have revised my estimated September minimum up from 3.7 to 3.9 million square kilometres of sea ice. This doesn’t come close to breaking the lowest ever minimum of 3.34 million, which occurred on the 16th of September 2012. If the sea ice pundits are correct though, the pack ice has never been thinner and the Arctic could see an unprecedented collapse in sea ice this summer  – we shall see how right they are in around four months time.

Figure 2 (yellow outlined series is my estimate)

Arctic sea ice on the mend (a bit)


The recent demise in the Arctic sea ice extent reminds me of a story that went around about the death of Mark Twain, which he refuted by saying “The reports of my death are greatly exaggerated”. It didn’t look good in September and October as the graph of the sea ice extent showed how slow the Arctic was creating new sea ice. So I decided to look at the figures from the NSIDC in a bit of a new way. The graph above shows daily anomalies, that is the value on the day, divided by the 1988-2016 long-term average [LTA] for that day of the year, multiplied by a 100 to get a percentage. The graph as you can see had two distinct dips, one in early September just after that early minimum when the anomaly was around 65% of the LTA. The sea ice bounced back very quickly from that minimum and before the end of October was above 76% of the LTA. Those gains were quickly lost though, and by the middle of October the values had dropped to less than 68%. But interestingly, in fits and starts, the sea ice extent has bounced back and is now above the 85% mark. If you look back at summer 2012 just before the summer minimum things were a lot worst than in September with the LTA  less than 55% of the LTA for a few days. Here’s a bit more of a close-up of the last few months.


This method of graphing daily anomalies is very sensitive to any daily changes in the sea ice extent, and I think it’s one of the best ways of keeping an eye on sea ice extent levels both in the Arctic, and the Antarctic, and talking of the Antarctic, here are the latest daily anomalies south of the equator.


The above graph shows very nicely the rise and fall of Antarctic sea ice in recent years. If you remember up until early 2015 Antarctic sea ice had gained a lot of new sea ice. In fact the 2014 season set a new maximum extent of over 20 million square kilometres, but since then things have been going down hill, and since early October of 2016 the sea ice anomalies have been tumbling. At the moment (3 December 2016) the anomaly stands at 84% of the LTA for that day, not a record low by any means, but certainly this season’s melt looks very aggressive, and these daily anomalies are the lowest since early 2011.

Sea ice extent is certainly in crisis, in both the Arctic and the Antarctic, but at least in the Arctic the season seems to be finally getting itself into gear a little.

The media seemed to have now picked up on this autumn’s shenanigans going on in the Arctic, a little bit late, but never mind.

Here is a list of recent articles that I’ve written on the subject, and I think it’s fairly evident that I have a bit of a thing for sea ice.

So you can see when I add a comment to the Met Office blog, and 10 days later it’s still sat there “awaiting moderation”, I do get a bit irked. I’m not sure what’s particularly wrong with it, there are no swear words or personal insults, just a few comments about the slightly arrogant tone of the piece, and how slow off the mark they’ve been (in my opinion). So here is what I said and you can be the judge, because after 10 days I think they binned them – so much for democracy!


Second lowest Arctic sea ice minimum on the 10th or was that the 7th? (updated)

The Arctic sea ice looks to have reached its minimum on the 7th September, which is four days earlier than average. The sea ice extent bottomed out at 4.083 million square kilometres making it the second lowest since records started in 1978 – well that’s according to the data file that I’ve just downloaded!

Strangely, according to the data that I download from the National Snow and Ice Data Center [NSIDC] the minimum occurred three days later on September 10th. As I said in my introduction on the 7th the value was 4.083, but according to the news item that I’ve included below, the value on the 10th was 4.14 million square kilometres and tied it with the year 2007, which according to the data file is third.


Here’s how I see the latest annual Arctic sea ice extent rankings using the latest data that I’ve just downloaded ending on the 15th of September.


And here’s a snapshot of what the data file looks like that I download just to prove to myself I’m not going mad.


To add to the confusion, the values in the NSIDC table of ten lowest Arctic sea ice extents varies ever so slightly from the ones in their data file which I download, which makes for changes in the rankings. You would have thought that the values in the data file should be the ones used to generate this table.capture6

Here are the daily values since the end of August. As you can see the 10th of September has an extent of 4.16 and not 4.14 million square kilometres as mentioned in yesterday’s news from the NSIDC, and on this date the extent is already increasing.


All these daily values translate into the following chart with the minimum occurring on the seventh and not the tenth of September.


I have monitored polar sea ice using data from the NSIDC for almost five years now and never had much problem until June of this year when the NSIDC transitioned to using data from the DMSP F-18 satellite due to issues with the F-17 satellite. I’m now totally confused, I realise that all the differences are very trivial when you are dealing with millions of square kilometers of ice, but I still wonder where I’m going wrong, If you know then please drop me a line!


Someone has kindly pointed out to me that the NSIDC now use a trailing five-day mean to construct all their graphs and ranked tables of the extent and time that minima and maxima sea ice extent occurred. So it is the same data that I download and use it’s just smoothed. Here’ a further snippet from their website that explains matters.capture16It looks like they adopted this new way of working in June of this year,  because a footnote further down that same news item goes on to say:-


Personally using a running mean has only muddied the waters regarding sea ice extent, should they have used a centred or a leading moving mean rather than a trailing mean when referring to maximum or minimum extents? Why not just accept the daily values warts and all, if they’re good enough to construct a five-day running mean why not just continue using the daily values? The daily minimum and maximum will most likely never coincide with the minimum or maximum gleaned from the five-day trailing mean, which to me is just plain confusing.