June 2015 Status

June data is now fully available, so here is my commentary on the current sea ice state and the month just passed. Sorry but this is a long post...



DMI Arctic temperatures use the region north of 80degN, and as that data is popular I have taken to using NCEP/NCAR to calculate the equivalent metric. June 2015 is back amongst the typical band of temperature seen for the last ten years, and is well above 2013 and 2014. The issue of whether we are entering some cooler phase looks less of an issue, especially as the weather isn't really great for melt. Nonetheless Blaine has an interesting idea about the recent poor melt years over at Neven's blog. I voice some reservations below that comment, but think it's the best idea I've read on the recent increase of volume.

Surface temperature may not be very informative as summer melting ice pegs temperatures close to zero. While potentially still being influenced by some surface effects 850mb seems to me to be a reasonable compromise between temperatures aloft and the surface. June 2015 temperature at 850mb pressure level is also well above 2013 and 2014 and back in with the behaviour of the last ten years.

A plot of surface temperatures shows the limiting effect of melting sea ice on surface temperatures, the Arctic being ringed by above average temperatures. Temperature anomaly shown below is the deviation from the 1981 to 2010 average.

Temperature north of 70degN covering most of the Arctic Ocean picks up more of these higher temperatures.

Despite the warmer temperatures extent has not declined much, but area has, and this has led to a decline in compactness (area divided by extent). This is seen in data for the Arctic Ocean.

Arctic Ocean extent continues to decline at roughly average rate, this manifests itself in the plot below as extent for 2015 being roughly level through much of June.

Meanwhile area has declined as in years like 2012, 2007, and 2011, note that the recent rise in anomalies (below average losses) also happened in 2012 and 2007. But before rushing to the conclusion that this means a crash is imminent, note that in those years extent fell at above average rate, which is not the case this year so far.

This behaviour of extent and area is leading to an above average decline in compactness, i.e. the anomalies fall. This might be taken as indicating an increase in absorption of insolation within the pack which will lead to substantial losses later this year, but whilst an increase in absorption of insolation is a reasonable inference, imminence of a crash is not.

Turning now to the Peripheral Seas of the Arctic Basin (Beaufort, Chukchi, East Siberian Sea, and Laptev Sea), this group of seas plays a dominant role in setting the minimum extent.

Extent decline in the peripheral region is at average rates throughout June (and up to 2 July).

While area drops.

And as with the Arctic Ocean, this has led to an above average fall in compactness. Note that compactness began the large fall in June on 10 June when a storm hit, I will return to this issue shortly.

So the peripheral seas show the same behaviour as the wider Arctic Ocean, this being because they are a major driver of that wider behaviour.

I have recently suggested that late June (10 to 30 June) average compactness might act as an early warning for sea ice crashes. Now, with full June data I can look at what this potential indicator might say. The title of that graph should read Peripheral Seas, not Arctic Ocean. Black bars show the size of new records below the previous record extents for September.

My suggestion had been that for the post 2007 period a drop of compactness to 0.7 or below would indicate that a large summer loss is due. The final data point of this plot is for 2015, and it does not suggest to me that this condition is fulfilled. This is a speculative suggested method, and this year should be considered a test, rather than a firm prediction, however I am allowing it to influence my assessment of prospects for the remainder of the season.

2012 saw a record loss of ice and record minimum extent and area, this year was dominated by a low index Arctic Oscillation set up, with a strong Arctic Dipole (Purple over the Siberian coast, yellow/green over the central Arctic and Greenland. This acted to draw air in from the Pacific and Alaska over the Arctic Ocean, a pattern that persisted into the summer of that year. This was the typical pattern for years 2007 to 2012.

This year the pattern in June has a smaller mass of high pressure over Greenland, and much of the pack was dominated by low pressure through June. It is worth noting that Screen et al find that perennial ice survival is favoured by dominant low pressure conditions early in the summer.

The first half of June shows a strong Barents low and a reasonably strong Greenland high, the track of the 10 June storm is seen across from Barents towards the Bering Straits.

The dipole between the high over Greenland and the low over the Barents sea seems to have supported strong export through the Fram Strait. The graphic below is a close up of vector wind, but it is rotated anti-clockwise. It shows strong southwards flow out through the Fram Strait (large red/yellow region).

This increased flow of ice kept area up through the first half of June in the Greenland Sea.

Subsequently Greenland sea area fell as the winds dropped reducing export and allowing melt to take over. During the second half of June low pressure came to dominate the peripheral seas of the Arctic Ocean.

 As mentioned earlier, the storm on around 10 June led to an above average drop in compactness in the peripheral seas, but this was not sustained. Wind vectors after 14 June show part of the reason for extent failing to fall, net winds were generally directed out from the Central Arctic towards the peripheral seas (lower half of the graphic below). I suspect that the resultant movement of ice has acted against recession of the ice edge, the lack of fall of extent anomalies suggests that extent loss has been normal for the 1981 to 2010 period, unlike most recent years.

I have previously discussed the US Navy's HYCOM model as possibly providing some evidence for a crash this year. The model I originally used (HYCOM-Global) has now been put behind a password, suggesting they didn't want it used for such forecasting or speculation. The HYCOM-Arctic model is still available though, and despite showing a better ice state than HYCOM-Global still shows stunningly thin ice over the peripheral seas.

Whilst I think a crash this year is very unlikely, the prospect for a substantial loss taking us down to 2007 or 2011 levels is a reasonable one. However in the Beaufort sea I previously suggested that loss would moderate as the ice edge hit the multi-year ice, this has indeed happened, with losses through much of June being slightly below normal despite an early start to melt in Beaufort. Beaufort losses will probably pick up, but a melt out in Beaufort looks unlikely to me.

Having touched on HYCOM, it is now time to turn to PIOMAS as that data is now available.

First, here's the usual plot of PIOMAS June average grid box effective thickness.

The June average thickness doesn't suggest much thinning off Siberia, however the final day of June daily average grid box effective thickness does show such a thinning.

Looking at volume from regions, the following graphic shows Central Arctic volume on the right hand axis, all other regions on the left hand axis. The scale for the Central Arctic is four times that of other regions.

It is clear that the volume increase from 2014 remains, as can be seen by comparing the regional volumes in June 2012 and June 2015.

This comparison is problematic, at a total of 2.023k km^3 above June 2012, the May comparison was only 1.300k km^3 above, the June comparison picks up on the large drop of volume in June 2012. Comparison with June 2013 gives only a 0.921k km^3 increase this year, whilst this is half that of the January 2013/2015 difference this may be impacted by differing growth rates over winter.

However Chukchi shows substantially reduced volume, and volume throughout the peripheral region is not substantially more than the same time in 2012.

I will leave comments about the spring volume loss until later next week as this post is already long enough. However in closing I will show PIOMAS Gice sub grid thickness contributions to overall volume for the whole PIOMAS domain.

Following the top line of the stacked bars shows what overall volume has done, this June 2015 reveals the persistence of the 'slug' of volume within the sea ice system. How will this fare through the coming summer and into next winter? A lot depends on how the season will pan out.

My prediction for SIPN stands (4.51 to 5.79M km^2 central value 5.15M km^2), I see nothing that makes me think it is doomed to failure. My CT Area prediction was 3.0 to 3.8M km^2, central value 3.4M km^2. The SIPN prediction includes years 2008 to 2011, 2013 and 2014, while the CT Area prediction includes years 2008, 2009, 2010, 2013, 2014. Using two datasets and different statistical methods these predictions are largely congruent. But both methods overshot in hindcasts for 2007 and 2012, here however my suggestion that late June compactness may predict years like 2007 and 2012 steps in, and that method does not suggest that a repeat of such losses is likely this year.

That said I think 2015 may finish at the bottom end of my range for SIPN, a small fail on the low side wouldn't surprise me (less than 0.2M km^2). Conditions in the peripheral seas suggest that although extent loss at present is at around average rates now, this may well increase over July. Furthermore another speculative pattern might come into play. I posted in 2014 about an apparent oscillation in August extent loss, with only three cycles this is very speculative. If this is a real effect then strong losses in the East Siberian and Chukchi Seas should happen in August 2016, but this year should be in the upswing to that, and we may see those regions contribute to strong losses in August this year.

So I don't expect a crash this year, but equally I would be very surprised if September extent were a high as in 2013 and 2014. I'm expecting a fairly normal post 2007 September minimum.