Tuesday 4 August 2015

PIOMAS July 2015

PIOMAS gridded data is out, well partially.As I type the gice data and daily data hasn't been updated, but I'll proceed with what heff (grid box effective thickness) shows.

In summer I often use NCEP/NCAR reanalysis temperatures north of 80degN to be comparable with DMI temperatures. July 2015 average surface temperature north of 80degN is the highest in the NCEP/NCAR reanalysis data since the start of that series in 1948.

So what of sea ice volume from PIOMAS?

First three map plots. Note that the following are for July average, not the end of July, and the thinnest ice thickness at 0m thick could probably be left out to give a plot that better matches expectations based on observed extent, however I have resisted the temptation to do this because I'd rather give the data 'as is' wherever possible. When the daily gridded data is uploaded I will post again with data for 31 July 2015.

July 2015.

July 2012, for comparison.

July 2012 and 2015 are rather similar, the major difference being the thickest ice in the Central Arctic.

And here is an animated gif of May, June, July grid box effective thickness for 2015, showing the evolution of the July average pattern.

Regional volume breakdown for July shows that July is not much down on 2014, data for 31 July 2015 might show a further decrease, but losses in the Central Arctic have not been remarkably large using July average data.

Turning to the volume for the Arctic Ocean I have, as usual, broken down the volume into contributions from various bands of grid box effective thickness.

Overall volume is slightly down on 2014, but there has been an increase in volume from grid boxes reporting thickness of 4m and above. So what of comparison with previous years?

Comparing each region's volume in July 2015 with July 2012, the difference is plotted below.

As is to be expected from extent and area observational data, the PIOMAS model shows more volume in Beaufort and Laptev, Hudson and Baffin also have more ice volume present through July 2015. However once again the Central Arctic shows the massive volume gain persisting.

That is not to say that volume has not been lost. Comparing July 2015 to July 2014 the volume of the Central Arctic has dropped by 430 cubic km.

It has been notable that the progression of 2015 is rather similar to 2012 in terms of extent, area, compactness, and the atmosphere (see previous post), so how has the volume increase since 2012 evolved since January?

Using data for the Central Arctic, the volume increase is seen to have been persistent and level since April. January, February, and March show an increase in the difference, this may be due to ice growth following the 2012 record being stronger than in the same period in 2015. (I got mixed up there, see my reply to iceman in the comments.) However this might not fully account for impacts through July, to look at that I need daily data to examine changes through the month.

So using monthly data there is little evidence that the volume increase is being eroded this summer within the Central Arctic region. The July 2015 increase above July 2012 is explored in the table below. Volumes from five bands of grid box effective thickness are calculated for the Arctic Ocean and Central Arctic as 2012 volume subtracted from the 2015 volume. Central Arctic volumes are then subtracted from those of the Arctic Ocean to produce volume for the region within the Arctic Ocean that is outside the Central Arctic, this region is known as 'Other'. The resulting set of volumes is then expressed as a percentage relative to the overall difference of volume between 2012 and 2015.

Percentage Contribution to Overall Volume Increase
Central 0 to 0.9m -22%
Central 1m to 1.9m 9%
Central 2m to 2.9m -12%
Central 3m to 3.9m 91%
Central 4m and above 17%
Other 0 to 0.9m 10%
Other 1m to 1.9m 3%
Other 2m to 2.9m 3%
Other 3m to 3.9m 1%
Other 4m and above 0%

The sum of these percentages is 100%, but there is some offsetting with 0 to 0.9m and 2m to 2.9m being lower in July 2012 than in July 2015. Including this offsetting, 91% of the volume increase from July 2012 to July 2015 is due to ice between 3m and 3.9m thick within the Central Arctic.

This situation can be seen in the following plots of volume distribution with grid box effective thickness.

The difference above 3m thick persists from the Arctic Ocean plot into the Central Arctic plot.

I have tagged on the following based on PIOMAS daily volume figures (not gridded) which are now out.

The PIOMAS 'Main Series' of daily volume figures provides data for 31 July 2015, which can be compared with past years.

July loss has been large with the Arctic Dipole that dominated July, giving a loss against the recent trend of reducing late summer losses.

As a result the spring melt has persisted through July, with further drops in volume relative to average after the summer solstice.

July has been such a remarkable month that using the monthly average to examine it isn't really satisfactory, so I will post again when Dr Zhang has uploaded the daily gice sub grid thickness distribution data.


Dreessen said...

Bang-up job as usual. Hey, I was wondering if you finished reading that Hansen paper yet? I'd do it myself, but you're far more of an expert on ice than I am, and I trust your opinion.

Chris Reynolds said...

Thanks Dreessen,

No I've not finished the Hansen paper, just haven't had the time. Sorry.

Chris Reynolds said...


See the 'Hello Dipole' thread re Hansen. I have finally finished it.

iceman said...

Your two plots of volume distribution with grid box effective thickness are quite striking in showing that the rebound in the thickest ice continues even during a more "normal" melt season than 2013 and 2014. This seems to reinforce the notion of the 2010 melt being so exceptional in taking out a lot of the ~4m thick ice.

In your analysis of the 2015 minus 2012 graph, I'm puzzled by "... may be due to ice growth following the 2012 record being stronger..." Aren't Jan-Mar in that plot after the 2011 melt season? The graph at end shows volume anomaly rising in early 2015, vs. approximately flat in 2012. And yes, the steepest Jan-Mar rise in the anomaly trace was in 2013 following the record low - but I don't think that's what the other plot shows.

Chris Reynolds said...

Hi Iceman,

Yes, I got mixed up.

The volume anomaly plot, last plot, shows that the Jan to March 2012 volume increase was about 0.5, this is similar to the increase in 2012 to 2015 difference of about 0.4. That is what I meant, but while writing the post I got mixed up.

Neven said...

Thanks a lot for this, Chris.