Greenland Calving as deduced from PIOMAS gains

[benjamin]

I find the detail of the PIOMAS volume data extraordinary and extraordinarily interesting.
As well as the average decline in Volume which Wipneus portrays so effectively, it interests me that the average annual re-growth of ice in the winter months has increased in recent years (but not as much as the annual loss in the summer months of course). It is especially odd as the average winter temperature in the Arctic has been well above trend in recent years…which should make re-growth slower. But a contribution at least to the higher apparent re-growth must come from the known increased winter calving from Greenland.
We know that individual glaciers have speeded up several times but it is difficult to get a holistic picture of total loss: could the PIOMAS data provide a clue to such?
Overall the total ice gains and losses are very up and down by year but a three year average smoothes it considerably. And a best fit of the smoothed data readily reveals the trend. See first two attachments.
 
 
It seems that the annual net loss was fairly steady at around 500 km cubed during 1994 through 2006 with Annual gains in that period unchanged. But after that the gains and the losses have both risen sharply (of course 2007 was a devastating year which changed the whole shape of the polar cap).  ( I have added an estimate for 2013 gains as the most important component ..the 2012 starting point …is already available)

Any increases in gains in the winter represent even more ice that must be melted in the summer to show a net loss over the year. I submit that at least some of this extra gain would derive from additional winter calving in Greenland. What is more it is certain that summer calving from Greenland has speeded up at least as much as the winter calving. All of this extra ice also would need to be melted before the PIOMAS net loss recording was struck.

Although one cannot easily impute the summer calving in the same way as the winter calving increase one might reasonably assume that the change in calving actually happens only in the summer and that it builds steadily from spring through the summer then reaching a new autumn plateau until the following spring. If so then the summer calving increase would be, in total, midway between the implied total winter rates either side.

(This increased winter and summer calving, as well as adding to the Arctic ice stock in the short-term also increases sea levels directly and immediately.)

Then, adding that “best fit” additional winter and imputed summer calving to the annual recorded net losses shows what I would regard as a more appropriate figure of ice loss from the “Pole plus Greenland”. See third attachment
 

The annual gross ice loss -blue- is rising very sharply and evidently much more sharply than the “simple” reported annual loss-pink (more than four times in 2012). The pink is shown as negative representing the remaining summer sea ice.

Of course we have been in the “honeymoon” period during which much of the net ice loss is from the floating stock without sea level impact. The summer floating ice has now all but gone. The length of the virtually ice-free summer period is already extending. So we have now entered the next phase during which Greenland is increasingly yielding up its ice.

Such imputed losses imply a very large loss of ice from Greenland in recent years and therefore a significant increase in sea levels (in the single year of 2012 the imputed Greenland loss represents 1.08cm of sea level). This is very much higher than the other present estimates built up from individual glacier measurements and by extrapolation.

I would be very pleased if others could comment on this approach of mine based entirely on the PIOMAS data which itself is seemingly gaining increased credibility.

[crandles]

it interests me that the average annual re-growth of ice in the winter months has increased in recent years (but not as much as the annual loss in the summer months of course). It is especially odd as the average winter temperature in the Arctic has been well above trend in recent years…which should make re-growth slower.

The main reason for this is that heat can be lost through thinner ice faster than through thick ice. The more heat that is lost the more ice that forms. So thinner ice regrows much faster.

I expect this completely dominates any effect from Greenland calving.

[ChrisReynolds]

Benjamin,

Interesting read. I have to agree with Crandles, I think this is due to the ice thickness/growth feedback. Thinner ice or open water generates ice more efficiently.

Furthermore I've not read of PIOMAS taking into account influx from glaciers. The PIOMAS model models ocean and sea ice response to NCEP/NCAR forcings. There is no mention in the literature I've read that glacier flow is included, nor is there mention in the gridded PIOMAS data.

However PIOMAS volume is below CryoSat 2, maybe glacier discharge is a part of this. Without figures of net discharge of ice from glaciers in N Greenland and the Canadian Arctic Archipelago I can't hazard a guess as to how much of a factor this might be.

If you need more to read about PIOMAS I recommend:
"Modeling Global Sea Ice with a Thickness and Enthalpy Distribution Model in Generalized Curvilinear Coordinates." Zhang & Rothrock 2003.
http://psc.apl.washington.edu/zhang/Pubs/POIM.pdf
Which is the base paper describing the details of the model. Frankly the maths is beyond me to delve into deeply, but sect 2a describes the model and by going through the text and the general form of the equations you can get a general feel for what goes into the model.

Schweiger at al 2011 states:

Daily mean NCEP/NCAR reanalysis data are used as atmospheric forcing, i.e., 10-m surface winds, 2-m surface air temperature (SAT), specific humidity, precipitation, evaporation, downwelling longwave radiation, sea level pressure, and cloud fraction. Cloud fraction is used to calculate downwelling shortwave radiation following Parkinson and Washington [1979].

Which runs down what the current version of the model is forced with.

It's good to see someone else using Excel! 

[benjamin]

Thank you for the insightful comments and also the link to the PIOMAS model.

It isn’t totally clear to me that the model excludes calved ice but it is implicitly as river run-off is excluded (albeit corrected for in salinity terms).

But as you say CRYOSAT data clearly must include all ice.

Although it is a relief to me if the PIOMAS doesn’t include the calving it still does show a big up-tick in summer warming in part reflecting the much lower albedo I suppose. And CRYOSAT data suggests that the summer warming is even greater than PIOMAS calculates.

Greenland of course isn’t warmed by the sea underneath in summer but neither will it benefit from this enhanced cooling in winter. So, greater heating from its lowering albedo and more rain from the surrounding warmer and open water still spell big and growing problems for its ice loss.

Have you seen any good efforts at calculating Greenland’s losses?

[ChrisReynolds]

You're probably best asking about Greenland mass balance in one of the posts on the Greenland thread. But first maybe try Google/Google Scholar. I only follow Greenland in passing so wouldn't like to advise.

[Apocalypse4Real]

Chris and others,

Check out the article on Greenland snow melt during January-February 16, 2013.

http://blogs.discovermagazine.com/imageo/?p=291#.USOZhOghM2U

[ChrisReynolds]

Actually, as there have been no other takers regards Greenland mass balance. Here's Shepherd et al 2012. "A Reconciled Estimate of Ice-Sheet Mass Balance"
http://xa.yimg.com/kq/groups/18383638/836588054/name/Science-2012-Shepherd-1183-9.pdf

[benjamin]

Thanks guys for the steers.
The snow site is interesting and I am very eager to crack into the mass balance.