Sunday, 31 March 2013

Well I'm not going to post on April 1st...

...that could cause all sorts of misunderstandings.

Here's a rather late Monthly Miscellanea post.

Last year I posted about a cold snap in early February, and what I saw as a link to the Arctic. Well a similar situation, linked to a large and persistent high pressure over Greenland has led to a cold air outbreak over Europe. As Stefan Rahmstorf points out this is a similar pattern to that seen in various modelling studies. So it seems that yet again we have severe winter weather associated with the loss of Arctic sea ice.

The BBC Weather Presenters have been saying this is due to the Jetstream, which is as daft as saying the house burnt down because of flames. However when I go searching for an article to back up some casual sniping on this matter I find this:
The coldest March for over 50 years is spilling over into a wintry start to April and the latest long-term charts bring little cheer.... ...I was struck, therefore, by the forthright views of a Peak District farmer I spoke to this week. He was adamant that the climate has changed in recent years and said that spells of wet and dry, cold and warm are more prolonged now. His belief is that we get stuck in ruts of weather nowadays and we lurch from one to another, impacting directly on farmers' livelihoods. The last year is testament to his views and the science backs him up. Be it through global warming, melting ice sheets or changes in the oceans and the jet stream, there is a strong signal that the climate is indeed turning more extreme. So while this April will eventually warm up and spring will emerge in earnest, it seems only a matter of time before we lurch into yet another spell of severe weather.
Will that be another record wet summer, is the new circulation simplistically related to loss of sea ice, or is there some intermediary, like the loss of spring snow cover?

From colder winters to cool wet summers, both due to the Arctic, when BBC Weather presenters start taking notice there really is a change in attitudes. BTW it's the coldest March for 50 years because of the very severe 1963 winter in Britain, 50 years ago.

Anyway this high pressure is part of the very low index Arctic Oscillation we've been seeing for some weeks now, AO Index. And the general high pressure over the pole is what caused the recent fragmentation event of the Arctic sea ice pack, as I discussed here, and as covered elsewhere. NASA have covered the event in an article, here. Anyway here's an IR satellite image showing the recent state of the pack.

Walt Meir of NSIDC is quote in the article, I'm still not convinced that the February storms were a major factor, to me it looks like the major fracturing occurred as high pressure shifted the pack clockwise, a movement that has now largely ended. However I do agree with what Dr Meir says in this passage of that article.
The age of the sea ice in this area was one of the key reasons this event became so widespread. “The region is covered almost completely by seasonal or first-year ice—ice that has formed since last September,” said Meier. “This ice is thinner and weaker than the older, multi-year ice, so it responds more readily to winds and is more easily broken up.”
I've been doing a lot of reading on PIOMAS, and the papers behind the model itself. I'm making a last effort to solve the problem of why PIOMAS shows a change in seasonal cycle following the volume loss of 2010. It's too intriguing a problem to go asking the PIOMAS team until I've exhausted all the resources at my disposal. Basically what it comes down to is this:

The graphic above, which I'll be using in a post to come, shows the thinning from April to September for all PIOMAS grid points with sea ice in April, against the initial thickness in April. There's something about ice over 2.5m thick that makes it thin less, and have less of a spread of thinning, than the ice below 2.5m thick in April. The spring melt is from April to June, and the difference is even more pronounced for that period. The question is - Why does PIOMAS show this behaviour, what specifically is it about the thinner ice that makes it have a greater spread of thinning and greater average thinning than thicker ice? It's not the characteristics of thicker ice with respect to brine drainage, because as far as I can see that's not accounted for in the model. The ice model is two layer, doesn't seem to age ice, so it's something to do with thinner ice alone.

I've also been looking into the possible causes of the new Summer Circulation pattern, covered by my Summer Daze posts. As a related aside I've been looking at a certain pattern of warming associated with snow line retreat. I'll post about that later today.

Right, now I'm committed to pulling my finger out and actually doing that!

PS, I'm not the only pessimist, Michael McCarthy of The Independent retires, here is his closing statement. Hat tip to Alexander Ac.


Kevin O'Neill said...

Chris - How does PIOMAS handle melt ponds? I assume the resolution isn't small enough to model them by physics, so there must be a parameter that they can tune to account for increased melt ponds on thin, young ice and the corresponding change in albedo.

Chris Reynolds said...


This is something I need to ask Dr Zhang. From the papers I've been reading there are two albedos - ice and snow. I don't recall melt ponds being accounted for.

Indeed when I search for 'melt pond' in Zhange 2003 "Modeling Global Sea Ice with a Thickness and Enthalpy Distribution Model in
Generalized Curvilinear Coordinates." I get no results!

This might explain why the anomalies increase then level after the spring melt. But as I say it's on my list of questions for Dr Zhang.

Kevin O'Neill said...

Chris - I don't find it surprising that ice < 2.5 meters thickness thins more than ice > 2.5 meters.

I'd guess the thinning should be non-linear and that the thinner the ice the more it thins.

I base this on a simple thought experiment. Take two slabs of ice, one 2m thick and the other 5m thick, and expose them to the same winter conditions. Come spring as air temperatures climb and incoming solar increases we should begin to see melt. Given the thermal gradient of ice, it will take more energy to melt the thicker ice to a depth of 1m than the thin ice to the same 1m depth.

I.e., there will be a lot more 'cold' ice near the surface in older, thicker ice. The ocean temperature sets the bottom ice temperature and the air temperature sets the surface ice temperature.

PIOMAS should be able to model the physics of this averaged across differing thicknesses without a problem.

Chris Reynolds said...

Thanks Kevin,

I'd been wondering about this mechanism, and very nearly asked if the thinner ice is 'warmer' in the post.

I think there's also the past pattern that thinner ice has been on the periphery of the pack that melts out, this accounts for the clustering near the line where thinning = initial thickness.

The relevant paper here is Winton 2000, "A Reformulated Three-Layer Sea Ice Model"