« Permafrost, the tipping time bomb | Main | PIOMAS March 2013 »


Feed You can follow this conversation by subscribing to the comment feed for this post.

Chris Reynolds


If I'm wrong I'm happy to say so and move on. From the point of view of my own understanding it matters a great deal to me when I'm wrong - recognising it brings me closer to what I'm seeking, a better understanding.

Actually I disagree that we're just observing, we're making and testing hypotheses.

Espen Olsen


Yes you are right, I am just trying to be humble, because we are all in new waters, and the behavior of the Vikings in this regard, tells me I better be!


Are the small spots and squiggles on the Sea Surface Salinity maps an indication of melting at big cracks?

And, isn't that a huge East-West crack that just opened up at the edge of the MYI, from at least 20W to 90W, at about 85N?

Nightvid Cole


The "crack" ice still doesn't have a snow cover...


A Team's wonderful animation of 3/9 03:03 shows that first the sea ice flowed out of the Bering Strait to the Pacific. Once enough was gone from the region north of Alaska, the remaining ice could no longer buttress the main pack and the collapse ensued.

Hans Kiesewetter

I just (re) found a short animation of 20 years ice movement, compressed into 50 seconds. Illustrative about what is (was) normal movements. http://iabp.apl.washington.edu/animations/7998-05Xfast.mpg

[Fixed the link, N.]

Chris Reynolds


I strongly disagree. The movement into Bering was incidental. What opened up the cracking was the net clockwise motio of the entire ice pack, this concentrated tension in Beaufort. There the ice could not move with the clockwise movement because it was against Canada and between the MYI and Alaska. This is what caused the initial pull away from the Barrow coast of Chukchi, and then the progression of arc fractures into Beaufort proper, along the Alaskan coast, with the arc fractures not penetrating the MYI.

It is at this point that failure due to loss of buttressing occurred. Once Beaufort had collapsed the net Alaska-ward direction of wind was able to cause the fracturing that entered into the MYI, as two large parallel fractures.


The weather patterns that caused the Beaufort Gyre to fire up, are now more or less gone and according to ECMWF it's going to stay that way in the week to come. So, things will 'calm down' and we can now concentrate on the max. Based on last year I'd say we're in for a relatively late max again.

Espen Olsen


But within the last 24 hrs. several large cracks / fractures are now developing along the top of CAA coast, only a strip of +/- 100 kms of fast ice (with no fractures) remains from Isachsen Island and northward.


Espen, that's why I put 'calm down' between those two '. :-)

I just meant to say that there will be less intense winds blowing in a clockwise fashion, and so there should be less stress on the ice. That doesn't mean things are over. The Sun is slowly starting to rear its beautiful head, but we have to wait until May for melting to start in earnest.


Espen wrote

"Music wise the melting season used to be like Ravel´s Bolero, but now a days it is more rock like Guns N' Roses - Paradise City."

Hmmm.... Another selection: ACDC - "Highway to Hell"


Ironic humor and double entendre intended.

Chris Reynolds


I see an intermittent string of dipole set ups with high over the ocean and lows along Siberia, still directing towards the Atlantic. So whilst the clockwise motion that's caused so much tensile stress in Beaufort may desist, there may still be a large transport towards the Atlantic, the extended area in the Atlantic will melt during the melt season.


I see what you mean, Chris. I focus perhaps a bit too much on where those isobars are exactly. During the great Beaufort Crack those isobars were over the exact area of fragmentation, and many of them too. Now there are less isobars, and not so much over the cracked area, but I agree that it's still a dipole for a couple of more days.



After glimpsing the latest Ellesmere AVHRR, we have shredded the MYI. This is beyond the earlier ice fracturing mentioned in the NSIDC reports: http://nsidc.org/arcticseaicenews/

Would love to see an NSIDC update on this, and what the relationship is to the stream of warmer winds that blew over that ice area yesterday.

I'm out of space on my Arctic ice imagery website, so I am posting three images on the Forum for before, during, and after comparison.


Chet Murphy

Too many comments for typepad? I am not seeing new comments. The last comment I see is from March 05, 2013 at 13:14.


Slightly off topic:

I noticed that the NRL has stopped updating the HYCOM/CICE sea ice motion forecast. That would be a usful tool about now.



Chris, this is basically a "chicken or the egg" question. There being first-year ice over such a large expanse means the sea ice pack cannot hold out against a multitude of factors.

The sea ice north of Alaska has been running out the Bering off and on since December.

Any sort of clockwise movement is going to result in fracturing, and today it is fracturing over the remaining thin multi-year ice: see


Steve Bloom

Wow. The MYI pack is looking less and less intact.

Steve Bloom

Chet, look for the tiny double arrows at the bottom of the comments and use them to get to the final comments page (on third one now). Easy to miss.

Steve Bloom

Although he won't see that comment unless he resolves the confusion otherwise, will he? :/

Steve Bloom

Tenney, IMO Chris is exactly right. There hasn't been enough ice flow through the Bering to have such a consequence, although whatever amount there was couldn't have helped.

David Sanger

I'd vote for a template design change to replace "« | »" with "previous | next"

peter prewett

Why not previous, next at top of comments then no scrolling down which is difficult on tablet etc.

Or even page numbers at top.



Here is a fairly decent site for research on sea ice and the Bering Strait:



Hi SB,

The region north of Alaska is like a tipping point or a point where the cork is out of the bottle when the ice is gone just to the east of Barrow.

Steve Bloom

Must disagree, Tenney. But if you're correct, process should continue even if dipole winds fade, right? Or is there some other test for that hypothesis?

Chris Reynolds


19 February 2013, a high pressure moved over the central Arctic.

Then a large scale clockwise movement of the pack commenced.

This resulted in ice in Beaufort constrained by land and MYI being under tension. That ice cracked apart in a series of fractures.
The red area is the rough extent of MYI from ASCAT image overlay. Note how the MYI constrains the arc fracturing just as much as does land. This shows that the fracturing event was a consequence of the pack being under tension. Similar failure mode is seen within the Canadian Arctic Archipelago during summer, when end buttressing is removed. So is this in support of your claim that Bering outflow is the cause? I don't think so. The small amount of Bering outflow didn't leave open water in its wake.

Furthermore the net movement of the pack is substantial: Here's the net movement of ice between 19 February and 5 March 2013.
Arrows connect common points that have moved between those dates. Dotted areas are regions of ice that show no movement.

I don't want to labour the point, but if there's a better explanation of the process I'd be interested in reading it.


I'd vote for a template design change to replace "« | »" with "previous | next"

Thanks for the tip, David. As usual, I didn't know this was possible.

Chris Reynolds

PS check out ASCAT:

See, for example, msfa-NHe-a-2013049 and 47, get them open and flick between them to see what Bering outflow was doing before day 50 (19/2/13).

R. Gates

In regards to the eventual end of the MYI (surely some year soon it will all be FYI, every year) one can only imagine what this kind of winter gyre rotation will do to first year ice.

Overall though, while the very detailed blow-by-blow accounting of the cracking and transport of the MYI has us all captivated, one can't help but think of a crowd gathering around an operating table watching a patient slowly die-- only in this case, and unfortunately, there isn't any medical staff willing or perhaps even able to save the patient. These are the final death gasps and contortions of an Arctic that has been the way it has during much of the rise of human civilization-- but shall be no more.



I concur. The models don't work. But again they are working on them. As has also been said, it is likely that when the Arctic ice finally does vanish in summer they will be able to incorporate those empirical data into the models to get a better picture of the future state.

However without the models we would be nowhere. Because lagging models at least set a baseline, which we can point to being exceeded.

I don't just read the cryosphere news but also a wide range of climate change material. It's clear that the politicians have very severe issues with accepting predictions in the 2050 - 2100 range. Without a total collapse of the Arctic ice, they will ridicule anything even approaching reality, causing more harm than good.

The models may not work but they do form a base from which we can move forward, comment, point out divergence etc.

I think that has merit even if they are daily being slated on the climate change reporting sites.

Just my position.

I Suspect Dr David Barber will be watching the "MYI" ice very closely to see if "rotten" MYI performs more like FYI or still like MYI....

Our world is becoming a more "interesting" place. I prefer boring.....

Dave C

It is somewhat morbid in a way. Still, I think of it as documenting history.

There have been plenty of local effects from global warming. Still, each individual event above the long term average could conceivably be dismissed as natural variation. Most areas haven't noticed huge changes from the 1 degree rise in temperatures.

The arctic ice disappearing is the first absolutely undeniable large-scale effect of global warming. There is simply no way to spin this one. If the arctic melts, global warming is happening. Period.

I don't know if humanity will effectively respond to global warming. But the melting of the arctic officially moves the debate our of the theoretical and into the actual.


"According to the BBC's Andrew Neil in a presentation in the City of London on Thursday, "Windmills" are far too variable and the Good 'ol US of A will soon be pumping oil and gas to the Gulf for export, whilst the Poor 'ol Arctic Ice Cap melts quicker than ever, swiftly followed by Greenland."
There is an issue that is neglected. Water. Most of NA NG and tarsand oil production depends upon large amounts of water. As seen last summer, because of the big stationary high over Greenland, a large portion of the US was in drought. Predictions for the future are long term drought and possibly long term flooding. Neither scenario is great for dependable water for oil/gas production. Also I would not bank on the AO as a help either as the big cyclone of 2012 showed ther could be big storms in that region, and predictions for that are bigger more frequent and longer lasting. All that is bad news for oil rigs as they may handle a 3-5 day GoM hurricane, but I do not think much can survive a 5-6 week AO cyclone.
All this saying, I do not think NA oil/gas is dependable as the big oil companies would like us to believe.
These events in Feb. just shows you that the Arctic should be the news centre of science for the next decades, because what happens there will impact the rest of the ice in the world and that will impact every nation directly. We then face the problem of how nations will react. Man in general has not shown itself to use the intelligence it was born with.


Perhaps they should focus the models more on the consequences of disappearing sea ice, instead of the disappearing itself (as they might not be able to model that before it's gone)?

Jim Hunt

Less in the way of CICE, more in the way of WAVEWATCH III in the Arctic Ocean?

Espen Olsen

Fractures are still developing north of CAA, and my estimate of the last fast sea ice is at about + / - 45.000km 2, a little more than the size of Denmark, and that is not a lot.


Artful D writes, "ftp site for L2 data from AMSR-2 has better resolution/more channels"

Anybody know the url for this ftp site? I have been looking at L3 data obtained by AMSR2 onboard GCOM-W1(SHIZUKU) but not seeing quite what you are referencing.

Thanks if you can describe route to imagery or post a sample of Arctic Ocean imagery!



We eagerly await any sign on the quality of the ice cover as we enter a summer that could bring an ice free North Pole for the first time in at least 5000 years (assuming there could have been some during the Holocene Optimum).
PIOMAS could give some info, IceBridge another bit. I still hold to my prediction 18 Feb, 4,0 mkm2 extent/2,5 mkm2 area. I’m not convinced that the anomalous cracking pattern secures an extra strong melt. A dipole and/or Sunny summer and ocean heat content will be more decisive.

I expect that PIOMAS will show further diminishing of the difference to last year. The cracks should have delivered anything between 50-150 km3 of FYI, while a lot of the earlier formed stuff has been pushed into the Atlantic sector. If PIOMAS does’nt, we’re in for trouble. Because it’s ‘all’ easy to melt FYI and a malign summer could really bite…

Neven wrote: Perhaps they should focus the models more on the consequences of disappearing sea ice, instead of the disappearing itself (as they might not be able to model that before it's gone)?

I tend to agree. At the present, we're like shopkeepers obsessing over the exact placement and angle of our store's security cameras, while thieves busy themselves emptying out the place. Or, for a possibly better analogy, we're like doctors so taken with guessing when I our patient may die that we fail to notice that he already has...

Werther wrote: We eagerly await any sign on the quality of the ice cover as we enter a summer that could bring an ice free North Pole for the first time in at least 5000 years

Of course, that will be followed by a "recovery" of historical proportions. I can already see October's WUWT headline: "In blow to alarmists, Arctic sea ice grows from 0 to 3 million km2 in just four weeks, the fastest freeze ever recorded!"


Just to catch up a bit on what Chris, Espen and others have already noticed: cracking in thick ice had stalled at the 120 W meridian for several days and it seemed the event might have run its course.

However it resumed yesterday with a vengeance, with six new parallel 700 km cracks opening in a few hours. (Some of these may reflect re-opening of 'hairline' cracks of a few days ago.)

There is still some relic thick ice left -- notably above central Ellesmere and in the Lincoln Sea (above Nares) -- but the event may be continuing. These are not so much arc-fractures but rather fan-out linearly from eastern Banks Island.

I animated Ascat for the first 6 months of 2012. It continues (with a 50-day offset) like 2013 but only up the Beaufort stage. So we are in new territory here. If you recall, 2012 was a record melt year.

The goat's head is clearly visible the whole 2012 season -- remarkably it is not a new feature. I am wrapping up dating its origin and quantitating its rotational and translational velocities over the last six months.

As Chris has been doing, we need to derive overall movement from radar directly rather count on canned or modeled products.


A-Team, if you want, I can use your material for a blog post on the Goat's head, if only for the title.


Neven -- sure, whatever (and re-position or re-purpose my posts as you see fit). It seems like we are pushing 1% of the total 25,000 comments right here in this one thread. Somehow we should be segueing into early melt season ... goat's head may not be the right handle for that.

This image below goes to show that a monkey sitting at the keyboard does occasionally peck out the letter-perfect Shakespeare sonnet. This colored radar image of 10 Mar 13-- somehow cleanly discarding first and second year ice -- comes about simply by converting RGB to CMYK, followed by discard of MYK! In other words, the cyan channel alone has the goods. (In Gimp: colors -> components -> decompose).

There are quite a few stable features in this older ice -- the goat's head is only one of these. That is, structures that can be tracked for five months are not artifacts of a particular day (such as local snowfall, melt or harder freeze).

Overall, I'd say an awful lot has been overlooked in archived imagery. A lot of these scripted model products are not coming to grips with the reality seen on the ground. These specific changes in the block properties of older ice may play a much more important role in the 2013 melt season than in the past -- perhaps determinative in their own end game.

 photo cmykThickIce_zpsfc7e7566.jpg



Thanks for the imagery, it makes this event more real and sobering.

This is not the end of ice fracturing. The 961 mb SLP over Kamchatka is forecast to move over the ESS, Laptev, and emerge near Svalbard as a 985 MB SLP during the next five days 0000 15 March. During the same time period, the 1040 mb high remains over the Beaufort Sea. See:


Espen Olsen

CAA / Ellesmere:

The Sea Ice Grinder is doing the job, less and less Fast MYI is surviving, and the process is continuing.

Craig Merry

Great posts here. Fascinating. Keep it up.

Neven, I believe it's a bit too late for models to concentrate on the "when" and focus on a world without -at least- MYI in the Arctic and vulnerable masses that could rise sea levels quickly in a short amount of time. As a Poli Sci major working in government, we have not done enough to prepare for at least the next decade.

Chris Reynolds

Anybody else had an email from Dr Schweiger apologising for the delay in releasing PIOMAS - gone to download and found it's still only up to 31 Jan?

I ask because I had a problem last year with others having PIOMAS data while I was repeatedly getting the old data.


Hi, Chris. Yes, I asked Dr Schweiger if there were any updates. According to him it should be updated now, but I don't see any changes in the graphs (downloaded the data two hours ago, but no change).


EO said , "fast sea ice is at about + / - 45.000km 2"

Could we take that as about the very lowest that sea ice extent could go this summer even if it is unusually stormy and warm?

Or will even this be continually degraded over the coming months?

Steve Bloom

Goat's Head Soup seems like way too good of a post title to pass up.

Neven, if the models can't first get the sea ice loss right it doesn't seem they would be very reliable in projecting the aftermath. Recall that maybe the single worst model failure is their inability to show the known extreme Arctic amplification (18C +/- 4C) of the later Pliocene (camels on Ellesmere!). I think we all strongly suspect that the sea ice behavior we're seeing now is not unrelated.


A source I have in the US White House responded to a query:

Arctic sea ice area and extent in March/April are no predictor of September minimums.

That said, the 2012 summer average Arctic temperatures were unprecedented relative for which we have instrumental records. (We noted they are still above average in Spring, 2013).

There is an upcoming paper in Nature showing that recent Arctic summer temperatures are unprecedented for the last 600 years.

The source concludes that historical records may provide a poor precedent for the future. Unusual events now could well be indicators of more to follow later in the season.


Really, did Obama say all that? ;-)


"Goat's Head Soup" Hah!

My thanks to Neven and all contributors.


On the 5th at 23:55 I wrote"If my guess above has anything to do with reality i expect the Alaskan bays to begin melting from close to the shore outwards, same in Mackenzie bay and adjoining areas,"
"Banks island too"
More ambiguous but somethings going on. There's also signs of an acceleration north of Barrow, and although the thick ice against the central coast[s] of of the archipelago appears to be growing I suspect this is just the prelude to it lifting clear in the tides over the next few days.
"Also if it makes it to Morris Jessup it will turn south hug the coast and separate the ice from the coast. Not too sure of the last might break through the archipelago instead, but give it 3-4 days."
Too difficult to see anything clearly related here but I remain convinced that the Pacific waters are eroding the seaward edges of Chris's Puramid http://farm9.staticflickr.com/8248/8534579476_7aa84d8919_o.gif
and driving the AW beneath it.
Tides more usually see the sea pulled towards the moon and then relax back in an elastic fashion, returning pretty much to where it was. Due to the particular geography of Bering strait any water passing through is too energetically charged to pass back again and thus we get pulses [of maybe 2-3 hours] of inflow twice a day and with the new moon tomorrow we should expect them to be larger. So I expect the cracks to start getting wider
the pyramid to shrink and the channels through the archipelago to become more active.
All that said I find it difficult to beleive this would be happening if the winds were'nt driving so much out of Fram


>"if the models can't first get the sea ice loss right it doesn't seem they would be very reliable in projecting the aftermath."

If you reduced the sea ice area without changing water temperatures, I can imagine that introducing artifacts and it might be difficult to tell what weather was caused by the artifacts and what by the lack of sea ice in summer. So rather than doing that I was thinking it might be better to reduce the thicknesses of the ice to get the volume down to an amount that can melt out.

Some tweaking of the thickness amounts should allow some sort of reasonable melt out pattern to occur. It is then a case of running a model with atmosphere and top ocean layer (slab ocean?) allowed to vary but with those sea ice extents and thicknesses prescribed. This atmosphere model would have to be done many times over with slight variations in initial conditions to get climate risk information.

The impacts we are interested in, floods and droughts, come from atmospheric weather. Is this possible without introducing major artifacts?

This would probably only cover a few years after sea ice disappears for the first time. Then you would make the ice thinner again to melt out faster in a full GCM to provide new prescribed ice extent and thickness data for the atmosphere model to cover a later as yet unspecified time period merely relating to when the ice cover disappears at that speed.

Cannot get a huge amount of advance warning like this but I suggest some is better than none. It may not be reliable to go far out in time like this but can't the atmosphere model produce something like an indicator regarding flood and drought risk?

Such an atmosphere model would not be trying to get the sea ice right (it is just given something we think is reasonable to avoid artifacts) and we are only hoping for it to give reasonable indicator of atmospheric effects.

Steve Bloom

That seems reasonable, Chris, but doesn't it amount to model tuning? I mean, if you have a model that can't replicate observed conditions to begin with, how reliable will it be in showing future states after those conditions have been forced upon it?

But re the larger question of long-term Arctic amplification, I expect the modelers have tried everything they could think of. My own gut feeling is that they're missing, or are at least very short on, several crucial factors.

Re short-term atmospheric circulation effects in particular, while it's very early days for the science on Arctic-jet behavior connections, note that another major model failure regards the extent and pace of expansion of the tropics (IMO a fairly obvious critical factor in jet behavior).

It would be very interesting to get feedback from a modeler on these sorts of questions.

Jim Hunt

@Steve - Crandles is over there already, but have you by any chance seen the forum thread on the acronymic DASIM.

All of which somewhat begs the question - "Is there a modeller in the house"?

Steve Bloom

Aha, this article on new results re models and Arctic sea ice just popped up. The paper seems worth a close look.

Kevin O'Neill

Steve Bloom - Ellesmere was 18C warmer than today during the Pliocene, but the mean annual temperature was still -1.4C. Even in the Pliocene Ellesmere was hardly a tropical paradise.

Climate variability in the Early Pliocene Arctic: Annually resolved evidence fromstable isotope values of sub-fossil wood, Ellesmere Island, Canada

I'm not sure why you think GCMs "fail" to capture the Pliocene warming. I know that NASA/GISS has run simulations that came up with 10C warming in the arctic. Pliocene Global Warming

Considering all the uncertainties involved, I would hardly classify the discrepancy as a failure - and I'd bet there are other models that performed even better at capturing the Pliocoene warming.



Despite your humor (I appreciate it), I was not being tongue in cheek about the White House. The person is placed where he can get answers that we will not and ask questions abour what is happening that will get replies that we may not.

The important piece is the last comment.

"... historical records may provide a poor precedent for the future. Unusual events now could well be indicators of more to follow later in the season."

In other words - we are not seeing sea ice activity witnessed previously, and they may be indicators of further change in September.

Chris Reynolds


Thanks for that.


Not totally sure what definition Espen is using for fast ice. If, as I suspect, it's the MYI region off the Canadian Arctic Archipelago(CAA) then here's my estimate:

Triangle in Beaufort very roughly 700*700 / 2 is 245000km^2.

Triangle off CAA and Greenland very roughly 2300*700 / 2 is 805000km^2.

Add the two together and you get something around 1M km^2. Very very approximately!

If anyone has Google Earth pro they could work it out much better, but I'm not willing to pay out for that just to calculate this figure.

Chris Reynolds

Oh, and Espen.

You were right! :)

Espen Olsen

Chris and Wili,

My definition of Fast Ice is the Sea Ice stuck and not yet fractured or seperated to the north coast of Ellesmere and to some degree the area north of Ellef Ringnes Island ( Named after a sponsor Brewer to the Fridtjof Nansen expedition). It now has the shape of a triangle sized 600 km x 650 km x 150 km somewhere around 45000km2

Espen Olsen


A "bit" earlier than expected though! ;-}


>"if you have a model that can't replicate"
Yes there is certainly an issue that isn't completely solved by thinner ice rather than area changes.

>"long-term Arctic amplification"
I don't know whether to dismiss this as not hugely important at the timeframe I am thinking about. (Long time ago, uncertain forcings etc....)

I see Dr Roy Spencer on his blog saying tropical regions warm 3 times too fast. That would really be a problematic tuning problem if that has to be done to get the global temperature series about right the tropics warm too fast and the poles too slowly. Shouldn't really believe it is that bad a problem, should I?

Yes, I would also certainly like to see such feedback from a modeler. (Maybe one day I will figure out how to get modelers to come to the CPDN forum.)

Chris Reynolds

Thanks Espen,

It was quick. And I've missed it. I had an afternoon off and watched a couple of films yesterday - forgot to get my IR from Environment Canada.

Anyway, here's 3 to 11 March (excluding 10 March - Doh!) as animated gif.

If anyone uses these images use the file option to save as I will be tidying up my Google Drive in the coming months and some of these links will be dead. But when sorted I may do a page on my blog with links to this stuff.


I guess we're all aware that models can't predict rapid shifts in state driven by unanticipated factors. Black carbon for instance which has no historical model to extend from.

The important message for me is that, unlike methane, which is a blip which can be factored in, sea ice loss in the arctic becomes a self reinforcing heat generator in the second coldest place on the planet. Something which should be factored into models but won't be for some time.

I do have another question though and something which I have been watching for a few years now.

I see a lot about weather patterns, MYI loss, sea heating and other factors. But I never see anyone talking about the correlation between the known spikes of CO2, annually and decadal and also the solar cycles.

For instance the huge blip of 98 CO2 happened at a solar low. Yet, as far as I can see, the impact was decadal long.

Then there is the Solar low from 2008 to 2010. It "should" have produced ice growth and building of more MYI. Yet, in fact, the 2007 loss (again fairly low solar), seemed to drive a new state change which was not overcome by the solar minimum. Leading us to the conclusion that the very large growth in CO2 from 98 to 2012 (averaging more than 2ppm per year), has blanketed even the large loss in heat from a solar minimum.

So who is modelling what happens over the next 3 years. We are almost at solar maximum. Granted it's lower than the cycle 23 maximum by a long way. But we have high CO2 growth (>2.6ppm in the last calendar year) at the same time as a solar maximum.

Bring this together with a large volume loss, more exports of MYI through the Fram and a changing weather situation and where are we? Ice free in 2013? 2014? Or even 2015 when we finally exit the estimated Solar maximum?

To me that is part of the picture which is missing. Is it important? I don't know but it does seem to correlate.

Chris Reynolds


You'll find some graphics from Notz & Marotzke here:
There's also a link to the original paper there.

The Notz & Marotzke paper is mainly looking at self acceleration (SA). But they use area/extent, not volume. In terms of all three I think SA can be ruled out for most of the period of sea ice loss but in recent years SA has taken over. That's if we accept the effect of the summer dipole as related to sea ice loss, not just incidental weather.

However solar forcing doesn't seem to be having as much a role as I had thought years ago. It seems direct and indirect role of CO2 is the main driver.

It's worth noting that we've been in a Grand Solar Maximum during the latter half of the last century. Yet as you observe despite dropping out of it the Arctic changes are accelerating. So I really doubt if the predicted reduction of solar output is going to have much of an effect.

I don't think there's much of a correlation between annual spikes of CO2 end sea ice (detrended doesn't support it). But the long wave forcing of CO2 is small, and together with warmer air/ocean, the net effect is mainly on volume, which retains the impact of losses/gains from year to year. Then volume impacts area/extent because thinner ice is more able to reveal open water for a given melt.



Another cycle to factor in is ENSO/El Nino/La Nina. We had a very strong El Nino the year CO2 spiked 2.98 ppm in 98. We may have higher CO2 jumps in the future, when another El Nino arrives.


The outer bands of the Kamchatka SLP are beginning to cover the Russian Arctic, we will lose a large part of our view for a good part of the week, but I anticipate that what we will have afterwards will be more stirring than what we havee seen so far...


I did get a couple of answers back in Oct 12 from one of the CPDN climate scientists, Friederike Otto:


Al Rodger

The spikes in atmospheric CO2 concentrations mentioned above (off topic, mind) are ignorable in the grand scheme of things. However they have their climatical importance inflate by skeptical myth-makers.
The biggest climatic wobble is ENSO which a few months later results in a positive wobble in global average temperatures and following later again is a wobble in the rise in CO2. Skeptics love saying this "proves" temperature is the cause of rising CO2 and often graphs are produced to demonstrate their point. There is even the occasional learned paper published saying the same thing, (this one, Humlum et al (2012) even getting peer review, mainly by not putting the effect it investigates into its true climatical context).
And it is not put in context because these skeptical messages are all a load of nonsense.
This graphic demonstrates it quite well.


Thanks for the replies.

I guess we'll just have to wait and see how it goes and how the analysis works out.

I was more interested in the decade on decade changes in CO2 rather than the year on year. I'm aware of the whole denialist arguments on that.


Talking about ENSO, reminds of what happened last year:

“This is one of the most challenging outlooks we’ve produced in recent years because El Niño decided not to show up as expected,” said CPC deputy director Michael Halpert

I wonder if they have an explanation for this?

Steve Bloom

Neil, IMO the best way to think about that issue is that the warming we're experiencing now is due to the CO2 levels of ~40 years ago.

Steve Bloom

Crikey, Typepad just disappeared a comment I spent nearly an hour on. It showed as posted, then went poof when I refreshed. That is a bug, not a feature. Fortunately I have it in buffer, and will try it in two pieces.

Steve Bloom

Kevin, bear in mind that even with winter temps below freezing, Pliocene warm period-like summer temps would be like applying a blowtorch to all the ice in the Arctic.

Also, that GISS link is very old. You need to check the dates on these things, although note that even if still valid that 10C Arctic amplification figure is only half of actual, i.e. pretty damned severe.

But here's a new paper on model ensemble results for the period. Not only is the shortfall still pretty bad, due to the long lead time needed for model intercomparison projects (about 4 years in this case) the selected CO2 value of 405 ppm turns out to have been at the extreme high end of the likely envelope (central estimate is more like 350 ppm IIRC). IOW, both overall sensitivity and Arctic amplification are low-balled.

Chris asked why we should care much about this.

Basically it's because the mid-Piacenzian was the last time CO2 levels were anything close to present values and so is recognized by scientists as far and away the best analog for our near future. If we simply keep CO2 at present values and wait, we will get something worse.

But we are not keeping CO2 at present values. Instead of slowly adjusting to equilibrium levels, climate today is adjusting on the fly to an unnaturally fast transient. Relative to paleoclimate, expect the unexpected.

That the model failure is freighted with significance is a widespread if not consensus scientific view. The discoverer of the Arctic camel fossils was interviewed on NPR a few days ago, saying in part:

I think the times that the camel was living is also of interest because at that time, it was the mid-Pliocene warm period, so the dating shows that this animal lived at a time of global warming. And this is really important because we're interested in global warming right now, evidently. And what the fossil evidence is showing is that the high arctic was actually much warmer than what the climate models are predicting.



Steve Bloom
So based on previous work, what we were able to show is that in the high arctic, the climate was 14 to 22 degrees Celsius warmer than it is today. And this is really significant because at that time, the global temperature is estimated to have been two to three degrees warmer than today. And, of course, that two to three degrees is of great interest to us because looking forward for the next hundred years, it's been suggested that that might be where we are headed.

So we really should get to know how high latitudes respond to global warming. And so the data that we're getting from the camel site and other sites are really important for ground-truthing our climate models. If we can hind caste properly, we've got a shot at forecasting.

(Emphasis added.)

But while the models being able to correctly handle the equilibrium sensitivity of the mid-Piacenzian would be a huge step forward, the big question we actually need answered is how fast Arctic amplification can proceed under the present fast CO2 transient. Looking at the poor showing of the models re the Arctic changes of the last decade, I hesitate to rely on results implying a relatively benign near future. IMO there's a rapid change mechanism implied by the amplification, and there's a good chance it's already been triggered.

BTW, note that the equilibrium climate of the mid-Piacenzian would not have included effects of permafrost or clathrate loss. They've had the entire Pleistocene to build up to levels that couldn't have existed in any past warm period.

I'll end by pointing out how ironic it is that the new Holocene hockey stick paper has gotten so much more attention than the camel paper even though it's the latter that has the serious implications for us.

(Note: There's a lot of terminological shifting back and forth between mid-Pliocene and mid-Piacenzian. This is because a few years ago the stratigraphers decided to remove the third period of the Pliocene and add it to the Pleistocene, making the mid-Pliocene no longer the same thing as the middle of the now-last period of the Pliocene, the Piacenzian. There's also the problem that many people know what the Pliocene is or at least have heard the term, whereas that's not the case for the Piacenzian.)

Steve Bloom

Hmm, should have previewed, The first two paragraphs of the second piece continue the quoted passage, although that's probably obvious enough.

Artful Dodger

Hi Steve,

When that happens to you, Typepad has eaten your post as spam.

Neven asks us to send him an email when it happens, and he'll fetch the comment out of his Admin spam folder.


Artful Dodger

A-Team asked | March 11, 2013 at 11:54

"Anybody know the url for this ftp site? I have been looking at L3 data obtained by AMSR2 onboard GCOM-W1(SHIZUKU) but not seeing quite what you are referencing."

Hi A-Team,

I have previously decided not to publish a link to L2 data on this blog since it can be 10s of GB daily for all sources and would be abused by certain parties without the intention of doing actual science.

Neven has the link in his email however since this issue came up as part of an extended kvetching session necessary to prove that CT data lags IJIS by 2-3 days. :^)

So, send an email to Neven asking him to forward my msg, or join us over at the ASI Forum and send me a PM.

BTW, i don't think JAXA's spin-up plan for AMSR-2 includes release of L3 data before 2014... it's on their website if you care to look into it.

Cheers, and KUTGW!


you probably know already that Tamino has started to double-check some analysis performed here by regulars. First part:


Bosbas, there's an ENSO thread on the Arctic Sea Ice Forum (ignore the warning and make an exception, as https is safer than http).

Steve Bloom

Thanks, Lodger. Although having been burned enough over the years by this sort of thing I habitually put comments into buffer before posting.

Neven, would you either replace those with the intact version or drop a block tag onto the start of the second one? Compulsive, I know. Thanks.

[Done, N.]

Al Rodger

Hi NeilT.

Decadal-scale CO2 changes show in the smoothed red line on this graph.


Below is a conceptual implementation of future ice pack movement in the Beaufort, loosely based on the vector field supplied in an earlier forum post by Chris

The first frame shows my initial choice of deformation nodes and deformation directions (using Tools -> Transform Tools -> Cage Transform in Gimp). Overall this works better than rotating a circular or elliptical region about a fixed point.

The underlying image is Jaxa color radar of 11 March 2013 enhanced to show ice structural classes. The goat's head is outlined for clarity but is not involved in the over-simplified depicted motions

Each successive deformation of the perimeter stretches the colors of the interior more or less continuously given this is a discrete 255x3 RGB color space. It may or may not retain ice thickness and age. (Grayscale might work better.)

Better to have done the whole Arctic, constraining to the land mask (instead of tearing off pieces of the CAA) -- didn't have the data for that yet.



Excellent stuff!

BTW, the HYCOM/CICE output is available again. Just restarted this morning.


The ice speed shows significant potential for churning the ice in the Beaufort over the next seven days.


Nice spotting, A4R. The Navy series is one of the best products out there. Here I took their speed and drift for 11 Mar 13, dimmed the colors way down, pulled out the Arctic portion, rescaled, and overlaid the arrows with partial transparency on color radar of the same date. Their gif imagery is technically difficult to work with because each frame only contains difference information with respect to previous frames and the baseline image.

Doing this on all frames would result in an overlay animation. Since they go forward a week in prediction, the idea would be to use the Gimp cage transform described above -- driven by their arrows in some quantitative fashion -- to pull along the underlying radar image to forward ice sheet structural deformations.

 photo overEasy_zps8de206ea.png



I'd save how you did this image, this will be valuable to see in the coming months!

From appearances here, the ice grinder is still on high speed. What is of interest is that the Bering Strait flow is reversed, but the Fram flow is high.

This is very helpful!


PIOMAS update:
Latest value: 2013-3-1 19.945

I have updated my graphics at ArctischePinguin for the latest data.

Note that the graphics have been moved to a subdirectory. If anyone is using permanent links you will need to update these.

Monthly data
Daily Anomalies
Daily data
Daily data with a "prediction" based on exponential trend


Hi Wipneus,

On the final graph, you have an uncertainty band indicated for the low point. Perhaps it might be an improvement to include an uncertainty band also for the last actual (modelled) reading - currently at 28 Feb.

Dave C

Wow. That is a massive volume gain for February. We are on pace for a record ice volume recovery.
It looks like the wind and cracks caused significant ice gain.

As of now my prediction is that this is going to be a recovery year.

The expected value is probably slightly below, but there is a decent chance that we won't even break last year's volume record.

Kevin O'Neill

Re: Dave C

Neven, I think you have your first Poe .... see
Poe's Law

Dave C

I am far from a global warming denier. You can look at my other posts.

I believe in following the facts though. Nobody on this forum predicted we would have record volume gain this winter. This is an unexpected fact.

While that fact doesn't fit in with the preferred narrative, you can't just ignore it. Ice has memory, and it is likely that the volume minimum for 2013 will be higher than most people on this forum would have predicted.


Well, there it is… PIOMAS February…back to volume as it was at the end of freeze seasons ’11 and ’12. As I posted yesterday :
“I expect that PIOMAS will show further diminishing of the difference to last year. The cracks should have delivered anything between 50-150 km3 of FYI, while a lot of the earlier formed stuff has been pushed into the Atlantic sector. If PIOMAS does’nt, we’re in for trouble.”

So we’re not in for trouble? Not straight away. Not immediately and alarmingly 'in' below the trend.

From now on, the character of the new season will be decisive. If it follows the trend, the North Pole could very well be ice free in August. After all, it is clear the MYI boundary is now about 100 km away in the direction of Greenland.

Maybe the slow path of the ‘Goats’ head’ will take it right over the Pole? There’s 250 km to go. The pace…about 3 km a day.
In comparison, winter temps gave a good show through Feb and March over the peripheral Seas. That won’t help come Summer.

If summer gets anything near '07, the game will get to the stage 'down and counting'.

Shared Humanity

Dave C.

I do not have enough knowledge to weigh in on this prediction.

Time will tell.


"That is a massive volume gain for February. We are on pace for a record ice volume recovery. It looks like the wind and cracks caused significant ice gain.

As of now my prediction is that this is going to be a recovery year."

Will see in September mr Dave C , but I didn't care about the maxima, because surface temperatures were and are still well below 0 C in winter. But the multiple leads and abnormal sea ice shapes contradict any sign of recovery, but to the opposite, sea ice is complex, on appearance its not black and white but really multiple shades of polarized colours. When you see most of Arctic Ocean surface as first year ice, you must ponder more before claiming its a recovery. Rather its recovered when the multiple year ice spans from Canada to Russia just as it was not so long ago.

Kevin O'Neill


2013 is the lowest volume for day 60 in the PIOMAS dataset. It's really difficult to characterize this as a 'recovery.'

Year day Vol
1979 60 31.249
1980 60 30.332
1981 60 28.952
1982 60 26.61
1983 60 28.571
1984 60 28.323
1985 60 28.455
1986 60 28.734
1987 60 29.876
1988 60 29.441
1989 60 28.852
1990 60 28.797
1991 60 28.784
1992 60 27.962
1993 60 28.727
1994 60 27.879
1995 60 27.27
1996 60 25.495
1997 60 27.448
1998 60 27.494
1999 60 26.413
2000 60 25.463
2001 60 25.46
2002 60 25.929
2003 60 25.408
2004 60 23.9
2005 60 23.877
2006 60 23.171
2007 60 22.132
2008 60 22.84
2009 60 22.949
2010 60 21.417
2011 60 19.981
2012 60 19.812
2013 60 19.945


Dave C, your prediction seems to be based on the 'memory'of the sea ice. If I have interpreted posts by FI Chris Reynolds well, that goes mainly for MYI.
Going on, it seems to me that about 0,5 Mkm2, or 450 km3 of mainly 2-year ice in the Sep left-over could have 'matured' this winter.

I'm afraid IceBridge will show that the all-important Basin pack hasn't profited much from the late winter cold period.

In that case, most volume gains stem from the peripheral Seas. It is nice to see the Kara Sea showing almost 1,5m on the CICE map, but it will melt with great certainty.

So a trend-break on volume isn't to be expected...


With the cracks still propagating northwards and the ice in it's very fractured state, checkout Chris R.s post at 21:38 above, I still think a record low is likely.
It's only one feature, but I compared this image
and this image
one, just looking at the western aspect of Banks island to see how early the thinning was, but 70 odd days ahead?

Espen Olsen

The midterm future will be good for both parties, since there will be record lows and record gains.
Anyone with an "pro" opinion on the situation in the Arctic Ocean, will still be able to make a dime, so don't worry!

The comments to this entry are closed.