« ASI 2013 update 8: the end is nigh | Main | IPCC crisis meeting »


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


OK…PIOMAS 31082013 5068 km3… my own graph take on CAD.
So it’s the high estimate. The splinter zone holds this 2000 km3 of ice above the 3000+ km3 left in the CAB pack.
That’s about it for the numbers. Prepare for the denial storm.

Andrew Xnn

Looking at the average thickness graph, I can see why it may be more than just a few year to an ice free arctic.

It has taken close to 30 years to lose 1 meter and there is still just over 1 meter to go.


Shouldn't the title be "PIOMAS August 2013"?


D, I thought about that last year and decided to do it this way. The updates usually arrive in the first week of the month (sometimes the second during winter). It sounds as if the title loses actuality if I put in 'August'. It's September now, so that's what goes in the title.

NSIDC does the same with their monthly analysis. For instance their latest one is called September 2013.

Lars Boelen

Neven, I completly agree with your view : I allways wonder about "the shop of the year 2012" award sticker on the door of a shop. It makes you think, "so you were great last year, but I'm interested in THOS years result".

Using the current month shows the readers that this is the latest and greatest.

On the technical part : very interesting that with all this "cold and cloud" the average thickness managed to set a fresh low record (for the month), must mean that botttom melt is a giant killer.


Lars, to me, same thickness as 2010-2012, means the ice did not recover to what was before 2010, after all

Colorado Bob

In terms of the winter climatological period — the months of June, July and August — it was the warmest three-month stretch at the South Pole since records began 56 years ago, according to Phillip Marzette, senior meteorologist at the South Pole Station.

It was also a windy month, with seven days that either broke or tied the previous peak wind speed record for those days.

"The weather has been really bad this year,” said Dana Hrubes, a scientist currently working on the South Pole Telescope who has wintered seven times at the South Pole Station since 2000. “I’ve never seen anything like it.”
Read more at http://scienceblog.com/66290/south-pole-experiences-more-record-heat-in-august-to-end-warmest-winter-ever/#FHHuEdKAxduc27A2.99

Rob Dekker

Thanks Neven,
Wipneus' Monthly Average Arctic Ice Volume graph showns very nicely how ice volume in 2008 also rebounded in June, July and August, after the devastating 2007 year. Almost by the same amount, I might say.
After 2008 of course, the downslide continued...

What I find interesting is which physical processes cause this 'rebound' that we see now after record setting years like 2007 and 2012. Does a next year 'know' about what happened the year before, and if so where is the 'memory' kept ?
Did the atmosphere "over-cool" itself last year, and now needs to 'refill' the battery ? Or the ocean ? Or both ?


Just saying; If PIOMAS continues as it did in 2007, it will almost catch up with 2012 by November. And if it sticks to the 2009 trend, it will be far below 2011 by the end of the year and far below both 2012 and 2013 by March.



"Those error bars, large as they are, didn't manage to catch this year's anomalousness. It's clear that statistics has its limits, but this is pretty amazing."
As I understood Wipneus those error bars are normal standard deviations. Therefore, about 68% of data should be inside the bars and 32% outside on average. So I would still stay with Wipneus, that the error bars are quite large, since less data than 32% is still outside the bars.


But if that really would be the confidence limit (typically 95% data inside, equal to 3 sigma or standard deviations), that would mean that something could be wrong with the estimation function or something perhaps systematically changed in nature... Only future can tell.

Kevin McKinney

Thanks for another solid update, Neven! Despite this year's uptick--and I think "Weather Strikes Back" is an excellent sound bite for what happened--the death spiral seems to me unbroken. Though if we got a couple more years of weather like this, the modelers who've been saying "ice-free in 2030" would look a lot more prescient than was the case last year.

From a political perspective, it's too bad, though. I was hoping that there would be some sort of 'kick' heading into the UNFCCC negotiations in Warsaw this winter. That process may not be in a death spiral, exactly, but it doesn't look too healthy, either.


Rob, those are the same questions I've been wondering about.

The answer that makes the most sense to me is that, the relatively late refreeze last year allowed more heat from deeper down to escape from the ocean during the beginning of the dark winter. Even though at the end of the re-freeze season, we had pretty respectable coverage, for most of the first half of the winter, we were at historical lows for area.

But that's just what makes most sense to me about where there could be such 'memory' in the system. I haven't investigated relative heat at various depths over the last 12 months. If anyone has good data on that, it might be interesting to look at.


As I understood Wipneus those error bars are normal standard deviations.

No, it is the 95pct confidence range, about 2 sigma.

Jai Mitchell

It shows how an averaged 1.5C of cooler temperatures during the days 160 and 200 of the year will greatly reduce the amount of ice loss during a melt season.


compare 2013 with 2012 and 2007

2008 had a slightly cooler start than 2007 but nothing close to the dramatic change for 2013.

2013 was the year that the jet stream failed and cut-off low pressure systems swirled backwards across the midlatitudes throughout the summer. The result was increased vapor transport into the arctic leading to much cooler temperatures throughout the season. it is unclear why the atmosphere acted this way but Drs Francis and Trenberth are battling out the idea that it may do with decreased snowcover on land masses in the late spring, affecting the strength of the polar Jet.

I think it may have to do with increased vapor loading of the atmosphere in the midlatitudes, possibly due to a combination of a Negative PDO (increased north hemisphere midlatitude surface temperature and evaporation) and possibly Chinese scrubber activity which moves large volumes of water vapor into the atmosphere. (about 500 gallons of water per megawatt hour)

As I understood Wipneus those error bars are normal standard deviations. Therefore, about 68% of data should be inside the bars and 32% outside on average. So I would still stay with Wipneus, that the error bars are quite large, since less data than 32% is still outside the bars.

Ah, OK. Thanks for the extra info. I just try to make it sound nice! :-)


The Daily Mail is now trying to introduce a 60% over 2012 meme. Do you all have the real numbers? People in the media need them right away. Thanks!


Jim Hunt

Tenney - David Rose is renowned for reiterating nonsense. How do you suggest correcting his inaccuracies, particularly when he provides no justification for his rash assertions apart from a couple of cherry picked images?

I endeavoured to do so on a previous occasion and my online comments never made it through moderation. I've been following events in the Northwest Passage this year and the bit about "more than 20 yachts that had planned to sail it have been left ice-bound" is yet more stuff and nonsense.

Will you explain to him all about the subtle differences between extent, area and volume, or shall I?

Jim Hunt

Can anyone see any sign of my first attempt yet?

Jim Hunt

How about my second attempt?

Jim Hunt

I've dug out the latest daily numbers. Do these sound right?

PIOMAS Day 243 3.480 => 5.077
CT 0.6794 2.35009 => 3.62455
NSIDC Day 249 3.558 => 5.236
IARC-JAXA Sep 7th 3312446 => 4893380

For some strange reason they all seem to be under 60%

Jim Hunt

My apologies for flogging a dead horse in public. The Mail Online has now seen fit to publish my 2nd and 4th comments, but not my 1st and 3rd:



Min 2012 CT area was 2.234. We are about 62% above that now.

My calculation of sea ice area from other ice concentration gives similar numbers.

So IMO a 60% number is quite correct. It is of course wrong to leave everything else out.


Rob and Wili,

There was some lively discussion of the "rebound" topic last year at http://neven1.typepad.com/blog/2012/05/arctic-sea-ice-loss-and-the-role-of-agw.html?cid=6a0133f03a1e37970b01630551734e970d#comment-6a0133f03a1e37970b01630551734e970d, referring to a paper by Tietsche et al that found recovery over a two-year interval following a shock. See also Chris Reynolds' discussion at Dosbat http://dosbat.blogspot.co.uk/2012/09/sea-ice-albedo-changes-and-seasonal.html of the Notz and Marotzke paper on AGW vs. other factors.

The evidence of 2013 is for two different causes: a rebound early in the year, probably owing to heat loss from open water at the end of the previous melt season, and later a slow melt from cloudy and cool conditions. Whether the latter is natural variability or a longer AGW-related causal chain, as Jai Mitchell suggests, is a good puzzle for the scientists.

[fixed links; N.]

Jim Hunt

Thanks Wipneus. The Mail article explicitly states that "A chilly Arctic summer has left nearly a million more square miles of ocean covered with ice than at the same time last year – an increase of 60 per cent"

Hence my own calculations weren't based on the 2012 minimum. The "million more square miles" is certainly not even in the right ball park.

A helpful skeptic has confirmed that there are certainly not "20 yachts... left ice-bound" in the Northwest Passage.

Of course, and please forgive my use of the Anglo-Saxon vernacular, when it comes down to volume 160% of sweet Fanny Adams is still sweet FA.


Jim, ha they fall easily to their beliefs of the looming coming ice age:

"A chilly Arctic summer has left nearly a million more square miles of ocean covered with ice than at the same time last year – an increase of 60 per cent""

The key word used here is chilly, was it ??? Look at my response


Suddenly their ignorance shines when they have the chance to see the same Arctic data as we do.


Thanks, Wayne,
I find the anomaly map even more telling:
 photo Geo500Mbanomalyjuntoaug2013small_zps0a014f11.jpg


There was some lively discussion of the "rebound" topic last year

Thanks a lot for those links, iceman! I had completely forgotten about that discussion, as I'm wont to do with 98% of things.

Rob Dekker

iceman, thanks for constructive feedback.

As Tietsche et al shows, Arctic sea ice is following a mean determined by forcing, with natural variability superimposed. I don't think anyone could possibly disagree with that.
However, the question that is more interesting in this is : where is the mean (in 2013) ?

The model runs (using AR4 IPCC models) reported by Tietsche et al suggest that the mean is around 5.2 million km^2 with a standard deviation of about 300 k km^2.
That would suggest that 2013 is right around the mean. However, in these same model runs, the 2012 extent minimum of 3.6 is more than 3 standard deviations below the mean, and thus, in practical terms "should not have happened".

In fact, when looking at Tietsche et al's figure 1 (similar to Notz et al figure 3), the results of 2012 (3.6) and subsequent year rebound (around 5.1) is more indicative of the period of "high variability" that comes with "weather" sensitivity of ultra-thin ice, which is (in the AR4 models) projected to be realized around 2025, when the "mean" is around 4.0 million km^2, and is starting to drop exceptionally fast after that.


So it almost seems as if the AR4 model runs are right on track in predicting the developments in the Arctic, but is running a decade behind reality, and the mean is about to fall off a cliff in the next few years...

Can this be real ?


Here is the today's ice cover compared with last year, seen by AMSR2 and processed by Jaxa. Blue is concentration now above 15% and below last year. Red is the reverse.

I am placing such images with some regularity on the forum. It is more convenient for me, and Typepad seems to particularly dislike me which is not encouraging.

Rob Dekker


I have not seen any paper reporting on heat content in the Arctic ocean over the past couple of years, but I agree with you that that would be very interesting data to investigate.

We all know that the deep Arctic contains a tremendous amount of heat, which may serve as a 'battery'. I have gone on record here at Neven's to assert that the GAC 2012 may have disrupted the upper and lower halocline in the Beaufort and CAB for a week or two, and thus maybe a lot of heat escaped from that 'battery'. If more heat escaped from the Arctic depths than what would typically be added during a year, then the Arctic may be cooler now than last year, which would explain at least a part of the reduced melting, and also identify the 'memory' that may cause rebounds...

Thanks for a clear line of thought.


iceman, let me join neven and rob in thanking you for those links.

rob, I hadn't considered the roll of the GAC 2012 in cooling the ocean. I know that in the tropics hurricanes draw their strength from warm sea surfaces, so they can have a major cooling effect. I just wasn't sure if Arctic cyclones had similar effects since they are somewhat different beasts.

What does all of this portend for next year's melt season?


why is anyone even discussing the fiction published in the Daily Fail?
there is more truth in the script of a Batman comic.
everything the Daily Mail publishes on scientific subjects is poorly written by obvious amateurs who have no clue what the word science even means. if they cannot explain the difference between a wasp, honey bee and bumble bee or publish correct pictures of these creatures how on earth are they going to write anything other than fiction on a more serious subject?

Jim Hunt

Philip - Well I'm discussing it because Tenney asked a question about it! My latest analysis of the "fiction written by obvious amateurs":


Given the number of things that David Rose has evidently managed to get wrong in just the first few lines of his article whilst discussing Arctic sea ice, how many more do you suppose he got wrong when he went on to consider "global cooling"?


I want to thank Jim and Wipneus for their comments.

I asked because this meme will be spread far and wide by the players of the Climate Denial Machine -- it is cognitively very easy for an unsuspecting public to absorb. I expect it to come out on Faux News soon.

Jai Mitchell


you are incorrect when saying that hurricanes produce a large cooling effect in the oceans. The amount of heat transferred to the atmosphere is miniscule when compared to quantity of heat in the ocean's surface layer in the tropics. The cooling that you see is actually due to mixing from deeper, cooler waters.


my rememberance of the Tietsche paper is that it theorizes a recovery mechanism of ice that occurs due to an increase in longwave radiative heat transport during the arctic winter. Less ice means less insulation. So more heat energy lost, (actually more than the amount of heat that was gained each summer by increased albedo!)

I think that this is a good point but it has a BIG variability that is due to increased cloudiness from the amount of increased evaporation. Also, the ice recovers much more quickly.

Wadhams and other's work indicates that the previous IPCC models of arctic ice loss to follow forcing (bringing an ice-free arctic in 2080) were completely wrong. Wadhams fameously said (?) that the ice may be gone in 2013. But the reality is that the amount of warmer waters in the arctic is not a linear as he predicted.

Still, the rate of ice loss is much much much more rapid than the rate of forcing increase. It is primary due to currents according to wadhams.

Jai Mitchell

*increased albedo-driven absorption*


If ice loss can increase planetary heat loss through long wave radiation from open water so that it exceeds heat gain through insolation, wouldn't this put an upper bound on global warming?

Allen W. McDonnell

Mdoliner43, The reduction in the albedo of the Earth from the loss of the ice is of greater magnitude by quite a large extent than the long wave radiation escaping into the atmosphere.

There is an upper bound on global warming from water vapor condensation effects, when water evaporates from the surface it takes a lot of energy with it, when the water vapor condenses into droplets at high altitude it releases that thermal energy as long wave radiation. Because the condensation takes place at relatively high altitude the thermal energy released has an easier time escaping because a large portion of the greenhouse gasses that reflect long wave radiation are below the condensation layer.

Because of this energy convection effect so long as Earth has surface water to evaporate and condense at high altitude there is a limit to how warm the surface can get.

Rob Dekker

Jai, there are many feedbacks in the Arctic, of which the summer albedo-driven absorption (a positive feedback) and the fall "ice insulation" feedback you mention (a negative feedback) are two important 'fast' ones (feedbacks that act with one year).

All feedbacks added up should give us a good idea of the overall "mean" response of Arctic sea ice to some external 'forcing' (such as AGW), and the overall "sensitivity" to short term perturbations due to internal variability such as cloud cover, sea level pressure variation in the various seasons, storms etc. Neither Dr. Tietsche nor Dr. Wadham appears to disagree with that.

However, since we cannot measure these feedbacks directly, their 'strength' has to be estimated in models, and since Arctic sea ice is most likely one of the most difficult things to model, this is where there is considerable uncertainty.

So, as far as I can see, Dr. Wadhams' work did not "indicates that the previous IPCC models of arctic ice loss to follow forcing .. were completely wrong". In fact, Dr. Wadhams clearly recognizes that Arctic sea ice loss is caused by external forcing.

Instead, Dr. Wadham is concerned that models (especially CMIP3 and CMIP4 to a lesser extent) appear to seriously underestimate positive feedbacks in reality, especially ice volume development, and I happen to agree with him.

And that was my point : it looks to me that CMIP4 models in Tietsche et al are underestimating some positive feedbacks in the Arctic climate system, and as a result the actual ice extent in 2012 and 2013 is more indicative of the CMIP4 model runs for around 2025, and thus the models are running about a decade behind schedule :


And if CMIP4 models are running about a decade behind schedule, then the "mean" in ice extent is about to take a serious nose-dive in the next couple of years...

Rob Dekker

Jai said :

Wili, you are incorrect when saying that hurricanes produce a large cooling effect in the oceans.

Actually I was the one who suggested the cooling effect of GAC 2012, not Wili.


Rob D,

"...high variability that comes with weather sensitivity of ultra-thin ice..." sounds like a good capsule description of the 2013 melt season - and as you imply, yet another instance where the models may need to play catch-up. We're seeing another manifestation just recently: as Neven observed in the OP, "...a large part of that record amount of first-year ice at the start of the melting season has been preserved..." So there is more remaining to melt towards the end of the season (from more conducive weather, bottom melt, wave action etc.) and the declines in area and extent are relatively steep for this time of year. I expect that PIOMAS will also have a drop in anomaly over the next month or so.

John Christensen

@Rob D,

I completely agree with your comments both on GAC 2012 and also the increased variability.


I don't see the relatively steep declines in area and extent for this time of year, not on CT or ROOS at least. What are the numbers you are looking at?

John Christensen

CT SIA grew by 4k km^2 yesterday.

Jim Hunt

I don't know which numbers iceman is looking at John, but over on the forum Wipneus reports:

Large declines for the time of year. The CAB, CAA and Greenland Sea substantially. All other regions contribute.


Allen W. McDonnell | September 09, 2013 at 18:36 - "Because of this energy convection effect so long as Earth has surface water to evaporate and condense at high altitude there is a limit to how warm the surface can get."

1. What is the predicted limit?

2. Would you share some links - references.

Allen W. McDonnell

JackTaylor, here are a couple I grabbed this morning, if you Google it yourself you have to sort out the sources that present scientific data from the headline writers who often use the most extreme terminology to get more clicks on their websites. Even National Geographic went overboard earlier this year by using the most simplistic model and then reporting the results as if they were significant. During the PETM, the Paleocene-Eocene Thermal Maximum 56 million years ago and all earlier world greenhouse events the temperature of the planet has always hit a hard limit when the climate was in the hothouse. Carbon Dioxide consistently shows a temperature increase of between 2C and 4C for each doubling of content. Taken to the greatest extreme you can go from 280 ppm (pre-industrial) to 560 ppm and get 4C rise, then from 560 ppm to 1,120 to get another 4C rise, and if you burn everything with fossil carbon we can get our hands on you might just barely double it one more time to 2,240 ppm giving you a total of 12C maximum rise. Of course the three doublings might also only give you 6C rise on the low end. Most climate scientists will give you the 6C-12C range for burning just about everything burnable.

    Recent observational studies show that these effects almost balance, but that the cooling effect is somewhat more important. From the point of view of global change, however, it is crucial to note that this small difference is about five times larger than the radiative effect anticipated from a doubling of atmospheric carbon dioxide (CO2), and that the individual components of the difference are orders of magnitude larger. In existing climate models about one third of the predicted warming due to increasing CO2 arises because of the predicted cloud changes. These predictions, however, are highly speculative because none of the models include interactive cloud physics.
from http://cgcs.mit.edu/research/focus-areas/convection-atmospheric-water-vapor-and-cloud-formation/


    The previous extreme global warm-up happened 56 million years ago, when Pangaea was splitting into separate continents. It is suspected that huge amounts of carbon were released into the atmosphere and oceans in the form of carbon dioxide and methane. The globe warmed 5 to 9°C (9 to 16°F). Most ecosystems were able to adapt—tropical mammals migrated to North America and Europe, and sea life swam poleward to cool down. But the rate of warming during the PETM pales in comparison to what we're now experiencing. Today, global temperature could be warming at a rate that is too fast for ecosystems to adapt.
from http://www.wunderground.com/climate/PETM.asp?MR=1
Jim Hunt

The BBC reports that:

The volume of sea ice in the Arctic hit a new low this past winter, according to observations from the European Space Agency's (Esa) Cryosat mission.

During March/April - the time of year when marine floes are at their thickest - the radar spacecraft recorded just under 15,000 cu km of ice.

In its three years of full operations, Cryosat has witnessed a continuing shrinkage of winter ice volume.

It underlines, say scientists, the long-term decline of the floes.




[Sorry, no Arctic/AGW Gish Gallops here. Take it to the other old, white males over at WUWT; N.]


Help please. Paul Douglas would like to know what percentage of the summer volume has disappeared.

So, there is the satellite era and the pre-satellite era. I assume that pre-satellite there were estimates made by people on submarines.

Thanks much!

John Christensen


We don't have the September PIOMAS data yet, which will include the minimum volume for the summer of 2013.

However, to get an idea of the volume loss, please see below the last 13 days of August in 1979 compared to 2013:

1979 200 22.79
1979 201 22.587
1979 202 22.364
1979 203 22.115
1979 204 21.913
1979 205 21.714
1979 206 21.503
1979 207 21.281
1979 208 21.066
1979 209 20.872
1979 210 20.702
1979 211 20.528
1979 212 20.347

2013 200 8.87
2013 201 8.698
2013 202 8.523
2013 203 8.335
2013 204 8.111
2013 205 7.87
2013 206 7.685
2013 207 7.568
2013 208 7.458
2013 209 7.355
2013 210 7.27
2013 211 7.191
2013 212 7.104

As you see roughly 66% of Arctic sea ice volume has been lost between end-August 1979 and end-August 2013.

John Christensen

Sorry, Skype massacrated the 2013 numbers. I hope you can still make them out.

Jim Hunt

Tenney - How technical do you want to get? There's a long "PIOMAS v CryoSat 2" discussion over on the forum.

CryoSat 2 doesn't work very well in summer however!

If non technical is what you're after see one of Andy Lee Robinson's PIOMAS volume videos, for example the one at the bottom of:



Thanks Jim and John,

Paul often gets questions by email from his viewers, and this one was about the percentage of sea ice volume decline in the summer.

So, the end of August data is just fine for this.

In fact, I really appreciate your help. I will send this info to him.

Thanks again!

The comments to this entry are closed.