« IPCC crisis meeting | Main | Freezing season 2013/2014 open thread 1 »


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

John Christensen

Yes, thanks to Neven for great covering of this unusual melting season!

@Watkin M,

Great observation of the ice breaking away. I am not sure this is very unusual for this time of year, and would also expect with freezing taking over that this will become solidly attached again by October/November.

Why this is happening, I would think the high centred sligthly towards the Siberian side (http://ocean.dmi.dk/arctic/weather/arcticweather.uk.php) has an impact, as this would cause northernly winds off the coast of Greenland and the CAA.

If you take a look at the animation feature below and select 'Surface Current', 'Arctic Sea' and 'play' you will see a small band of current along the north side of Greenland and the CAA, and it actually builds into increased currents in the Pole area and towards Laptev and ESS. So expect a lot of movement in the coming days, and that the open water near the Pole will stay a bit longer:


Finally, I also spotted on the weather link above for the first time this fall a broader area of -10 - - 15C is spreading between the CAB and ESS, clearly showing how a late season high is lowering temperatures with clear skies and decreased sun radiation.


Jdallen_was sorry for name typo.
Ot. Something that I have been trying to figure out for many years and maybe you can help.
Earth is like a sphere floating inside a round balloon/greenhouse. I have trying to figure out the BTU Heat energy being released everyday into the atmosphere by mankind. That would be the btu heat over and above the natural btu heat energy released everyday by forest fires, volcano's ect..

Jim Hunt

NJSnowFan - Not trolling? Off topic? Take your question to the Antarctica section of the Arctic Sea Ice Forum then.


Hi all, Chris, Wayne,

I follow the discussion closely, dismissing the parts that hold no content.
The state of the ice is clearly different from what it used to be. I’ve taken in the numbers (A/E/V) and accept them for what they are.
When I find the time (all winter while hibernating), I will try to get some work done that is within my special interest. I’m not going to speculate before there is some result worth mentioning.

I’m also keen to start comparing all the decadal-mean atmospheric data I assembled last fall to what comes up this year.
The enormous open ocean area on the Atlantic side makes me curious. The ice boundary is still retreating over the Nansen Basin. And the persistent ice SW of Severnaya Zemlya, in the NE Kara Sea is in this late stage of the season reduced to ‘insignificant’.

We’ll see what it fits into…

Chris Reynolds

Werther, Wayne,

I've still got a lot of work to do on the ice/atmosphere issue over the coming months, so perhaps discussion at this stage is premature. I need to get to grips with the evidence first.

Right now I'm writing an assessment of a method for prediction using CT Area, I can pull the prediction date back to 23 June as a sweet spot, arguably one could go earlier, but no earlier than 18 June due to the CT Area cliff. I'll be blogging on it later today.

But for now here is the result of my 6 July prediction for CT Area at minimum.

The red bars in September are the higher lower and central bounds of the prediction.


Jdallen_wa The all time extent record is right for Antarctica.

Neven had good post on sea ice area verse extent here.



NJSnowFan, because I'm in an exceptionally good mood, I'm going to give you one last chance. But one more link to sites containing too much dis- and misinformation, and you're out.

Put yourself in my position: I want to keep discussions on-topic and troll-free as much as possible. If you keep bringing up debunked fake skeptic talking points, it is costing me too much energy to handle that. And though everyone here is doing an excellent job at not feeding, I can't ask of them to keep doing that indefinitely.

The reason I started this blog, is that I'm worried about potential consequences of Arctic sea ice blog. Most people here are. I don't have much patience for people who maintain there aren't any potential consequences whatsoever. That the risk is zero.

So there you have it. If you don't adapt to this alarmist echo chamber, there's always the wonderfully scientific WUWT.

Martin Gisser

NJSF: trying to figure out the BTU Heat energy being released everyday into the atmosphere by mankind

Forget it. It is negligible: The added GHG make for 20x more than all energy use by mankind, or: 400000 Hiroshima bombs per day.
(Watch http://www.ted.com/talks/james_hansen_why_i_must_speak_out_about_climate_change.html 7:50)

Jai Mitchell

Antarctica. . .(yawn)

I just found out that the mass equivalent ratio of carbon dioxide emitted from coal after combustion is 1.87X the original mass of the coal being consumed.


. . .In other words, the amount of CO2 produced by burning coal is almost two times the amount of coal being burned, by weight.



Neven, The question I asked was for you about Antarctica Extent and yes off topic. Snow balled some in a way I did not want to. I did learn everything about Extents and Area when it comes to sea ice today though from reading one of your blog post and some other info on some sites.

I do feel their are consequences if the Arctic becomes Ice free in time. There are so many things(puzzle prices) that can make sea ice melt faster or even preserve(summer of 2013)/grow it faster. Some are Natural and some are man made I feel.
I have not seen this data link on the 20013 Arctic Extent minimum posted on this blog so here it is. I came across it while looking into Antarctica extent data.
Also shows sept 12 being extent low.



One more time


Kevin McKinney

Belatedly, and FWIW, I think the call is good. I had a feeling at the time that might turn out to be 'it'--though it's easier to say "I thought so" now!


Not working try 4 pieces cut and paste. Delete all++ and delete my second attempt . Thanks NJSF



2013 9 9 5.07939

Extent minimum was on September 9 Not the September 12 like I posted above.
Sorry, was posting mobile today.


Melt season ending September 17, 2013


Andy Lee Robinson

By "20013" there won't even be a maximum!


@NJSF - about 480 exajoules, or 4.55 x 10 ^ 17 BTU. That's about 1/12000 th of the annual energy we get from the sun by my shoot-from-the-hip calc.

It is not contributing to warming in any significant way.

As to the minimum, Wipneus suggested either the 9th or 10th in the forums, but some folks (myself included) want to wait thinking we could still see 50-60 thousand KM2 in additional melt. For practical purposes as far as the season is concerned, I think we are at minimum and won't see any more serious drop in extent and area.


Jdallen-wa Does not look like you will get that 50-60km melt.
Loop of the Arctic shows it locked up with no systems entering or exiting. Cold is locked up and expanding.
go to Northern Canada and Arctic Ocean IR loop


Andy Lee Robinson

I had two typos, 2013 and Extent was on 09/09/2013 not the 12th I posted.



This seems to be the time of year for thanking you Neven.

Thanks and congratulations.



@NJSF - in the the three years I've been following this blog, it's very seldom there is a poster certain about what's going on up there.

Would you comment about whether my "graph" of Arctic Sea Ice for prediction of 2014 Minimum will have a Red or Blue Line?

Colorado Bob

Antarctic Sea Ice Reaches Record Highs Due to Increasing Polar Winds



Jack Taylor that is off topic and I will reply just one time to it. I do not want to get the boot from Neven. I have to respect his rules on his blog.

My feelings.
I would say blue line again but not by much more. All the ice that made it is now multi year ice now. That multi year ice will become more thicker by the end of maximum season. I predicted Arctic sea ice for 2013 melt season in late winter of 2013 to be above 2007 minimum based on What sun spot cycle 24 was doing. I did not give exact ## for minimum.

I will predict minimum for 2014 melt season now based what sun spot cycle 24 now.

Based on Sun spot cycle 24's lack of sunspots and CME's now and if trend continues. The winter in the Arctic will be average to below with temps.

My prediction for 2014 minimum Area will will end at or near 5.7 mil km.

Kevin McKinney

"All the ice that made it is now multi year ice now."

Er, no.

Lord Soth

Well actually its not Multiyear ice until Oct 1, as that the convention that the Ice Service uses for the ice birthday.

Even then, it will be a mixture of multi-year and first year ice.

If we get anything resembling a normal summer, the first year ice will melt out, warming up the water around the multi year ice, and by the end of August 2014, a good chunk of multiyear ice will be gone, and we will be in a race for a new minimum.

2013 was an outlier, and pretty close to a black swan. I don't expect to see an arctic summer like 2013 for a very long time.


Kevin McKinney "All the ice that made it is now multi year ice now."

"Er, no."
Ok 12 days from now, All the Ice that did not melt this year I do feel is and should be named Multiyear Ice ON the date of the Minimum is declared( we already know that, Area, Sept 09 Extent & Area Sept 12). If more melts that is Ice from last Maximum then subtract that. Everyone on this blog has access to all the satellite maps on what is Ice from last maximum and first year ice that is growing today.

"Lord Soth"
"2013 was an outlier, and pretty close to a black swan. I don't expect to see an arctic summer like 2013 for a very long time."

That is your option but I do feel otherwise..

"Even then, it will be a mixture of multi-year and first year ice."

Yes and Ice grows Ice faster then open water grows melting waves.
New First Year Arctic ice has grown 100+km since Sept 12 in a week.


Jack Taylor I posted before from Mobil, Trick Question on Graph, my 5.7 mil km was according to your graph.
My Real World Est for 2014 minimum is 4.1 mil Km Area..
Good night

Jai Mitchell

If the anomalous and increasing lack of lower stratospheric moisture vapor, caused by a (very) unusually cool stratosphere that is apparently creating a barrier against normal tropical moisture convection flows continues, then the massively increased amount of tropical moisture that is maintained around the 250 mb height level will continue to move into the mid latitudes and eventually, once agin, into the arctic.

If this happens then next year's season will continue to produce lower arctic temperatures and anti-coriolis effect winds as this increased moisture and jet stream convulsions push more midlatitude moisture into the arctic circle where it will, once again, feed storms whose winds will reduce the amount of ice moving into the fram next year.

of course, if this happens, the tropics will continue to experience extreme precipitation events, the jet stream will continue to dip down and cut-off creating fantastical cut-off lows that last 21 days and drift backwards across the u.s. And we will most likely continue to experience record rains in the southeast and droughts in the west.

However, extreme mid latitude flooding events will continue, and probably increase in intensity. But, hey, at least the late march high pressure dome forming over the U.S. Midwest from the (now blocked) descending edge of the Hadley cell will not set up next summer and our breadbasket states' grain productions will be saved from a 2.5 month super heatwave/drought for another year.

the real question is, what could possibly be causing a barrier to convection currents in the lower stratosphere?



"what could possibly be causing a barrier to convection currents in the lower stratosphere?"

A very steep and strong inversion at the the tropopause, because the stratosphere is so cold that is not making immediate sense. The lower stratosphere should be unusually warm for this to happen.

The clouds, like extensive cloudiness hanging over the breadth of a strong Arctic anticyclone is plain weird, and your explanation makes sense in terms of cloud seeding wider areas than the tropical Pacific.

AMSR2 keeps on showing very little no new ice, but for the NW passage, which incidentally was temporary very thin ice, made when it was colder than -11 in light winds, disappearing when it warmed.

The said clouds which cooled a great portion of the Arctic during summer are continuing. There is a cloud induced temperature turn around point after the equinox, the clouds which cooled the arctic are now warming it.

John Christensen

@NJSF and Lord Soth on ice age classification:

- Young Ice: Has several sub-categories with First Year Ice being the oldest sub-category within Young Ice.

- Old Ice: Typically, you split between Second Year Ice (has survived two melting seasons) and Multi Year Ice (more than two melting seasons.

FYI therefore become SYI, and then a year from now MYI, if it lasts next summer as well.

John Christensen


You said on 9/17:
"Chris, area is surely the better stat, but consider this, compaction ridges the ice as opposed to a thin very large spread , nature of minima 2012 is not at all the same as 2013. A layer 50 cm thick is considered the same as 150. As said by Werther there was rubble pancake ice for about 2 million square km. Squeeze this ice together by 50% and you have a total area of 2.5 million km2. Same number or less than 2012."

While it is true that compacted ice take up a smaller area, you need to consider that the ice this year has not been spread any thinner than last year - both years have average thickness of 1.15-1.20 meters by the end of August (http://psc.apl.washington.edu/wordpress/wp-content/uploads/schweiger/ice_volume/Bpiomas_plot_daily_heff.2sst.png).

As you also see from PIOMAS, there was about 3.48k km^3 ice by end of August 2012, whereas we had about 5.2k km^3 ice by the end of August 2013. That is just a lot more ice to spread..

In other words, we have on average the same degree of spreading in 2013 as in 2012, and the ice cover therefore is so much more extensive in 2013.

John Christensen


Thanks for pointing to that article from Wunderground, which I had seen earlier as well.

As stated on Wunderground:
"The observations
We haven't been able to observe water vapor in the stratosphere very long--accurate global measurements only go back to 1991, when the HALOE instrument aboard the Upper Atmosphere Research Satellite (UARS) began taking data (Figure 1). Stratospheric water vapor showed an increase of about 0.5 parts per million by volume (ppmv) during the 1990s. But after 2000, a sudden drop of 0.4 ppmv was observed, and this decrease has persisted into 2009. To see how these changes impacted the amount of global warming, Solomon et al. fed the observations into a specialized high-resolution model that computed the change in heat from the fluctuating water vapor levels. They found that the increase in stratospheric water vapor in the 1990s led to about a 30% increase in the amount of global warming observed during that decade, and the decrease of 0.4 ppmv since 2000 led to a 25% reduction between 2000 - 2009."

Combined with the comments that climate models don't understand water vapor or tropical cyclones well, it leaves you with the impression that any prediction or modeling of future climate will have a significant level of noise or error, if you try to align the model with observations - especially if the model predicted the warming in the 90's, but did not include the water vapor impact, which is argued to have caused 30% of the warming..

This leaves me to focus on the ice, while others fix the model..


Hi Neven, I visit your blog occasionally but I think this is my first post here, so for transparency sake, I'm a joe public skeptic with a relatively weak scientific background.
I do have a couple of questions which I'm hoping someone here may be able to answer. It is in connection with Chris' link to his blog (reply to NJSF) showing correlations to PDO/AMO/CO2 etc.
I've often heard it suggested that the 2007 melt may be linked to the 1998 El Nino because the lag time for poleward ocean heat transport is in the region of 8-10 years. I'm not completely wedded to this idea but I do find it plausible.
Taking it further, this would imply that the very positive MEI period from around the mid seventies to the turn of the century is in part responsible for the gradual decline in arctic summer ice, with this year on year internal 'forcing' leading to volume loss. Of course, a simple correlation could be obscured by other influencing factors such as weather, solar cycle, AMO, and perhaps 'tipping points' caused by the gradual loss of ice volume. I'm thinking here of the 2012 arctic storms which probably had a far greater effect on the sea ice than a similar storm would have had 30 years ago due to long term volume loss.
So, finally to the question! Is this all just speculation or have papers been written to support or dismiss the concept. And secondly, in the event that there is some connection, how well might one expect to see a signal considering the significant potential for noise?


Additionally, one question for Chris Reynolds. From your plot correlating sea ice with CO2, do you consider the increase in CO2 as a generic proxy for temperature increase or is there more to it. For example, are you saying that atmospheric CO2 itself (irrespective of global air temperature) may be well correlated with sea ice because of the specific wavelengths of outgoing longwave radiation at average polar temperatures. Sorry if that sounds confusing, but I recall reading somewhere that CO2 absorbs at wavelengths which correspond to low temperatures.

Kevin McKinney

"I recall reading somewhere that CO2 absorbs at wavelengths which correspond to low temperatures."

Sounds like you are thinking about Planck's Law:

Every physical body spontaneously and continuously emits electromagnetic radiation. Near thermodynamic equilibrium, the emitted radiation is nearly described by Planck's law. Because of its dependence on temperature, Planck radiation is said to be thermal. The higher the temperature of a body the more radiation it emits at every wavelength. Planck radiation has a maximum intensity at a specific wavelength that depends on the temperature. For example, at room temperature (~300 K), a body emits thermal radiation that is mostly infrared and invisible. At higher temperatures the amount of infrared radiation increases and can be felt as heat, and the body glows visibly red. At even higher temperatures, a body is dazzlingly bright yellow or blue-white and emits significant amounts of short wavelength radiation, including ultraviolet and even x-rays. The surface of the sun (~6000 K) emits large amounts of both infrared and ultraviolet radiation; its emission is peaked in the visible spectrum.


Gases do emit and absorb specific frequencies, depending upon their quantum mechanical properties. For CO2, several of these lie in the infrared range--which is 'low' compared with, say, the temperature of the sun.

I doubt, though, that the Terrestrial temperature range is very precisely 'parceled out', since I know the responsive frequencies 'smear out' in practice due to the effects of differing temperatures and particularly pressures. But maybe someone much more knowledgeable about the details of radiative transfer than I may care to comment.

Kevin McKinney

"Ok 12 days from now, All the Ice that did not melt this year I do feel is and should be named Multiyear..."

As has been pointed out, it's not so simple as ice turning 'multiyear' on its anniversary; second-year ice is not the same as older ice, even if the brine rejection process has made some progress.

Perhaps more importantly, notice that at this time of year the same argument applies *every* year? That doesn't make for a solid basis for prediction!

True, the survival of a relatively larger amount of ice means that there will be more second-year ice this coming winter than there was last year. That doesn't imply a trend--as the examples of 2006 and 2009 illustrate.

Here's a chart from March of this year showing the evolution of the Arctic sea ice 'demographics':

Note the trend!


Thanks Kevin.
My limited understanding of the wavelengths stuff is that the absorption bands for CO2 are at the longwave end of the scale and that this corresponds to a very cold temperature, such as that found at top, or near top of atmosphere. I suppose what I'm asking is, if the distribution of OLR at the poles from the lower troposphere level is therefore more heavily biased to wavelengths that CO2 absorbs.


Intriguing thoughts, Lateintheday. While aware of the specific character of CO2 and how it interacts with IR to trap it, I hadn't thought about it from the standpoint of shifts in black body spectra related to surface and atmospheric temperature. It would imply also that different concentrations of CO2 would have different "preferred" temperatures.

Kevin McKinney

You are welcome, 'late.'

I should probably shut up at this point, having apparently failed to make a complete fool of myself yet, but the question is interesting.

I think the problem with your idea is that the objects emitting radiation are 'greybodies' that emit over a wider spectrum, not at the more precisely 'tuned' frequencies of the gases. For example:

Objects at room temperature will emit radiation concentrated mostly in the 8 to 25 µm band but this is not distinct from the emission of visible light by incandescent objects and ultraviolet by even hotter objects (see black body and Wien's displacement law).


So shifting temperatures of emitting graybodies by, say, 30-50 C may not have such a dramatic effect on the flux at the CO2 bands. (I'm thinking tenths of a nanometer, but those adept at applying Wien's Displacement Law could answer that more precisely/reliably.)

Kevin McKinney

Oops, just noticed that my ice chart was too big. Here's a (hopefully better) size.

(Yes, better, if I can trust the preview.)

Jai Mitchell

So, Very, Dry. . .

map of 400mb water vapor departure from normal for tropics from 7-06 to 8-30. caused by SE asia and Amazonian basin Aerosols? whatever the cause, I believe this is causing the shift in the jet stream and increased mid-altitude moisture transport into the northern latitudes from the tropics.


…joe public skeptic with a relatively weak scientific background…
“So, finally to the question!”

I like your style, Late!

Maybe Chris Reynolds will step in. But first, what’s the purpose.

OLR is absorbed by CO2, though I am not competent to know for sure I use the right words for the physical process of restraining heat within the atmosphere.

Your premiss is that CO2 has some physical alignment to absorb the specific OLR wavelength at low temps.
So what?
Why would there be a direct correlation between “sea ice” and CO2 content in the atmosphere over it? So you dismiss global air temps. Because it is obvious that an increase in those would also have an impact on the Arctic. And I guess you are sceptical about that?

But you agree that CO2 content has risen. So are you suggesting that it works specifically and regionally on sea ice? So the loss of sea ice could be seen as a remarkable but not really harmful side effect of fossil fuel burning?

And then, what is the idea of mentioning “average” and “low” temps? Do you really suggest that the CO2 might have had that regional effect on Arctic sea ice because the temps were, ´randomly´, average during the last decades? Implying the sea ice loss trend could reverse?

So, is there anything to learn here?

Pete Williamson

Beaufort Sea continues to look interesting.


Just from eyeballing the CT area chart it looks like this year wouldn't be out of place in the late-1980's and late-1990s. It looks like for each year since 2007 the Beaufort Sea has been in negative anomaly territory for more than half the year. This year it's looking like it's going to be negative for little more than 2 months. Given the endless observations of cracked and rotten ice in the region it seems rather remarkable.


Cryosphere Today animation latest few days make most of the perimeter of the pack shrink, compaction continues. AMSR2 has even more details of broken ice or open water which surprises. But CT showed a decrease in area by about 10K.

There is a couple of things to consider with Piomas John C.
Is it using the same area numbers as CT and AMSR2? Then there are issues to analyze, like fresh water at the surface between small ice floes, easily freezing and thawing daily. Fresh water does not follow the -11 C rule, it freezes at very near 0 C. A mix of less saline sea water than normal does likewise. There are examples even on todays CT where there is apparently new ice near Cornwallis Island while there isn't. An apparent disconnect between Piomas data and the very large extent of broken ice field which is vanishing first. A geographic illusion of sorts, triggered by extraordinary wind conditions.

Chris Reynolds


CO2 levels in this sense act as a proxy for wider effects. Strong amongst these effects in terms of Arctic sea ice are increase in ocean heat content and atmospheric warming.

Francis & Hunter find a link between the sea ice edge and a feedback between ice/ocean and atmosphere which increases downwelling IR.

This feedback is due to increased humidity of the atmosphere, not directly CO2 levels, but increased CO2 will cause a small increase in downwelling IR, which is significant particularly in the winter.

However there is also a direct effect. Eisenman finds a significant impact on sea ice in models due to the large spread in modelled downwelling IR.

Again this is particularly important in winter. And as I note in that post:

Jakobsson et al find that the change in July 65degN insolation that lead to near ice free conditions in the Arctic around 6 to 8k years ago was around 40W/m^2. This gives an indication of the significance of the 40W/m^2 spread of IR radiation in GCMs, especially when you consider that July insolation isn't being applied year round, as the spread in IR is.


werther, although I'm a sceptic I'm not completely deaf to the CO2 theory and so for me, there is always more to learn. I frequently visit a number of blogs of both pursuasions and try to fathom the various arguments best I can with limited science/math skills.

I was interested in Chris' link (a reply to NJSF) in which he showed poor correlations between arctic sea ice area and the PDO and AMO (from memory) but a strong correlation with CO2 levels. I may have misunderstood the plots.
Chris has now replied (thanks) to clear up a point which evaded me. The strong correlation between CO2/SIA is assumed to be wider effects rather than a very localised and direct effect caused by the absorption properties of CO2 and the specific temps at the poles.
I'd assumed (still assume) that the polar amplification of warming was to do with increased heat transport from mid latitudes and that because of lower humidity (compared to tropics), a larger temp increase from the same amount of energy would be expected. I thought I'd maybe missed something here. That possibly, polar amplification was somehow tied in with the spectral physics of colder places.

Chris - thanks for the links.


lateintheday My limited understanding of the wavelengths stuff is that the absorption bands for CO2 are at the longwave end of the scale and that this corresponds to a very cold temperature, such as that found at top, or near top of atmosphere. I suppose what I'm asking is, if the distribution of OLR at the poles from the lower troposphere level is therefore more heavily biased to wavelengths that CO2 absorbs.

The shift in the emission band to longer wavelengths is small, a somewhat higher fraction of the band will be in the CO2 absorption band but the total emitted will be will be more effected, perhaps about 50% less than at 25ºC.


Jim Hunt

The ice edge off "Russia's northern shores" doesn't really look like it's refreezing just yet:

Click the image for a larger version, along with much more on David Rose's investigative inadequacies!


Nice work Jim, there is many many places on EOSDIS without ice while CT shows some. Suffices to study like North of Southampton Island, South and west from Cornwallis etc. I did see fresh ice at some nights disappearing in the morning, but temperatures are basically too warm for massive ice formations:


clouds causing summer cooling now provide fall time warming.

Chris Reynolds


Most of the polar amplification is actually due to sea ice loss. e.g. Screen & Simmonds 2010 "The central role of diminishing sea ice in recent Arctic temperature amplification."

I've got the workings of a blog post I never seem to get round to posting - will do this autumn. During March April May (MAM) the relationship between sea ice thickness (PIOMAS) in the Siberian sector of the pack and temperature is: -6.02degC/m, the R2 is 0.698 (R2 is 'R squared' and ranges between 1 for a perfect 1 to 1 relationship, and 0 for no relationship). In other words, as the ice has thinned in the Siberia sector temperatures when that region is all ice (open water isn't a major factor) have increased from an average of about -17degC to -11degC, while the ice has thinned from about 2.6m to 1.8m.

The thinner ice is letting more heat through in the spring winter. The resultant warming profile is biased towards the surface.

This pattern is consistent throughout Sept to May, but in the autumn open water plays a big role.

Intelligent questions deserve considered answers, and I'm always happy to oblige.

Pete Williamson

So the possibility of another crystal-ball gazing game in the arctic.

This graph shows that the winter maximum of the CT ice area hasn't broken 14 million for 10 seasons


Given the spooky melt season this year what are the chances that that particular upper limit might be breached.


cycle lenghts of ocean layers overturning are of course not very well constrained. some have suggested ~30 years for some cycles, presumably unconnected of the ENSO state, this might be possible, this would require f.e. the water mass exchange between south Pacific and north Pacific to switch location respective to the ENSO state. What happens under Cromwell current? Possibly some buoys currently operating could answer to this, but these have been on service for too short a period. Meanwhile the surface layers get more acidic (or less basic, if you're semantically oriented person (the two words 'acidity' and 'basicity' are used to describe the same thing)

Jim Hunt

The pole and nearby polynya are disappearing into the darkness from MODIS's perspective. However the sky is still clear on the Laptev side of the CAB. Here's a close up:


Smap filter caught legitimate comments by Chris Reynolds and Erimaassa.

Jim Hunt

In the latest Arctic Sea Ice News the NSIDC have called the minimum:

On September 13, Arctic sea ice reached its likely minimum extent for 2013. The minimum ice extent was the sixth lowest in the satellite record, and reinforces the long-term downward trend in Arctic ice extent.

Harking back to that Mail on Sunday article, they also reveal that David Rose's "unbroken ice sheet" was no such thing.

Jim Hunt

And I nearly forgot, the Arctic paints a very pretty picture for us today:


Whenever lakes freezes over in the area where I live, cold and clear skies are not always the things that matter the most, but rather how much snow that falls on topp of the ice. I mean, cold is of course a very important factor, but my experience is that a thick insulating snow cover can prevent the ice from growing properly even in the coldest of times, and vice versa, you don't really need that much cold to grow strong, thick ice if there is no snow on top of it.

I wonder to what extent these insulating abilities of the snow might effect the regrowth of sea ice during fall and winter. If, lets say, a storm where to dump a abnormal amount of snow of newly created sea ice, or if snow depth remained 50% above average for the entire winter, how much, if any, would it effect ice thickness and the following summer melt in that particular area?


Hi folks,

Hithero I've been a longtime lurker and value the realtime analysis and projections for the Arctic Sea Ice end game.

I'm a bit concerned that when LateInTheDay says "I recall reading somewhere that CO2 absorbs at wavelengths which correspond to low temperatures." (s)he actually means something fairly simple, like the idea that because the Arctic is 'cold', CO2 affects it more, though dressed up a little by referring to terms such as Outgoing Longwave Radiation.

The confusion is between a commonsense understanding of 'cold' and the physical term for a 'cold' radiating body. The upshot being the formulation of a skeptical 'inverse canary' argument: "Arctic amplification is due the coldness of arctic sea ice specifically causing CO2 to have an effect; CO2 doesn't affect warmer parts of the planet and therefore we can ignore the Arctic".

In reality, Arctic Amplification is due to other factors, e.g. changes in albedo.

I'll get back to lurking now :-)


Lars Boelen

I thought this article was very interesting : http://theconversation.com/is-global-warming-in-a-hiatus-18367

It states that in the past 15 years SURFACE TEMP has stayed the same (it does mention heat absorbed in the oceans kept increasing) where they should have declined by 0,4C because of solar minimum and back-to-back la nina's.

Learning every day, I hadn't realised that all the extreme melting going on took place during a solar minimum. Makes you wonder what's going to happen once the cycle heatsup (even if it only adds 1 W/m2) and el Nino's start re-occuring.

John Christensen

wayne wrote (9/20):

"There is a couple of things to consider with Piomas John C.
Is it using the same area numbers as CT and AMSR2?"

No, PIOMAS is a distinct model based on ice concentration and thickness data and with a distinct grid model:


It is still a model, but from discussions on this blog and elsewhere, it appears to provide quite reliable data on ice volume.


OT - just want to compliment everyone on how they have responded practically, politely and conscientiously to people with opinions that differ from their own. I value immensely the "collegial" atmosphere of this blog and Neven for guiding the discussion.


Jeff Master's blog http://www.wunderground.com/blog/JeffMasters/comment.html?entrynum=2531&page=1#commenttop has some interesting things about what happened in August and some figures as to what happened in 2013 as to costs in weather. Note at the end he does mention the Arctic and that his Tues. blog will be about the minimum. Should be good as he and his team understand the correlation between AGW and weather. eg no single event can be attributed to AGW, but tends definitely do.


Very interesting onset of ice near Cornwallis Island, before the ground was completely covered with snow, at first with temperatures lower than -11 C but now that its has warmed to -8 C the ice spread out fast to Barrow strait, only to be blown away, in a little wind.

Does anyone have the link to the sea surface salinity daily composite map of especially the Arctic?

I think that salinity plays a big role in delaying the re-freeze, since 2013 appears to have less SSS in the colder area of the Arctic, the refreeze seems faster. However salinity is stronger elsewhere stronger with much higher temperatures. The map for this refreeze may be with the spread of where all the salt is.

John Christensen

@wayne on SSS;

DMI has a lot of cool stuff to check out, and a model and forecast of surface salinity (You select the needed paramaters):


There actually seems to be an upwelling near Barrow, which should delay refreeze right there. You also clearly see the outflows from Russian rivers on the Siberian side.


There is a ship in the field of view of North Pole webcam #2.


Near where there was a lead almost all past winter.


John Christensen - Thanks! I hadn't seen that DMI link. SSS is one of the factors I've been mulling over regarding the odd behavior of ice as we move into refreeze.

Clive Mitchell

Sorry for being off-topic but with the IPCC/temperature increase hiatus debate, does anyone have the wherewithal to plot flattening temperature rolling average and ice volume minimum rolling average on the same graph, crossing over around 1998, please?

Or let me know where I can find the data?



Hi folks, does anyone have the September PIOMAS Arctic ice volume figure please? I estimated it as 5000km^3 but it might have gone down a little since then. I know we won't have the final figure just yet.



Thanks John, DMI shows more than NASA's Aquarius which has scant swaths which tend to have less salinity in the Canadian Arctic.


Clive Mitchell

Further to my former post, I found some data, thanks.


In a four or five year rolling average plot of surface temperature change superimposed with rolling average September Arctic ice volume against time, the ice volume appears to decrease at a steady rate then shows a distinct change to a new steady rate, the change corresponding to the flattening in the temperature curve.

Of course, I'm an amateur and this could be bunkum but if the surface temperature rose until the late '90s when a resultant small increase in the rate of ice melt was sufficient to absorb all the heat formerly going into surface temperature increase, what sort of temperature rise can we expect when the ice melt rate levels out?

Let's hope it's just amateur bunkum!


I guess my question is:

Now that we have the minimum, when do we start plotting the Maximum.

No, not the Maximum SIA or SIE, but the Maximum anomaly for 2013 on the CT plot.

It's growing rapidly as the ice area and extent stalls. As the ice does not regrow quickly.

I'm wondering if we will, yet, as we did in 2006, have another interesting statistic in this most interesting of all ice years?


Also If I read my NOAA recent global monthly mean CO2 charts correctly, peak to peak, 2012 to 2013, was a 3ppm growth. I don't know they actually do the calculation but even the 2013 low is looking like being higher than the 2011 high. It would have to drop more than 2ppm and I haven't seen that before at this stage in the chart.

Interesting to see what that impact will be. The last time we saw this much rise in CO2 was 1998 which was driven by 1997 events.


A handy two page summary of the IPCC report --


Chris Reynolds

Clive Mitchell,

Here's a quick and rough graph for you, I've just started working on a post about all the the GW stopped bollocks.

We should not necessarily expect a more massive rate of loss once AGW starts off again. It is quite possible that we will hit a floor as Autumn ice growth retards winter thinning. PIOMAS shows this behaviour in future projections.

The current lack of warming may in part imply a continuation. While the ENSO component (1,2,3) seems to be random, and should bring about a warming once it desists, there is a possibility that human emissions of sulphates are having a cooling effect (1) which seems likely, and this is not likely to desist any time soon. And solar activity (1, 2) is rising to a new maximum, but this is going to be lower than the previous several maxima. Furthermore there is the issue of loss of Arctic sea ice affecting autumn snow line advance and cold winters in Eurasia (4), that factor doesn't seem likely to abate soon.

However the recent lack of warming in the annual mean seems to be a consequence of ENSO, solar activity, and sulphate aerosols. So a persistent cessation of GW looks unlikely, we hit 400ppm this year, this increasing radiative forcing cannot be hidden for many more decades.

What does this mean for sea ice? That the ice volume at minimum has declined at an accelerated rate over the very period of the GW hiatus shows that sea ice is chasing down to an equilibrium that is already set at lower sea ice volume than current levels. In other words the hiatus in GW is irrelevant for sea ice. This is further supported by the regional warming of the Arctic.
That shows a plateau of sorts around the same time as the GW hiatus (globally), yet the ice volume loss has accelerated over that period.

1, Kaufman et al, Reconciling anthropogenic climate change with observed temperature 1998–2008.
2, Foster & Rahmstorf, Global temperature evolution 1979–2010.
3, Kosaka & Xie, Recent global-warming hiatus tied to equatorial Pacific surface cooling.
4, Cohen et al, Asymmetric seasonal temperature trends.


Sept PIOMAS volume data?

It's 27 Sept - go figure.


Hey Chris...

I'm going to back yup up even further.

"We should not necessarily expect a more massive rate of loss once AGW starts off again."

That in *itself* is a misconception. AGW has not stopped, slowed down, or in any way checked. Not even temperatures have altered particularly.


If air temps are lower, its because more energy is being picked up by the oceans. Net estimates of specific heat in them all support that. I think that is *directly* reflected in increased heat reaching the arctic via warm north Atlantic currents, and may have had a key role in how far back the Barents/Greenland Sea margins pulled back this year.

The heat hasn't stopped. CO2 forcing hasn't changed. There is no lack of warming. There is only changes in where the heat is being redirected.

Chris Reynolds

JD Allen,

I agree that to conclude that CO2 forcing has stopped is plumbing the depths of stupid.

However with regards the surface air temperature that we commonly use as an index of GW there has been a pause in the upwards trend since 2002. Those saying since 1998 are clearly incapable of reading graphs so should be ignored - too witless to bother with. The papers I cited (1 to 4) together show a plausible narrative as to what has been going on. ENSO seems to be the largest part, but as IPCC AR5 SPM notes, determining the exact contributions is difficult. However what is clear is that the pause in surface warming is due to processes separate from the underlying forcings (CO2 being the largest single forcing, but CH4 (direct & indirect) is over half that of CO2).


Does anybody know if there is somethng like a "cummalative rotation energy of the atmosphere". I'm asking because I have a feeling that the atmosphere is not rotating as fast as it used to. Here in Holland we seem to have less SW wind and more N or S winds. If this were a global fenomena the rotation energy of the atmosphere would be lower. This would be a proxy for the condition (heat content) of the poles : warmer poles = smaller delta T with tropics = less energy to drive the jetstream.

On a sidetrack : Neven, how is your house getting along?

Clive Mitchell

Thanks, Chris. Interesting papers.
I'm not sure about sea ice chasing down to a lower equilibrium. If excess heat inflow is consistently available at a temperature which will melt sea ice, sea ice will melt until sea ice is all gone - the melting point of ice is not a continuous variable to be adjusted for equilibrium purposes. The new equilibrium will be reached when the heat inflow and outflow balance and I fear this will involve a temperature which is too high to sustain sea ice.

On a global scale, less heat is being employed raising temperature and more heat is being employed melting ice, but as you imply, there are a great deal of other mechanisms.


On a sidetrack : Neven, how is your house getting along?

There's a thread on the ASIF where I posted some images a couple of days ago. Will update later today. The combination of computer work, physical work and tension has resulted in a wonderful migraine, lasting for the entire week. But we're nearly done with all the big stuff, and then things will become more flexible.

Susan Anderson

Larsboelen, I'm wondering if you would be interested in Jennifer Francis's work. She is working on the hypothesis that as the temperature differential between the poles and the equator gets to be smaller, the jet stream is getting slower and more eccentric, with blocking patterns. Forgive me if you know all this, or I've oversimplified. She has a number of longer presentations (AGU, for example - a search should find some of them) but this, if you can skip over some of the drama, provides a shorter version to introduce you to the topic. The jet stream comes up after minute three (Arndt as well):

Chris Reynolds


Are you talking about a specific season?

I recently had a similar comment from one of my email correspondents, again without specifying season. When I sent this image to him he said this was what he was talking about.

This shows flow of air around the planet (left/right not up/down). The plot is of difference from climatology (1981 to 2010 average) with purples being more east to west than usual for three recent cold European winters mentioned in the literature as being connected with the loss of sea ice ( in order - Petoukhov/Semenov, Cohen et al, Cohen press article).

Clive Mitchell,

OK, 'equilibrium' isn't the correct term as the whole state is in flux. However my point remains that indications are that the ice is being forced into lower volumes by a bias acting against ice survival. An apparent equilibrium may be about to assert however, it could have done so post 2010. For PIOMAS the three recent winters have seen near equal maximum volume (Cryosat however shows a decline). PIOMAS forecast shows that autumn ice production eventually stems the loss of volume causing a long tail of ice survival for some decades.

See - Zhang et al 2010 "Arctic sea ice response to atmospheric forcings with varying levels of anthropogenic warming and climate variability."

In that study post 2007 shows an increase of September volume (figure 1 of the pdf paper - which uses the old PIOMAS model version). In the new PIOMAS version data (main PIOMAS series) September average volume rises after 2007, but falls from 2009 due to the 'post 2007 recovery' working through the sea/ice system followed by the 2010 volume loss. This year will see an increase of Sept and annual average volume, which being due to weather is uninformative. The volumes to watch out for with respect to the autumn growth limiting volume loss will be from Sept 2015 onwards. Although it is conceivable that the 2013 volume gain will work through the system by next September, I personally doubt it.

However May (peak) average volumes for 2011 to 2013 have been level at around 21.7k km^3, which may be the start of a tail of persistent ice survival. If we see volumes in May 2015 onwards tending to 21.7k km^3 then I think it will be strongly suggestive that we're in for a long tail, not a fast crash.

John Christensen

NeilT wrote on 9/27:

"Now that we have the minimum, when do we start plotting the Maximum.

No, not the Maximum SIA or SIE, but the Maximum anomaly for 2013 on the CT plot.

It's growing rapidly as the ice area and extent stalls. As the ice does not regrow quickly."

I don't quite see the rapid growth of the CT SIA anomaly and ice area and extent numbers are in fact going up at normal to strong rates:

Arctic ROOS:
(http://arctic-roos.org/observations/satellite-data/sea-ice/ice-area-and-extent-in-arctic) and
CT: (http://arctic.atmos.uiuc.edu/cryosphere/arctic.sea.ice.interactive.html)

While it is correct that in recent years the negative SIA anomaly on CT has increased into October (http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/seaice.recent.arctic.png), September-November freeze happens mainly in eastern Kara, Laptev, ESS, Chuckchi, Beaufort, CAA, and down into Hudson, where SSTs in most places are below normal (http://ocean.dmi.dk/satellite/index.uk.php).
I would therefore expect the max anomaly for 2013 to have been reached in the second half of July around -1.5M km^2.

For November-January, however, the picture may change, as freeze then will move into Barents, Baffin, Bering, Okhotsk, and Greenland, where SSTs are higher, so max extent and area numbers for march 2014 may not be increase compared to march 2013, but let's see.



I wasn't thinking quite so much about right now, although there appear to be stalls in the growth for a few days at a time.

Look at 2006 compared to 2005 and around day 300 timeline.

I'm wondering if we are going to see similar.

2006, much like 2013, started low and melted fast. Then just stopped for so long that it could never catch up. Sound familiar?

Yet when winter 2006 came on, it started to rise like every other and then, just stopped, for such a long time, that 2007 was set up for it's epic run.

I'm just wondering if the cycles really do run that way or if 2006 was a one off.


Hot hot hot off the press.


Jai Mitchell


thanks for the link, figure 4.15 (cumulative ice mass loss for Greenland) shows an exponential curve with a doubling of sea level equivalent every 4 years.

if this curve continues according to it's current exponential growth through 2052 (10 doublings) the ice melt from Greenland will contribute 8.2 meters of sea level rise.

Similarly, figure 4.16 (Antarctic ice mass loss) shows a doubling every 7 years which will produce .35 meters of sea level rise by 2054.

If these trends continue then a cumulative sea level rise (from 1990 levels) will be 8.5 meters by 2054.

Peter Ellis

Jai: And if the curve continues exponential growth through 2060 (12 doublings) it'll be many times that. If it continues until 2100, the Earth will collapse into a black hole made of water *

My point is that melting the entire Greenland ice sheet would cause about 7 metres of sea level rise. Unconstrained extrapolation of exponentials is insanity.

* Well, maybe not. I wonder how long it would take?


We're all wondering the same. Part of the scientists seems to think it will take at least a thousand years. Another part seems to fear it may take as little as three or four centuries, if we continue with BAU.


Part of the scientists seems to think it will take at least a thousand years. Another part seems to fear it may take as little as three or four centuries, if we continue with BAU.

The 64 dollar question. The betting money is on the side of it taking longer. At 334KJoules/KG, it will take an awful lot of heat to melt 2.85 million KM3 of ice. Even more than doubling the current average melt rate to 400KM3/year we are still looking at a minimum of 700 years for the cap to melt out. At the low end, that time frame is over 14,000 years. In the short term, I suspect there will be more sea rise as a direct effect of thermal expansion than by way of melt.

I do think we could see more dramatic peripheral melt back, but the core of the icecap is pretty well constrained by the odd configuration of mountain ranges along the circumference of Greenland. With that ice stacked up over 3KM deep over wide areas of the island, it will take a very long time for the heat to get to it. As a comparison, there are estimates that the relatively small ice "erratics" that formed many kettle ponds in my native New England may have taken centuries to melt even after detached from the glacial front.

I don't see Greenland as a near-term threat at all, and as a species, if we are smart enough, we will be able to cope with its now almost inevitable melt-out. I'm much more concerned about the short-term weather disruption which will come hand in hand with arctic sea ice melt out, and related problems associated with ocean acidification.

Jai Mitchell

correction, I decided to actually plot the data, the greenland curve follows a best fit (R^2=.999) of y=.0209x^2+0.949x+0.9768

which charts the following cumulative sea level rise due to ice melt as follows:

year mm
2012 8
2022 19.2
2032 34.6
2042 54.2
2052 77.9

VERY different from the exponential increase before. Of course, the rate of increase in the acceleration of warming is going up so these values are an underestimation.

Jai Mitchell

Antarctica has a much higher curve fit due to the rise in the rate of land-based ice mass loss (fig 4.16)

The curve fit has an R^2 value of 1 and its equation is (y = 0.003x^3 + 0.0227x^2 + 0.2924^x + 1)

its outputs are as follows

year mm
2011 9.1
2021 39.9
2031 111.2
2041 241
2051 447.4


"Even more than doubling the current average melt rate to 400KM3/year we are still looking at a minimum of 700 years for the cap to melt out."

You mean 7000 years, I suppose. Then again, over the past three years GIS lost about 400 km3/yr on average. Let's say this will continue until 2020. And then could this double in another 20 years to 800 km3/yr from 2020-2040? And then again to 1600 km3/yr from 2040-2060, and again to 3200 km3/yr from 2060-2080? Then the rate would be about 9 mm/yr of SLR contribution from GIS. If that rate could be sustained, with positive and negative feedbacks about in balance, then it would take about 700 years for near melt-out of GIS.

I don't know if this is possible. And the risks for Antarctica seem higher. But scientists like Jim Hansen do suspect such accelerations could be possible under BAU. So I think it's too early to completely dismiss such scenarios.


"With that ice stacked up over 3KM deep over wide areas of the island, it will take a very long time for the heat to get to it."

Are you sure?

Seems from this paper that all that hot melt water takes the latent and sensible heat to the centre of the ice sheet, meaning they will melt like microwaved butter from the inside out, they call it cryo-hydrologic warming of the internal and base of ice sheets. Basically heat gets into the ice sheet a lot faster than previously thought, and not in any melt models yet!

Philips T. et al, "Evaluation of cryo-hydrologic warming as an explanation for increased ice velocities in the wet snow zone, Sermeq Avannarleq, West Greenland"

Journal of Geophysical Research: Earth Surface.



Hi all,

it's been a while...

there is a live online discussion on the Guardian about Arctic drilling, possibly here...



In Physics Today (Oct 2013), a nice summary:
"The Arctic shifts to a new normal"
by Martin O. Jeffries, James E. Overland, and Donald K. Perovich

Shared Humanity

I certainly have little to contribute about the possibility of accelerated melt of the GIS. I simply do not know enough about any of this stuff. I can contribute with regards to the behavior of systems, particularly "growth" systems. All growth systems display the characteristic of exponential growth and charts displaying an exponential growth trend are evidence that we are measuring a growth system. Growth systems display these trends because of positive feedbacks. Examples of this type of growth system behavior are everywhere (compounding interest, population growth etc.). Figure 4.15 certainly suggests that GIS ice loss is just such a system.

Are the currently understood positive feedbacks (forcings?) sufficient to cause GIS ice loss to behave as a growth system? Put another way, are CO2 or surface temperature rise (or any other positive feedback) sufficient to explain this behavior?

This is the question I cannot answer because I simply do not have enough knowledge. If the answer to this last question is no, it does not mean that GIS ice loss is not a growth system. It does suggest that there are positive feedbacks that are not identified or understood that are driving this exponential growth.

System behavior (exponential trends in the case of growth systems) will continue to behave in a predictable manner as long the system is operating within the constraints of the system. All systems have constraints. In the case of population growth, the major constraint is the absolute carrying capacity of earth. As the exponential growth of population approaches this constraint, negative feedbacks (famine, disease etc.) will stop this exponential trend.

Given that the GIS ice loss has only recently displayed this exponential trend, I suspect we are not even close to any system constraint at which point negative feedbacks predominate system behavior. We should expect the exponential growth trend to continue for a long time.

Shared Humanity

jdallen_wa suggested just such a system constraint in his comment above. Perhaps it is a true constraint in the GIS ice loss growth system. If this is the case, we should still see exponential growth in ice loss until this constraint can alter or constrain system behavior.

Shared Humanity

I have also read articles that suggest the topography of Greenland could serve as a constraint. The bowl or valley in the center of Greenland and its very thick ice has been proposed as a constraint to ice loss. If true, the accelerated ice loss will continue along the edges of the GIS until it bumps up against this constraint. What rate of sea level rise could we expect if the perimeters of the GIS would experience exponential growth in melt until it disappears? How quickly would it disappear given this exponential growth?

Shared Humanity

I believe there is some historical record regarding previous warm periods where the Greenland topography seems to exhibit just this type of impact with ice receding around the perimeter and along the southern third of the land mass while persisting in the center.

If topography plays such a role in ice loss and I can't imagine that it doesn't than my attention is drawn to figure 4.18 which is both fascinating and terrifying. The WAIS sheet displays no such topographical constraint. I've read numerous research articles linked to here that conclude that this ice sheet poses a far greater threat with regards to sea level rise. An exponential trend in WAIS ice loss would continue unabated until the entire ice sheet collapses. Are we seeing such a trend now? If so, how fast would this ice sheet melt given a continued exponential trend?


Nicely put.

The strength of positive and negative feedbacks will determine the outcome, together with the GHG-forcing. And it seems we all do not know enough about how the ice caps will respond to this kind of forcing. GIS also seems to have two glacier fjords which extend deep into the interior of the ice cap, at Jakobshavn and Petermann glaciers. So although WAIS seems more unstable than GIS, this may be a pretty vulnerable ice sheet as well.

Jai Mitchell

Models show that once the arctic is ice free, air temperatures are going to go up in a step change by 5-10 degrees C during the late summer/early fall months. This is peak melt season.

At the same time, an increase in atmospheric moisture is supposed to increase precipitation.

In the short term, the increase in snow fall will make up for some of the GIS loss, but the rate of loss will continue to accelerate until about 2065 when the Thermohaline Current drops to about 1/12th of its current rate. At this time large amounts of subtropical moisture will move further northward and rainfall on the GIS will increase melt rates, both on the surface and internally, on a massive scale.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Your Information

(Name is required. Email address will not be displayed with the comment.)