« DMI SST maps back up | Main | NSIDC Arctic sea ice news mid-August 2011 »


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


This is of course rather good news, if true. I remember we discussed an ice-free Arctic in winter (based on the Tietsche et al. paper, IIRC) and found some of the arguments for an ice-free Arctic in winter quickly following an ice-free Arctic in summer quite convincing.

Too bad we can't turn down CO2 down in reality. I'm still a bit apprehensive what happens to NH weather patterns when things continue as they do (and of course the other thing I'm in denial about), but it's good to hear that things may reverse more easily than feared.


Currently the Arctic is pretty cut off from the rest of the worlds oceans, so it is pretty expected that it could freeze in winter, even the Baltic does that.


C02 45 million years ago- ten million years after the PETM around 550-600ppm- the Planet was ice free.

With C02 at 450ppm sustained the arctic should be ice free most of the year- per Dr. James Hansen.

At 550ppm (doubling of C02 from the preindustrial level) the Planet should see an ice free arctic year round- assuming a 3 degree C rise or more in global temperatures. The western antarctic ice sheet will collapse. The study seems to fly in the face of other models- which see a much greater sensitivity to warmth.


First, i don't see what's so interesting about these papers. The finding, that the ice will regrow, when the planet cools, is trivial. Of course there will be ice at the north pole when temperatures are cold enough. You dont need a climate-ice-model to come up with that idea.

Second, the usage of the term "tipping point" in these papers seems for me a little bit flawed. They seem to imply, that - once you have passed a tipping point - there is no going back. But that is IMO simply not true. A tipping point is simply like the top of a hill.

A _.--._ B

Imagine, you have a heavy metal sphere, which you want to roll from A to B. You need to push really hard, to get it there, because you have to climb the hill. If you stop pushing before you reach the top, it just will roll back. But once you have passed the top, it will roll to the other side. So, you have to stable states, and the tipping point separates both of them and tries to keep you at the current state.

If you have no tipping point between A or B you simply can push the sphere much more easily, because there is no potential, that you need to overcome.

A ________ B

But in this picture, a tipping point can be passed in both directions. You can always go back, but you need in any case enough power, to surpass the hill, dividing the stable states.

IMO it is clear, that sea ice is more like the hill-case. As long as the ice is there, it cools the earth by reflecting sunlight and thus working against the effort to melt the ice. But once you have melted it, the open ocean absorbs more heat and makes the world warmer and works against the effort to grow the ice back. So, you have two stable states, and the ice resp the open ocean are acting like the slopes of the hill and try to preserve the current state. But with enough cooling or warming you cleary can pass the point in every direction. It just need a stronger force, as if there would be no tipping point.

So, if we pass a tipping point in the arctic, warming intensifies and it will be much harder, to go back to previous temperatures. You would need stronger cooling forces as if there would be no tipping point. And IMO this is exactly what you would expect by the ice-albedo feedback.


After the brow of the hill there could be a steep drop or cliff meaning there is no coming back but these studies are telling us it isn't like that.

So we can think of either relatively shallow slopes on each side or maybe the hill just continues upward for a long way yet?

This is good / a relief but returning in that direction anytime soon seem unlikely.


@Newcrusader, the ocean circulation has changed considerably since then including the closing of the Panama Ismuth. The past can guide us on what to expect but will not be an exact mirror.

Seke Rob

We just need to think of such things as the continental configuration now and then and land cover then, no 7 billion to cut down the forests. Fossil Fuel drillers use MRI since 20+ years to find the original orientation of geology. Drill 2-3 holes, and position the cores and water mols and you've got a pretty good idea of the long ago orientation. They've been knowing for a very long time and won't have anyone interfere with the future Arctic earnings potential. New field just yesterday in news of a 500-1500 barrel new discovery in the North Sea... good for not 20 days consumption on a global scale, at 83 million barrels daily (data 2009).

Artful Dodger

The paper can be obtained here:


Al Rodger

Re climate 45 million years ago.
As well as no Panama isthmus, the Himalayas & Tibetan plateau would still be in the early days of formation, both of which are often mentioned as factors in creating the cooling climate.
Climate-wise 45 millin years ago is like a different world.


Thanks for a very clear explanation, Reasonablemadness.

the Himalayas & Tibetan plateau would still be in the early days of formation, both of which are often mentioned as factors in creating the cooling climate.
They have a high albedo at a low latitude hence reflect a lot of energy out again, also are responsible for sequestering a great deal of carbon by rock weathering.


But there are strong cooling forces.

The loss of heat to the atmosphere due to prolonged open water into the freeze season is shown by Tietsche's model study to cool the region, by both radiation to space and reduction of inward atmospheric heat transport. This happens to such a degree that even with a total removal of sea-ice the ice recovers to it's equilibrium state prior to the removal of the ice.

Regards tipping points, one would expect one to exhibit hysteresis e.g.
So to get back to where you started you have to back-track to well before the point the state flipped. The research is telling us that the Arctic is not like that (and there's a growing body of such research using models of various complexity).

For me the most surprising result is the extraordinary high CO2 needed to hold a perennial ice free state. I'd thought it was around 1000ppm and above - based on paleo studies. The increased atmospheric heat transport implied by the Arctic Dipole could account for a lower level of CO2 maintaining year-round sea ice free. Watanabe et al 2006 used MIROC3.2 and found the AD was reproduced well. I don't know if the model used in this study, CCSM3, reproduces the AD. But as MIROC does, perhaps the high CO2 finding takes into account atmospheric heat transport. In which case we may only cause a seasonally sea ice free state, not a perennial one. Interestingly both CCSM3 and MIROC were two of the models chosen by Wang & Overland in their 2009 study "A sea-ice free summer Arctic within 30 years?" In that paper the ensembles of CCSM3 used seem to be indistinguishable from observations. Yet the ensemble members closely agree on a seasonally sea-ice free state around 2050. With a virtually (<1M km^2) state around 2040. So perhaps I'm being too pessimistic in expecting a virtually sea ice free state by next decade.


Oops! Should have signed the above.

Chris R.

Michael Stefan


So, if we pass a tipping point in the arctic, warming intensifies and it will be much harder, to go back to previous temperatures. You would need stronger cooling forces as if there would be no tipping point. And IMO this is exactly what you would expect by the ice-albedo feedback.

That is the way I have always seen it as; the notion of an irreversible tipping point is clearly absurd or the planet would be ice free right now (e.g. once it became ice free, it would always remain so). In any event, once the Arctic becomes ice free in the summer, it will likely require a much cooler climate than today to restore the ice. Also, while it is true that the ocean circulation/geography 45 million years ago doesn't have much bearing on today, paleoclimate is still useful in assessing the climate response to forcings (e.g. a recent study that found that the climate is twice as sensitive to CO2 as models show, never mind 16°C warmer at CO2 levels projected for 2100, or that the Arctic was ~19°C warmer than today a few million years ago with slightly higher CO2 and similar ocean circulation/geography). Note also that the Sun was dimmer in the past (much more so than over a solar cycle), which is also a big factor in comparing past to present climates.



Ocean circulation has not changed that much

that would require 3200ppm CO2 instead of 550ppm for an ice free planet.

YOU are debating a straw man- 3.5 million years ago in the mid Pliocene when C02 was around 390ppm- there was little ice in the arctic year round- Ocean currents where not that much different then today.


Al Rodger

and shall I repeat- 3.5 million years ago
in the Mid Pliocene- when C02 levels where near today- there was no ice in the arctic most of the year- and sea levels where 20 meters higher

you are debating A straw man here.


OF COURSE there will be ice in the winter. After all, in our present climate even the Great Lakes (U.S.), Gulf of Bothnia(NW Europe) and northern Caspian Sea (Middle East/Asia Minor) are NOT "ice free" in the winter, despite having MUCH, MUCH milder and shorter winters than the Arctic. If we get 1000 ppm, my intuition is the Arctic Ocean will be the way Hudson Bay is now - ice-free for a few months and up to 1m ice thickness in late winter/early spring...


Also, the ice albedo effect is grossly exaggerated in most popular discussions - after all, the seasonal timing of snowpack establishment and retreat in the N Hemisphere landmasses also seems quite stable (and snow has more albedo than ice with melt ponds). This is in large part because ice and snow have lower albedos in the infrared. Since 50% of the energy content of solar radiation is infrared at wavelengths > 710 nm , thinking only in terms of visisble-light albedo is at best misleading and at worst a delusion!



intuition is not science.

You will see an ice free planet at 600ppm.


I am actually curious about the albedo question.

How much faster is the entire planet likely to warm once the arctic is seasonally ice free?

Most global warming questions are easy to find answers to, but I've had trouble finding much on this.

Ocean circulation has not changed that much
Really? The cutting of off the Panama Ismuth is considered one of the biggest changes in the past 62 million years at least.


Artful Dodger

It a fallacy to state that heat loss to the atmosphere due to prolonged open water cools a region. It does not, and can not.

Heat loss from open water can only act to return the system to the freezing state, which will always be reached later with the added heat content of warmer water. Freeze up ALWAYS takes longer with warmer water as a starting point.

Further, not all that heat is lost. Increased water vapor, CO2, and CH4 act as a thermal blanket to retain more of that heat, warming the atmosphere. This slows cooling of the open water as a positive feedback, and again extends the length of the ice free period. This is often called the green house effect (Al Gore made a movie about it). See the Hudson Bay, Jan 15, 2011 if you have doubts:


A thinner ice cover develops because of subtraction of the early freezing season, making it easier to melt and advancing the arrival of Spring. The cycle repeats without bound due to these positive feedbacks. Eventually, first freeze date == onset of melt, and the system becomes perennially ice free.

AS LONG AS WE KEEP BURNING FOSSIL FUELS... or maybe even LONGER, if mankind triggers the methane clathrate bomb before we shake ourselves awake.

Imagine you are an Egg. Your name is Humpty. You are peering over the edge of a Wall...


A box of ice full or half full is still going to keep the air around it cold with the same amount of outside forcing. It may not be as cold the further away from the ice but it's nearly the same.

A box of ice full, half full, or empty is not the same.

without the ice the entire dynamic of it is different. We see this in March here. after a snow fall it could be 37 the next day when it would have been 57.

the closer we come to little to no ice the fast the climate changes will occur.

Artful Dodger

Let's have a look at the recovery in SSTs in the Hudson Bay. The following animation compares Aug 19, 2010 to Aug 19, 2011:

Real-time, global, sea surface temperature (RTG_SST_HR) experimental analysis

Not too much cooler after last year's extended period of open water, wot? Any takers for a January 2012 freeze up?


This paper reinforces what we have already heard about an Arctic Sea tipping point. I think we mostly agree on this point.

So what points remain hanging around?

A tipping point for land based ice sheets is a big one.

Details about how arctic amplification works and how the present is a novel case of warming is another.

The impact of rapid warming vs historical events... How is this different and why does it matter?

What else?

Artful Dodger


I doubt that ANYONE her has read this paper in the 17.5 hrs since Neven posted the Science Daily article, and in the 9.5 hrs since I posted a link to the actual paper.

Which points do you suggest this paper makes that we now agree upon?

Other than rhetorical points, I mean.

Daniel Bailey

"our findings are expected to be most relevant to the assessment of sea ice thresholds under transient warming over the next few centuries in the absence of substantial land ice sheet changes."

I had a problem with this presumption. I am also not confident the warming of the currents surrounding Antarctica nor those warming currents rounding the northern tip of Greenland & entering the Fram (to name but two) were accurately represented in the study. Lastly, the 1 degree resolution is way too granular. Maslowski's next generation simulator has a much-greater eddy resolution than that & will be the de facto prediction system to use when made available.

But then, I'm not ANYONE. ;)


Artful Dodger,

Great comment! What point(s) am I alluding to?

Based on recent posts on this and related topics, I am alluding to 'tipping points'. ie. the earth is not on the verge of an irreversible condition without artic sea ice

Seke Rob

Re: Artful Dodger | August 20, 2011 at 03:58

By the sole correlation [not to take as a firm causation] with a rebuilding La Nina state, I'd say that Hudson Bay refreeze will be laaaaater. Small refreeze, then a good pile of snow on it [more vapor going around these decades], so that blanket will help to hold things below there nice and cuddly for next melt season.

Artful Dodger

Seke Rob:

I tend to agree! When I get a chance, I'll check the archive of snow cover depth over Hudbay, and see if I can put together an animation.


Kevin McKinney


My intuition initially matched yours--however, the Arctic Ocean is not like smaller, land-locked bodies of water. Lake Superior usually doesn't freeze completely over anymore, even as inland lakes well to the south of it do.

The Arctic Ocean of the future may be expected to act somewhat similarly--you'll see ice formation in sheltered waters, probably, but the warmer, more agitated central basin, with great thermal inertia will not freeze.

As to the albedo effect, which D also asked about, I saw a paper about this, and linked it here, maybe a month or two back. I'll try to find it again if I have a chance. But IIRC, the global mean forcing from an ice-free Arctic was calculated at 7 W/m2. Not huge, but not insignificant, especially since the effects won't be distributed equally.


If you wanted to cause confusion in the minds of the general public and provide political cover you would sponser this report. If you think like I do that we are going to experience a massive melt this decade -- then you would want to say oh the science indicates that the condition isn't permanent and the Ice can recover. This study is a deniers wet dream. When this fails to be enough you will start seeing studies indicating that the ice free arctic is not responsible for the weather changes being experienced. This is the second study that has been discussed here that is designed to prevent you from being absolutely sure that the ice loss is permanent. The question is not that a forever after tipping point is crossed but how much hysterisis is in the system? How much warmer will the average temperatures get solely because there is less ice to cool the north pole? How much lower than present would CO2 need to be to reverse these changes? Furthermore the current level of CO2 guarentees that temperatures are increasing in the arctic not staying the same or reverting back to decades ago levels. These studies are designed in advance to provide political cover for deniers when the arctic ice melts away. Oh well it's not like its permanent or any thing!


RuninCircles - Well put - We know how to add 1% of co2 per year - Nobody has explained how it would be possible to remove an equal amount

Chris Reynolds

Artful Dodger,

I've read the paper, thanks for the link.

For me the most important point, aside from (yet again) finding no tipping point, is:

"The results presented here illustrate a hazard of using factors
such as an increase in variance as generic ‘early-warning
signals’ of an approaching tipping point..."
and the paragraph following.

Regarding your earlier message at August 20, 2011 at 03:06: There is an observed lower troposphere heating during the early freeze season (NCEP reanalysis), and a reduction of atmospheric heat transport into the Arctic (Smedsrud 2008). Both of these are found in the Tietsche model study. That we're observing them now (in the post 2007 environment) suggests that the same process is at work and explains why despite 2007 there has been no further crash in the following 4 years. I include 2011 in that count as we're not going to see a substantial drop below the 2007 benchmark.

Chris Reynolds


Is it not possible that this paper is a scientific examination of the idea that the Arctic has a tipping point?

Is it not possible that the denier's real dream is the anticipation that: Within a few years when the Arctic doesn't undergo a rapid transition and they can use it to clobber action regards AGW by citing more alarmist overstatement?

I think those that are saying we're on the verge of a rapid transition should examine their reasons. Do they have strong arguments? Is it possible that they are misreading what is going on? What will be the consequences for effective action against AGW if they'e wrong?

I strongly doubt we're going to see a rapid (this decade) transition. If I'm wrong I can publicly declare it's a lot worse than I thought. Many real scientists will do the same. At least for a time blind eyes may turn and worry may lead to action.

If you're wrong what will you say and what impact on the pubic's perception of AGW will it have?

That the Arctic will rapidly transition this decade is a dangerous idea that should be critically examined with the evidence at hand.


If you're wrong what will you say and what impact on the pubic's perception of AGW will it have?

I don't think this will change the public's perception from what it is already. The public is generally very susceptible to whatever it is that pseudo-skeptics are feeding them, because the public has been brainwashed into its lifestyle and will only change it over its dead body. If we alarmists turn out to be wrong, it really won't change things much. Damned if you do, damned if you don't.

The last three papers on ice being 50% of what it is now during the Holocene Thermal Optimum, extent potentially increasing during periods in the coming decades, and now this one, are very interesting. But they are also excellent spinning material, as we have seen here and there on the bunkosphere.

In the grand scheme of things it doesn't change much. But it's also good material for this Arctic Sea Ice blog.

Janne Tuukkanen

No need for desperation. I firmly believe in positive thinking, and I firmly believe people are smarter than simplistic opinion polls make you think. Even I am smarter, than I usually feel :)

(mis)Quoting a late American president:

"We will do it. Not because it is easy, but because it is hard!"


Interesting to read how Chris R's perception is 'we're not going to see a substantial drop below the 2007 benchmark'.
I think that benchmark is 'history'. There is so much less ice in the Arctic now in terms of volume. It's the physical form of how the ice is vanishing and our arbitrary limit of counting (15% per grid), that is blurring us to see what's really ging on...


I don't want to have an argument about what is and what isn't science. But I will let you answer your own question about loss of sea ice. Ask yourself how thick is the current ice. How many centimeters does the ice go down every year. The trend is clear the answer hasn't changed since 2005 when Dr. Maslowski first drew the line on chart.

Chris Reynolds


I'm not sure what the situation is over your side of the Atlantic. But here in the UK of the 3 other engineers I work most closely with: one doesn't express an opinion, the other two are convinced and very worried about AGW. All of my mates are convinced about AGW. I know a couple of people who doubt it, but their position is so poorly considered that they've given up arguing as when they get into arguments they lose. I don't consider them mates because they're jerks (and not just on this issue). From what I see doubt over the reality of AGW is in the process of being manoeuvered into the same place as other consipracy theories; a minority position espoused by whackos and pub-bores. From what I read, over here in Europe the situation is ripe for a major AGW disaster to drive public demand for action.

Obviously I was referring to extent, the main measure and the one based on direct observation, although an areal crash is likewise extremely improbable at this stage. Since reading the 2011 Schweiger paper I think PIOMAS is probably on the ball. Yet PIOMAS doesn't show an ice free September by 2035, more here. So if you're prepared to accept current volume trends from PIOMAS as showing an imminent crash, why does that very source you use not support your contention?

Is it possible that the physics models that serve PIOMAS so well are telling us not to interpret the drop in volume as a prelude to a total crash? If I can ammend what Armour et al (the study Neven's posting refers to) state:
"The results presented here illustrate a hazard of using factors such as an increase in variance [or rapid decrease in volume] as generic ‘early-warning signals’ of an approaching tipping point..."

Chris Reynolds


You may not have read the posts on my blog about sea-ice thickness, here and here. I am all too aware of thickness and volume trends, when forming an opinion I concentrate on the evidence that stands against that opinion. This is why I first posted In Flux, before going over why I don't think we face an imminent transition, here.

Wang and Overland used models that correctly reproduce the size of the seasonal cycle to revise projections of when the Arctic will be sea-ice free. The point is that the seasonal cycle is driven by an external forcing agent - changes in sunlight. So Wang & Overland argue that these models should correctly reproduce the sea-ice response to the external forcing of CO2. They state that when extent is around 4.6M kmsq there is a period of rapid loss, we're around that period now, but the models still show that the earliest we can expect a seasonally sea-ice free state is around 2030. One of the models they use is MIROC, Watanabe et al find that MIROC accurately reproduces the Arctic Dipole - a major factor in the accelerating loss rate (I'll be blogging on this later today).

I know that Maslowski has expressed concern about model resolution. I've read several of his papers. But publication wise - he's patchy and sporadic on this matter. I haven't read any recent papers from him that detail his findings w.r.t. an imminent sea-ice free Arctic. Serreze said he is very intelligent and his views should be considered, I do not disagree. However my attitude to him is that I've placed his work on the back-burner until he publishes - until that I have no way to assess his work and what it may mean for the future (soundbites in the press are useless).


Rather than point to misc studies how thick do YOU think the ice is now? What do you think is the thickness decline rate per year?

Ian Allen

Let us remind ourselves of the physics of freezing water.
Water is a highly unusual substance by allowing surface freezing rather than bottom-up due to ice being lighter than water.
Freshwater has max. density at 4C thus fresh lakes only need the water column to be below 4C for ice to form via surface chilling.
above about 20 ppt salinity this breaks down, and water sinks until freezing point, thus open-sea-water needs a subzero substratum. different salinities at depth complicate the picture, which is why the worlds ocean currents are termed the thermohaline circulation, bringing 2C antarctic bottom water on a world tour.
The Baltic sea, and especially Gulf of Bothnia is less salty than much processed food and soup and so is effectively fresh as regards cold water floating phenomenon (and fresh enough for most folks to survive on)

Considering the extremely shallow seas N of Siberia, if the whole e.g. 20m of the water column warms to 2C above freezing, this excess heat would all be needed to bottom-melt 50cm of ice- 80C latent heat/2*20
there is a lot of fresh water pumping in here as well.
However, for the deep parts of the basin, once the whole column warms it is not coming back in a hurry- the main reason it's cold is due to the fragile mirror of ice. Once it goes, the poles of cold will migrate to land and ameliorate as N America and Asia become separated by a warm sea. storms will also roam around stirring the sea as we are already seeing around the edges in summer.

We never ever see ice in the open sea W of UK or Norway because the sea keeps us warm despite pathetic winter insolation, because the seas are salty and not extremely shallow.

Chris Reynolds


When I point to miscellaneous studies I'm not doing so to intimidate or to confuse. I refer to them because that's where I got the info and if you want to know full titles, even links to paywall free copies, I can give them and you can check I'm not talking rubbish. I encourage scepticism (but I loathe denialists).

At the 2010 minima PIOMAS volume & IARC/JAXA extent implied an average thickness of around 0.9m. At present (as of 31/7/11*) it's around 1m.

As for the rate of decline: In 2003 & 2004 the thickness at minima was a little under 1.4 metres. 2007 was a notable step drop, as was 2010.

In 2007, 2010 and now 2011 the biggest anomaly drop was in April to June.

When/if I get permission from IARC/JAXA I will blog on what PIOMAS shows in more detail. If I don't get permission I'll blog anyway, but regretably won't be able to include thickness.

*PIOMAS only update monthly so that's as far as I've got.

The comments to this entry are closed.