Melting of the Arctic sea ice

Thank you Climate Dialogue staff, for starting this venue with a discussion on Arctic sea ice developments.

And thank you Dr. Meier, Curry and Lindsey for providing a great overview of scientific findings attempting to explain the sharp (unprecedented in at least a few thousand years) reduction in Arctic sea ice cover from the past 4 decades.

Since the Arctic is a major regulator of Northern Hemispheric weather patterns, our understanding of the process is extremely important for future estimates of Northern Hemisphere weather projections. In that regard, the overview given by Dr. Meier and Dr. Lindsey, sustained with references to scientific publications, are very helpful.

In the discussion of “causes” of Arctic sea ice decline, it seems important to note that “natural variability” is not an explanation for a trend, since it can goes TWO ways. The best we could say is that such-and-such part of the decline COULD be caused by natural variability if natural variability is currently “negative”.

Here, it is important to note that Arctic sea ice extent dives way below “natural” variability, and there is no “recovery” in sight :
http://neven1.typepad.com/.a/6a0133f03a1e37970b017744cf5360970d-pi

So, I think that before we start speculating about how much of Arctic sea ice decline is “natural” and how much is “anthropogenic”, we should FIRST understand why GCMs are underestimating Arctic sea ice decline even if we add both these factors in the models !

Here, I noticed that GCM projections are lagging some 20 years (CMIP5) to 40 years (CMIP3) behind actual Arctic sea ice decline. Apparently, we are currently experiencing Arctic “climate” conditions that were not expected until another 40 – 80 ppm increase in CO2 concentration (which implies some 0.5 – 1.0 W/m^2 forcing).

So, could it be that there is an additional heat source that is not correctly modeled, which causes the models to ‘delay’ the response to anthropogenic forcing ?

Derksen et al 2012 noticed that snow cover is quite significantly underestimated in CMIP5 simulations :
http://www.nature.com/news/arctic-snow-cover-shows-sharp-decline-1.11709
We are talking about very significant reductions in snow cover, which just set a record of a whopping 6 million km^2 snow cover anomaly in June 2012 :
http://climate.rutgers.edu/snowcover/chart_anom.php?ui_set=1&ui_region=nhland&ui_month=6

What kind of forcing would such a large anomaly in snow cover create ?
Tamino did a first estimate using observational data and a simple model for insolation, and assessed that the albedo changes for NH ice and snow over the past couple of decades causes an extra 1150 TW to be absorbed by our planet :
http://tamino.wordpress.com/2012/10/08/snowice-by-request/
1150 TW is some 0.9 W/m^2 if spread out over the entire Northern Hemisphere.
If this albedo-induced heat is kept closer to the Arctic, that number goes up even more.

This is certainly in the ballpark of the forcing mentioned above, and thus can quite plausibly explain why Arctic sea ice decline is 20-40 year ahead of GCM estimates.

If underestimate of high-summer snow cover decline is the cause of GCMs underestimate of Arctic sea ice decline, then the search for “causes” of Arctic sea ice decline would shift to finding causes for the rapid decline in Northern Hemisphere summer snow cover.

Which would take the scientific discussion to a slightly different area of the planet, with slightly different physics and different implications about the future….

I have not seen any comments about Antarctic sea ice extent. My reading of the data is that Arctic and Antarctic sea ice extent are negatively correrlated. This does not indicate that CAGW plays any role in the decrease of Arctic sea ice extent that has been observed in recent years. Rather it indicates that clouds may play a major role. The albedo of clouds is greater than the surface everywhere on earth except Antarctica, where the albedo of the ice is greater that that of clouds. So more clouds cause cooling everywhere else except Antarctica, where they cause warming; and vice versa. A change in cloud cover is more indicative of what is happening to total sea ice extent.

A note to those who run this blog. It seems to me that discussing Arctic sea ice extent in isolation is a very biased way of starting discussions on this blog. If you want to appear to be unbiased, then you ought to have included Antarctic sea ice extent in the intiial discussion subject.

Hi, good to see the discussion has started (dialogue to me sounds like there are just two sides, which would be an oversimplification!) May I ask how these three scientists for the first discussion were selected? Are they chosen to represent the full range of opinions in climate science about this topic? I ask because I have the feeling the scientists who are most concerned about the Arctic are missing. For example, I would like to hear what Peter Wadhams (Cambridge)has to say about this. Or Wieslaw Maslowski, an oceanographer who made a model in 2007 that predicted an ice-free Arctic in the summer of 2013.

Jim Cripwell, before discrediting this blog could you please elaborate on your theory that the observed changes in Antarctic and Arctic sea ice extent/volume are caused by changes in cloud cover? I don’t see why the albedo of clouds above Arctic sea ice is larger than above Antarctic sea ice, I’d also like to see some studies/observations of that.) Preferably I would like to see some observations and studies backing up your conclusions.
I have seen studies and observations backing up the theory you seem to call CAGW (?, though it is not clear what is your definition of that?). I have also seen studies describing why we expect to see large sea ice changes in the Arctic (see the discussion above), and not in the Antarctic and possible reasons why we observe slight increases in Antarctic sea ice (e.g. see (very recent) http://www.the-cryosphere-discuss.net/6/931/2012/tcd-6-931-2012.pdf , refs and especially discussion therein). (FYI: sea ice changes in the Arctic are 4x larger than in the Antarctic, changes in the latter are barely significant; global sea ice trend is strongly negative.)

To Marcel Crok.

It is clear that you do not read Judith’s Climate Etc. She routinely chooses topics for discussion which assume CAGW is correct. Surely it is not up to me to choose experts who disagree with CAGW; that is your job. I merely point out that the three who you chose just happen to be proponents of CAGW. I hope that this is not deliberate.

I seriously doubt that there are not people who have published on Arctic and Antarctic sea ice, and who are climate skeptics, but who they are I dont know. I suggest you contact people like Anthony Watts, Richard Courtney, Christopher Monkton, Richard Lindzen, or thousands of others who are well known climate skeptics, who would be far more likely than myself to know as to who would be suitable candidates. I just have doubts that you have searched very diligently.

Please note, I am a retired scientist who has no access to what is going on in the scientific world, except what I can glean from the Internet. Surely you have far better access as to who should be chosen as experts than I do. But if you want this blog to be impartial, I suggest you dont introduce a topic unless you can guarantee that both sides of the discussion are represented as experts.

I would like to say I appreciate the efforts by Drs. Meier, Curry and Lindsay to participate in this dialogue. All three of the guest blogs are really informative. Also, I happen to disagree with the comments made by Jim Cripwell concerning the choice of experts: all three are excellent. Although I sometimes disagree with the comments made by Dr. Curry on her blog, she is certainly one of the most rational, influential and well-informed of the ‘skeptic’ voices in the blogosphere.

Judith Curry rightly asks: “What percentage of the recent decline would you attribute to anthropogenic greenhouse gases?”

We ought to keep in mind that anthropogenic forcings are the sum of (at least) greenhouse gases, ozone, direct and indirect effects of aerosols and also soot (‘black carbon’), if you want to mention that separately from aerosols:

http://www.ipcc.ch/publications_and_data/ar4/wg1/en/fig/figure-ts-5-l.png

And I guess we can forget about Urban Heat Islands in the Arctic. 😉

I would suggest the proper answer to Dr. Curry’s question might well be: “> 100 per cent”, because anywhere from 20% to 70% of the effect of extra greenhouse gases may have been masked by extra aerosols (as you can see in the IPCC radiative forcings table, linked above).

A separate reason that the attribution of this decline to extra GHG’s may actually be “> 100%”, is that naturally the insolation at higher Northern latitudes is now (very slowly) decreasing – because of orbital forcings (Berger and Loutre 1992). Since we are very slowly nearing the later part of the current interglacial, one would actually expect a trend towards an *increase* in Arctic sea ice.

I would appreciate it very much if all three experts would briefly comment on these considerations.

I agree with both Judith and Ron that there is still fairly high quantitative uncertainty in terms of the amount of influence of greenhouse gases. But I think we’re all in agreement that GHGs are playing a significant, but not exclusive, role in the long-term decline of Arctic sea ice with the magnitude of the decline is enhanced to some degree by natural variability. What the balance is between these two influences is still a largely open question.

The 50-70% range for GHGs that Judith mentions is probably a reasonable spread in capturing the potential range. I would lean more toward the high side of that range because the I don’t see the AMO and PDO having a large magnitude influence on summer ice. The AO does have a larger influence, but that has largely been lost in recent years.

One final comment on Judith’s about the data. While our most complete dataset, the one we have the highest confidence in, is the passive microwave record, there is fairly complete coverage from operational ice charts back to at least the mid-1950s. And there are Russian ice charts for the Eurasian Arctic back to the early 1930s. Though not complete, these do extend the record and I think provide some sense of the interannual and decadal natural variability of the ice. There are indications of lower ice in the 1930s in the Russian Arctic [Reference 4 in my post], suggesting the influence (AMO?) of a multi-decadal cycle, at least in the Russian Arctic. But the data show a different character in terms of the seasonality and regionality of the lower ice conditions compared to the recent decline.

Regarding the melting of Arctic Ocean ice. Undersea volcanism has increased worldwide and it includes the Arctic Ocean. In 1999 and explosive volcano erupted under the Arctic Ocean with such force it left a 10 sq km debris field. The cause of that explosive volcano was a CO2 buildup in the crust under the ocean, according to the Woods Hole Observatory. They also stated it released 10X more CO 2 than originally thought. It continues to erupt along a vent line, heating the water to 752 degrees Fahrenheit.

Dear editors and advisory board,
A very good initiative to facilitate a professional scientific discussion about the very interesting subject: the climate of the world we all live in. I find it also laudable to give the interested public an opportunity to react or ask questions.
Thanks very much.

I see there is some kind of agreement between the experts on the attribution of the decline in the Arctic sea ice to greenhouse gases (50-70%). Dr. Meier and Dr. Lindsay state that without the long-term warming trend of GHG’s this decline can’t be explained, with some references to research indicating that this decline appears to be unique in at least the last 5000 years.
Does Prof. Curry agree with this statement?

The natural variability could even cause the Arctic sea ice to grow again for some decades (Kay 2011), but due to the seemingly unstoppable trend in growing GHG concentrations in the atmosphere the long term trend will be one of less and less ice in the Arctic. A unique ecological habitat could be lost then.
According to Yamamoto et al 2012 the rate of ocean acidification (dropping CO3(2-) concentration) in the Arctic is not only influenced by the rising of the atmospheric CO2 concentration but also by the loss of sea ice. This also has a negative influence on ecological life in the Arctic Ocean.
What’s the opinion of the experts on this matter?

Sorry, here’s another try for a better layout:

I agree that it would’ve been nice if Dr Peter Wadhams of the University of Cambridge had been invited to give his views. This initiative was announced in a Dutch news paper as a negotiation between alarmist scientists and skeptics. Dr Meier and Dr Lindsay are not in the least alarmist. Wadhams could be labeled as such, despite the fact that he has been largely right about the unfolding of events in the Arctic so far. An unfolding which, of course, is alarming.

It’s good to see though that all expert participants agree on an anthropogenic factor of at least 50%. Given the magnitude of the event and the rate at which it is progressing, one then inevitably moves on from ‘could there possibly be a problem ever?’ to ‘what can and should be done about this anthropogenic factor?’. It seems that after one day ClimateDialogue has been so successful in reaching agreement concerning the reality and risks of AGW (which is clearly reflected in the Arctic situation) that it has made the dialogue about all other scientific aspects of climate change irrelevant. Well done, chaps.

Now, of course, many will say: But is there a problem related to the disappearing of Arctic sea ice? I personally think there very well could be several problems and have co-written an article about it for the Arctic Sea Ice blog, called Why Arctic sea ice shouldn’t leave anyone cold. Maybe Drs Curry, Lindsay and Meier could expand some more on potential consequences of disappearing Arctic sea ice? In my view saying “we don’t know” isn’t really comforting. Rather the opposite, with, again, the magnitude of the event and its rate of progress in mind. We absolutely need to know this, right? This is the crux of the matter.

So some more please on Arctic Amplification, Northern Hemisphere snow fall patterns (positive anomaly in winter, very negative anomaly in spring and summer) and the influence on atmospheric patterns. Permafrost and methane clathrates are less relevant at this point.

It’s good to see though that all expert participants agree on an anthropogenic factor of at least 50%. Given the magnitude of the event and the rate at which it is progressing, one then inevitably moves on from ‘could there possibly be a problem ever?’ to ‘what can and should be done about this anthropogenic factor?’. It seems that after one day ClimateDialogue has been so successful in reaching agreement concerning the reality and risks of AGW (which is clearly reflected in the Arctic situation) that it has made the dialogue about all other scientific aspects of climate change irrelevant. Well done, chaps.

So some more please on Arctic Amplification, Northern Hemisphere snow fall patterns (positive anomaly in winter, very negative anomaly in spring and summer) and the influence on atmospheric patterns. Permafrost and methane clathrates are less relevant at this point.

Judith writes: “So . . . what is the bottom line on the attribution of the recent sea ice melt? My assessment is that it is likely (>66% likelihood) that there is 50-50 split between natural variability and anthropogenic forcing, with +/-20% range.”

That is, there is at least a 2 in 3 chance that the contribution to sea ice melt will be in the range of 30nv-70af to 70nv-30af. And perhaps a 1 in 3 chance that it will be outside that range. In short, anthropogenic forcing could be a very minor element (say 20%) or a very major element (say 80%). We don’t know; as Ron Lindsay says, it “is rather uncertain.”

The bigger question is: does it matter? Is the extent of Arctic sea ice a great concern? Should we fear or celebrate a reduction? If the latter, then the question of attribution is of only academic interest; we should be pleased if human emissions are accelerating ice loss.

Judith Curry writes in her guest blog:
“Focusing on CO2 as the dominant influence on the time scale of two decades seems very misguided to me.”

And:
“The first issue to debunk is that an ‘ice free’ Arctic is some sort of ‘tipping point.’ A number of recent studies find that in models, the loss of summer sea ice cover is highly reversible.
The impact of September sea ice loss on the ice albedo feedback mechanism is interesting. The minimum sea ice occurs during a period when the sun is at low elevation, so the direct ice albedo effect isn’t all that large. Less sea ice in autumn means more snowfall on the continents, which can have a larger impact on albedo. The impacts of the freeze-thaw over the annual cycle influences ocean circulations. But sea ice would continue to freeze and thaw on an annual cycle.”

And:
“Would melting sea ice trigger some sort of clathrate methane release into the atmosphere? Well in terms of thawing permafrost, it seems like more snow fall on the continents would inhibit permafrost thawing. Same for the stability of the Greenland ice cap.
These are all qualitative speculations, but I am not seeing a big rationale for climate catastrophe if the see ice melts.”

This would all be quite comforting if it would make sense. But does it? Not according to the arguments of scientists like Jim Hansen.

What’s the influence of CO2 on the time scale of two decades? Over the past two decades CO2-concentrations increased by about 36 ppm (from about 356 to about 392 ppm, almost 2 ppm/year on average). Over the past two centuries CO2 increased by more than 100 ppm from about 280 ppm to the current 392 ppm, so more than 0.5 ppm/year on average. The current growth rate is more than 2 ppm/year:
http://www.esrl.noaa.gov/gmd/ccgg/trends/

Jim Hansen gives some useful context to these data in this article with Makiko Sato:
http://arxiv.org/ftp/arxiv/papers/1105/1105.0968.pdf

The natural rate of CO2 change during deglaciations is about 100 ppm in 10.000 years (from about 190 to about 290 ppm), so on average about 1 ppm per century. At the current growth rate, which shows no sign of slowing yet, we can expect to reach 560 ppm, a doubling from preindustrial, before the end of this century with an average rate of increase of about 1 ppm/year over three centuries. That would bring us back to a CO2 level that earth has not seen in probably more than 40 million years, so before the start of the glaciation of Antarctica, with an average speed over three centuries that is about 100 times faster than during recent deglaciations.

This is the focus of Jim Hansen, and that seems very reasonable to me. I don’t understand why Judith Curry seems to think this focus is ‘misguided’.

Hansen thinks the loss of Arctic sea ice is a potential ’tipping point’ because of several potential positive feedbacks, such as what he calls the ‘albedo flip’. Loss of albedo does not start at ice area minimum, but much earlier in the season when insolation in the Arctic is at its peak. That makes the ice albedo effect potentially much larger than Curry seems to think.

More snow fall on the continents in autumn/winter doesn’t automatically mean more snow and albedo in spring. To the contrary, as Rob Dekker points out, studies show that there seems to be less and less snow in spring on the continents, so less albedo, faster warming and more melting of permafrost.

This also affects warming and melting on Greenland, just as the warming ocean water does, with the risk of fast and large sea level rise. Why does Judith Curry seem to take this risk not very seriously?

Melting of permafrost and clathrates would be another positive feedback by releasing CO2 and methane, although there seems to be quite some discussion about the potential speed/time scale of that feedback. If however Arctic sea ice melts away much faster that thus far expected, this feedback could also kick in much faster than expected.

I don’t see why we should ignore these quite serious risks, as Judith Curry seems to advise.

There has been a bit of recent research that indicates that the shorter term climate oscillations like the PDO and AMO may have much longer term oscillations or be parts of other longer term ocean oscillations related to Thermohaline circulation. Tuggweilder et al for example indicate that variation of surface wind velocity in the Antarctic convergence can have significant, (10 to 30 Sverdrup) impacts on the North Atlantic AMOC on time scales of ~150 years. Since there has been a shift in the diurnal temperature range trend starting in ~1985, where Icelandic diurnal temperature ranges are increasing, would not this be indication of stronger natural internal variability impact?

Arjan, you are trying to make my remarks mean far more than I intended. I must go back to Square 1. There are various blogs which discuss CAGW. To take but 3 examples, there is RealClimate, which is blatantly warmist, and WUWT, which is equally blatantly skeptic. In between there is Climate Etc. which is definitely warmist, but far more neutral than the other two blogs. When I heard of Climate Dialogue from the Global Warming Policy Foundation, it seemed that this new blog would be somewhere between Climate Etc. and WUWT.

So when I came here intially, I was deeply disappointed to find a blog that, at first blush, is just an unmoderated version of RealClimate; a blatantly pro-warmist blog. The first subject was Arctic sea ice, heavily biased to the warmist side, with no experts from the skeptic side. What I tried to do was point out how biased this first article was, in the hopes that Marcel might change his approach. In support of this I tried to suggest that it would be far better to include a discussion of Antarctic sea ice, and see if there was an reason, other than CAGW, which explains the differences between the two poles.
Now I am fully aware that the warmists have all sorts of excuses, some published in the peer reviewed literature, that try and dismiss Antarctic sea ice, so that they only need discuss Arctic sea ice, and show that it really does prove that CAGW is real. All I was attempting to do is to suggest that there is a hypothesis which tries to explain the difference between the two poles, and which does not involve CAGW. Has this hypothesis been fully tersted? Is there lots of data etc to show that it might be correct? No, of course not. It is merely an idea which tries to deal with both poles at the same time; something this first artcile on the blog fails miserabely to do.

I sincerely hope that in the future, Marcel will take this blog so that it truly lies between Climate Etc and WUWT. Should this fail to happen, I wont be around at all.

As per so many of the above comments, good initiative.
It is fitting for the dutch to start with it, and that is not meant sarcastic but a compliment.
Throughout the ages the dutch have always been the more likely nation to accept that there are two, or more, sides to every issue.

It is clear from all of the above that other then “belief” in computer models and predictions no one has got any real clue as to what is going on.
What is also clear, like it or not, is that similar ice disappearing discussions were in the news in the 1930’s with similar predictions of doom.
We can of course go back 1000 years and ask the Vikings for advice on how to farm on Greenland. Or even further back and ask Hannibal about crossing the Alps with elephants, that path is blocked by ice and snow at the moment.

What would also appear obvious is that nothing that is happening at the moment is new or unprecedented except in the mind of some and those that wish to follow every prediction of doom.

We are not seeing any weather event that is outside natural limits or has not happened in similar fashion in the last 100 years. We have just put more people in harms way which is just about the only anthropogenic part in the whole issue.
Simplistic reasoning? Perhaps.
But so is the notion that an increase in CO2 is the cause of a slight increase in temperature after a 30 year cooling period. Whereas the CO2 has been increasing supposedly since the mid 19th century and has by now increased about 33%, assuming that the starting point is correct.
There has been no increase, some would say there has been a decline, in temperature for the last 14 years. Which by itself flies in the face of the theory.
The sea level is not rising any faster then it has done on average for the last 100 years.
Etc.
I know they are all the old arguments from a non AGW believer.
But then I like to live with facts rather then models which, the 3 experts here agree to, are not accurate.
The debate here would appear to be about how inaccurate the models are and how can we defend the predictions.

History and facts are clearly not on the side of the models.
Life is looking up and if it is warming in the long term rather then part of a cycle it will, overall, be good for the human race.
Our species has always done better in times of warmth. Darn history again.

No problem with finding long term energy solutions which do not require carbon based fuels on the basis that logic would indicate that within the next 150 years we will have used most of it and we will be forced to rethink those energy sources.
But we are certainly waisting excessive mental energy and public funding on CO2 is causing unprecedented warming models and trying to explain every gust of wind with that model and classify it as “predicted”.

Let’s put the mental energy and public funds into something useful for the future.

I would like to understand Judith Curry’s confidence in attributing the remarkably low levels Arctic ice plunged to this year to the brief Arctic cyclonic storm in August. Cyclonic storms in the Arctic are rare but not unknown. Why hasn’t such a storm caused this sort of thing before? If this storm was in some ways unique, would not climate change be expected to be itself causative in the formation of such a storm, so it cannot be referred to as natural variability in the first place?

NSIDC reported on the storm here: http://nsidc.org/arcticseaicenews/2012/08/a-summer-storm-in-the-arctic/ including the interesting comment – “Summers dominated by low pressure systems over the central Arctic Ocean tend to end up with greater ice extent than summers dominated by high pressure systems.” Why did that not happen this time? Attribution without a physical causative mechanism doesn’t sound like scientific explanation to me…

Climate science cannot have it both ways. Either anthropogenic warming (AGW) exists and affects all of our weather, or AGW has not occurred and therefore affects none of our weather. It is an all or nothing proposition. The trend of temperatures tells us that AGW is real, and therefore affects the entire weather system, all the time.

AGW has been affecting all of the weather on Earth as long as AGW has been perceptible. We were ready for variations in the old, lower energy weather system. The early effects of small amounts of extra CO2 were tiny, and, we just did not notice that the weather was intrinsically different. We thought the pattern of warming was just natural variation or another natural cycle of warming – the natural variation of a system in equilibrium. Now, greenhouse gas concentrations have increased, the system is not in equilibrium, and AGW is much greater. We are not ready for the more powerful weather system that results from the extra energy from AGW. It is time to get “real” and tell the truth.

The truth is that energy from AGW is a part of the heat content of all phases of our weather system. That includes the 2012 weather process that produced “Sandy”. Sandy, without the energy from AGW, would not have been “Sandy”. AGW was a part of the 2010 Russian Heat Wave and the Pakistan Floods. AGW was a part of the 2011 Texas Drought. Without the extra heat from AGW, these would all have been very different events – “normal weather”. Likewise, the great snow events were internal work done by the global weather system, driven to greater intensity, by the extra energy from AGW.

AGW provided the extra heat that reduced Arctic Sea Ice extent. This is a special set of weather processes because Arctic Sea Ice extent affects all other northern hemisphere weather. Loss of sea ice leverages the impact of AGW on the global weather system. Anyone who doubts this needs to go back and think about the thermodynamics of weather.

AGW is extra heat that humans and Earth’s ecosystems are not adapted to withstand. AGW is that straw, which when added to a load of straw, breaks the camel’s back.

This post claims that there is a “trend” in Arctic sea ice extent. A trend, in order to exist, should be demonstrated to be statistically significant. There has been no such demonstration, by anyone. Moreover, I do not believe that there could be such a demonstration, given the shortness of the time series.

In other words, the claimed “trend” might be reasonably ascribed to random variation. As an analogy, if we flip a coin four times and get the same side each time, that does not imply that the coin is biased: getting four in a row can reasonably be ascribed to random variation (the probability of getting four in a row is 1/8).

The techniques for evaluating significance that are taught in introductory statistics courses are based on assumptions that do not hold for the Arctic ice (e.g. the assumption that there are no autocorrelations). Thus, more specialized techniques need to be used.

Rob Decker says, “But we better hurry up fixing our models, since the trend is clearly accelerating…”

Knowing how wrong is the basic climate energy equation, we sure have to hurry up in fixing our models-The potential energy of the atmosphere is missing from the equation, and there cannot be any worst than that. Bureaucrats have to hold onto the climate regulations until the climate models are fixed and print something that makes sense. Otherwise, most of us will be poor and poverty is the worst enemy to the climate.

Dr. Meier wrote:

Our expectations for how the ice responds to natural variability is based upon a thicker ice cover, which may no longer be valid.

Bingo…

Schuur you write “Your suggestion to invite people like Monckton or Courtney holds no ground at all.”

I do wish people would read what I actually write, instead of assuming that I wrote something they would have liked me to have written. What I said was “I suggest you contact people like Anthony Watts, Richard Courtney, Christopher Monkton, Richard Lindzen, or thousands of others who are well known climate skeptics, who would be far more likely than myself to know as to who would be suitable candidates.”

In other words, I suggested that Marcel consulted some of the real skeptics of CAGW to find out who the experts on sea ice are, and who believe CAGW is a hoax. Not pseudo-skeptics like Judith Curry. Maybe you want me to look foolish by suggesting that the people I mentioned wrote artciles on sea ice, but that was not what I said.

If Marcel really wants a dialogue from both sides of the debate, which I am beginning to doubt, then he must actively solicit the views of people who, like myself, strongly believe that the true physics shows conclusively that adding CO2 ot the atmopshere has absolutely no bad effects whatsoever; and certainly has a negligible effect on raising global temperatures. CO2 is a plant food, and the more we have of it in the atmosphere the better. It will help us feed an ever growing world population.

I believe one of the future subjects to be discussed will be climate sensitivity. If I am right, then let me suggest here and now, that he invite Terry Oldberg and Tomas Milancovic to write on the subject.

Bert said :

alt Meier concludes that ‘the current decline appears to be unique in at least the last 5000 years’. But in the report ‘Past Climate Variability and Change in the Arctic and at High Latitudes’ (US-CCSP, 2009) I read that, somewhere between 750 and 1500 AD, ‘the Northwest Passage was open, at least during some summers’.

It took me a bit of time to find that report, but here it is :
http://downloads.climatescience.gov/sap/sap1-2/sap1-2-final-report-all.pdf
Regarding your claims, take a look at the evidence :

Figure 6.8 shows the bowhead whale bone findings on which this statement was based.

They found 5 specimens of bowhead whale bones on the shore of the Beaufort sea that dated around 1000 AD.
There was no significant increase in bowhead whale bones found in the Central Canadian Archipelago over the 750-1500 AD period, nor any significant increase on the East side of the Canadian Archipelago.

So, maybe the Beaufort sea was not completely frozen over in summers during that time, but I think that the authors do not present enough evidence to claim that the Northwest passage was open during some summers over that period.

What do you think ?

So… all three experts think natural variability has contributed substantially to recent sea ice loss in the Arctic. But they can’t point to a significant source of this variation from outside the system (the only candidate, the sun, fails to meet the criteria). So it’s internal variability we’re talking about. This implies that this variability is not part of climate change, just part of the present climate. And that in turn means the effect should reverse in the not too distant future, adding extra ice. Indeed all three seem to think that 2012 ice cover was an exception, which will be followed by years with more ice.

I’m not an expert, but I am confident that they are wrong. That’s because sea ice volume in the Arctic has steadily declined since 1979. A virtually ice-free Arctic seems inevitable before the end of 2015, but could happen next year. My question to Walt Meier, Judith Curry and Ron Lindsay would be: if this happens, will it change your view about the proportion of ice loss that can be contributed to anthropogenic greenhousgases? Especially Walt Meier, since you call total ice loss in 2015 highly unlikely. If it happens, that will surely change your view?

I have been asked by the moderator to comment on James Annan’s comment

“Curry’s “attribution” statement implies that she believes the forced component to be about half of the recent trend, with natural contributing an equal amount.

Under the reasonable assumption that the forced trend will stay roughly constant, and that natural contribution can be either sign, she is expecting the future observed trend to be half the historical one, albeit with substantial uncertainty.

I think this is obviously nonsensical, it is certainly contrary to all published research in this area, and call upon her to either withdraw her statement, or clarify how I have misunderstood it (and to be precise, let’s hear her prediction for future trend in Arctic sea ice), or
Justify it with reference to quantitative analysis.”

I find this comment and inferences to be incomprehensible. As a challenge from a climate researcher (rather than an anonymous commenter), I find it to be very poorly argued and of a tone that does not promote reasoned dialogue. But since I have been asked by the moderator to respond, I will.

I refer James Annan to read my published papers on Arctic sea ice and climate dynamics at the link provided in my biosketch, and also my 3 more extensive blog posts on this topic.

The relatively high attribution to AGW comes from climate models, which have substantial problems in simulating the Arctic climate, among other issues that I have discussed at length elsewhere. Further, none of these future projections consider alternative scenarios of solar and volcanic forcing. Not to mention that these climate models underestimate natural internal variability on multidecadal timescales. So exactly why should I believe climate model simulations as a basis for future projections of sea ice extent? I note that my group has just submitted a paper for publication analyzing the CMIP5 simulations of arctic sea ice. I am frankly trying to figure out how these models manage to produce any kind of sensible sea ice given that most of these models are biased cold (many are biased cold by 2C or more). Our old friend ‘model calibration’ I assume, whereby 5 wrongs might make a ‘right.’

My point is that I DONT KNOW with any high level of confidence what the sea ice will look like in 10 years or 100 years. If all other things remain the same as the period 1980-2000, plus additional greenhouse forcing, then yes the ice extent will shrink. Do I think that all other things will remain the same as the period 1980-2000? almost certainly not: we have seen a shift to the cool phase of the PDO and most projections are for reduced solar forcing. Further, the response of the climate is not linear with external forcing, so making projections on timescales less than 50 years solely on the basis of greenhouse forcing is unjustified

Go back and read my paper Reasoning About Climate Uncertainty. http://www.climateaccess.org/sites/default/files/Curry_Reasoning%20about%20climate%20uncertainty.pdf
I know you have read it once, please read it again.

I know you are also not a fan of the Italian Flag. But the error in reasoning that you are making is to assume that a collection of publications that find a large contribution of AGW to sea ice melting means that this must be correct and that scientists should accept it as such. There is a great deal of ignorance on this subject: inadequate data prior to 1979, imperfect models, and incomplete understanding of the complex radiation and climate dynamics environment of the Arctic.

As an example of the latter, in Peter Wadhams Scientific American article, he suggests geoengineering to modify the arctic clouds to keep the ice cool during summer. A very scary thing to do. If these were subtropical stratocumulus clouds over the ocean, then we have a pretty good idea that seeding the clouds would decrease the solar radiation reaching the ground. The situation is very different in the arctic, largely owing to the low sun elevation and the high surface albedo. In fact, for all but about ~2 months of the year, more liquid in the clouds would actually act to warm the sea ice surface. For background, I have published a number of papers on this issue, here are a few:
http://curry.eas.gatech.edu/currydoc/Curry_JC5.pdf
http://curry.eas.gatech.edu/currydoc/Benner_JGR106.pdf

My point is that I feel that I have a fairly deep understanding of the dynamics and thermodynamics of sea ice, the radiative environment of the Arctic, and the regional climate dynamics of the Arctic and how it interacts with the global climate. The depth of my understanding is relative to other people; in terms of absolute understanding I acknowledge a great deal of uncertainty and ignorance (i.e. the white part of the italian flag).

I offer this section from Timothy Casey’s article on volcanic CO2.

http://carbon-budget.geologist-1011.net/

“Irrespective that some authors may neglect to allow for significant volcanogenic CO2 input to the atmosphere, volcanoes represent an enormous CO2 source that is mostly submarine. Furthermore, volcanic activity beneath both ice caps and localized to the regions of most intense melting has demonstrated an obvious cause of stronger Spring melts at the Poles. It is evident from the observations of Sohn et al. (2008) & Reves-Sohn et al. (2008) that the Northwest Passage was opened up by powerful volcanic activity under the Arctic Ice along the Gakkel Ridge, while West Antarctic melting (as opposed to thickening of ice throughout the rest of Antarctica) can be explained by recent volcanic activity beneath the ice (Corr & Vaughan, 2008). Moreover, there are simply too many volcanoes to deny that the atmospheric concentration of the most erupted gas next to water is predominantly controlled by the balance or lack thereof between volcanic activity and photosynthesis. Furthermore, there is simply no established volcanic CO2 fingerprint by which we may distinguish atmospheric proportions of anthropogenic and volcanogenic contributions. This leaves us with no empirical method by which we may attribute the 20th century rise in CO2 to human energy consumption.”

Dr. Curry said

Recent published estimates of attribution of sea ice melt range from 5-40%. These estimates do not include solar cooling or account for the shift to the cool phase of the PDO and a like shift to the cool phase of the AMO. Since all of these lean towards cooling, I drop the AGW attribution to 50% with a large error bar.

It seems to me that this statement has little or no scientific evidence to back it up.
The 50% of “natural variability” factors Dr. Curry mentions are either neutral or already negative :

The PDO is negative, with the harshest winter in at least 30 years in the West Arctic this year, the sun is particularly weak in this solar cycle, and the “natural” part of the AMO is hardly noticeable : North Atlantic waters have warmed, but not warmed more than the rest of the planet over the past 30 years.

So why did Arctic sea extent decline another 2 million km^2 (30%) even after the attribution studies that were presented in 2005 ?

And in absence of any multi-decadal “natural variability” cause, can anyone scientifically disprove the hypothesis that 100% of the multi-decadal decline is caused by external forcing (AGW), and that natural variability is just doing it’s normal thing : be variable, year-to-year ?

Regarding Arctic sea ice extent prior to satellite records, DMI has been making ice extent maps since 1898, based on ship logs.

Since some “skeptics” refer to the 30’s as having a “lower” ice extent, it may be interesting to look at this DMI 1933 report, to see how “low” that extent really was :
http://brunnur.vedur.is/pub/trausti/Iskort/Pdf/1933/1933.pdf

Look at August 1933 (close to minimum extent), and notice that ice is still present in the Kara sea, ice packed against the coast of Siberia, just east of Yamal, and on the other side against the coast of the Beaufort (we would need to go back 30 year to see that in August), some ice still in Baffin Bay, and Svalbard still has ice on the east coast.

In fact, it looks like there was more ice on the Atlantic side at the minimum extent in August 1933 than there is now in November 2012.

Dr. Curry herself referred to the work done by Chapman and Walsh, who used these DMI charts, and also the Russian AARI data, and a spectrum of other pre-satellite reports, and came up with this chart for Arctic Ice extent since the 1880’s :
http://nsidc.org/noaa/iicwg/presentations/IICWG_2011/Fetterer_Back_to_1870_Plans_for_a_Gridded_Sea_Ice_Product.pdf

Kinnard et al 2008 confirm these 20th century results, using paleo-climate data :
http://tamino.files.wordpress.com/2010/10/polyakfig2.jpg

If ice extent (or variability) earlier in the 20th century was really completely different than what Chapman and Walsh or Kinnard et al reconstructed, then why did nobody publish any paper contradicting their results ?

Further back, Kinnard et al 2011 reconstructed Arctic sea ice extent 1400 years back, using paleo climate proxies :
http://tamino.files.wordpress.com/2011/12/kinnard_2011_sea_ice.jpg?w=500
Notice that, yes, this does go all the way through the LIA (which seems to have a interesting dip in ice extent (maybe “warm Arctic, cold continent” effect)) and the MWP as well.

Just to indicate how absurdly low Arctic sea ice extent has gone, somebody added 2012 minimum into the Kinnard et al graph :
http://img849.imageshack.us/img849/5994/kinnaird2012.jpg

Does anyone still want to argue about the “low” ice extent of the 30’s, the MWP, or suggest that somehow, after decades of AGW ice thinning, this cliff that Arctic sea ice is falling off in the past few decades is “natural variability” ? Or that there may be a 50/50 chance that it would “recover” somehow ? Or that “we don’t know how to make decadal predictions of sea ice”, or “I DONT KNOW with any high level of confidence what the sea ice will look like in 10 years or 100 years” ?

Sometimes you don’t need to be a scientist to make a prediction about the future.

A picture says tells more than a thousand words.

So can we all please face reality by taking a good look at this last picture, and see where this is going ?

Thanks Judith Curry, Walt Meier and Ron Lindsay for pointing out your view points and interaction so far.

You seem to agree it is very complicated to put a number on the anthropogenic contribution in the sea ice decline. Nevertheless, you do have different positions in the debate. To summarize: Judith uses a wide range of 30-70 percent but in her story she focuses on the natural factors causing the decline and the uncertainty that almost prohibits to do a meaningful estimate. Also she believes that in the next two decades natural factors will be much more important than GHGs. Walt seems to believe that 50% is the best guess, where he focuses on the fact that thinner ice is much more prone to be melted due to natural factors (i.e. the Beaufort sea changing from a nursery into a graveyard). Ron tends to put a higher number on the anthropogenic contribution, based on model outcomes, using Ice Thickness in stead of Ice Extent.

Important issues to discuss next in my opinion are:
– Do you agree that Ice Thickness is a better indicator than Ice Extent to determine, among others, the relation with GHG-forcing? Walt already indicated he agrees (right?), but Judith didn’t comment on this yet.

– Walt Meier stated that “our expectations for how ice responds to natural variability is based upon a thicker ice cover, which may no longer be valid”. Is that the reason that climate models do not show the observed steep decline of sea ice extent/volume? And if so, what processes are missing?

– As brought in by Walt, Peter Wadham said in a recent interview that: “[…]the albedo—the fraction of solar radiation reflected back into space—drops from 0.6 to 0.1, which will accelerate warming of the Arctic […]the loss of the remaining summer ice will have the same warming effect on the earth as the past 25 years of carbon dioxide emissions.” I am very interested in your opinions on this, also because Judith Curry wrote in her introduction: ‘The impact of September sea ice loss on the ice albedo feedback mechanism is interesting. The minimum sea ice occurs during a period when the sun is at low elevation, so the direct ice albedo effect isn’t all that large. Less sea ice in autumn means more snowfall on the continents, which can have a larger impact on albedo.’

– In this context the comment from Rob Dekker (first public comment) is interesting. He indicates that GCM projections are lagging 20 (CMIP5) to 40 years (CMIP3) behind, equivalent to 0.5-1.0 W/m2 forcing. He shows (with references) this might be caused by the fact that snow cover decline is underestimated by models. This seems to be contradictory to Judith’s claim that snowfall (and thus snowcover) has increased.Judith, do you also have a reference to a study that supports this view?

@Rob Dekker 2012-11-15 08:32:38 (and various other bloggers):

Natural variability / climate feedback on a multi century timescale

The ocean circulation is an important climate feedback on a multi century timescale. And Arctic sea affects the intensity of this world wide circulation. This is roughly how it works.

In the (sub) tropics, the sun heats the (shallow) ocean. On its turn, the ocean radiates (IR) to the atmosphere where heat is trapped by GHG’s. But not all the heat is released instantly. The reason for that is that the shallow (heated) ocean is part of the global ‘ocean conveyor belt’ (THC/AMOC). The ocean redistributes this excess (sub)tropical heat towards the polar regions. There the deep ocean is fed by freezing cold water (NADW and AABW). As a consequence, enormous bodies of warm and cold water are being pumped incessantly around the globe. During the process, waters are ‘mixed’, warm water wells up and heat is released to the atmosphere. Note that the heat storage capacity of the ocean is thousand times larger than the one of the atmosphere, while the circulation period is more than a thousand years. However, the strength of the THC/AMOC is not constant. One of the factors that affects the intensity of the THC/AMOC, is the formation of North Atlantic Deep Water (NADW) in the Arctic. In this formation process, Arctic sea ice plays an important role.

Now let’s go back to the Little Ice Age (LIA). LIA was – probably – caused by an extremely inactive sun in combination with a series of volcanic super eruptions. As from the 18th/19th century, the sun regained strength and the heat flux to the (shallow) ocean increased. Ocean’s circulation starts to change, Arctic sea ice (and glaciers) start to retreat, earth’s albedo decreases, etc . Due to the inertness of the ocean system (heat capacity, circulation period, interplay with cryosphere, release of heat to atmosphere), it may take several centuries to reach a new state of equilibrium. So, in my view, a part of present day atmospheric warming can be explained by the presence of a secular warming trend since the end of LIA.

Bert,
Yes, these types of natural climate forcings seem to get lost on many people. Determining accurate natural changes are critical to calculating manmade changes. Without these determinations, the models are uselss.

Jim Cripwell on 2012-11-16, @13:13:40: Respectfully, your assertions are so wrong, as to not EVEN be wrong. I could point you to long-established (over a 150-year period, in fact) of exactly how CO2 does cause warming, causes it as it rises (ever heard of a really hot place, called “Venus?”) but something tells me you wouldn’t accept ANY rational scientific data as valid, given the wildly-incorrect nature of your statement. Saying it loudly, a lot, and all over the Internet does NOT make your opinion anything more than what it is: an opinion, and one not well-supported by any extant scientific data and research.

A better start would be to ask how ocean temperature data, ENSO, PDO, wind data, solar wind changes, planetary dynamics, etc etc can explain the ice loss. Without a handle on natural variabilty, the model results are necessarily junk.

Thus I conclude that this discussion is a very poor start for any scientific “Climate Dialogue”.

I am surprised that the focus has been totally on the Arctic, though clearly that is where the ice is melting. We are however actually dealing with the entire planetary system and at the opposite pole the Antarctic Ice cover is at a maximum since satellite coverage began. It is then difficult to argue that the change in Arctic Ice cover can be related to atmospheric CO2 content since it is essentially the same at both Poles, the difference between Northern Hemisphere and Southern Hemisphere atmospheric CO2 is minimal and certainly not in the range to explain the differences in behaviour of the Polar Ice. This suggests that the the Arctic melt is not related to AGW but occurs as a result of the major oceanic oscillations. It is the oceans which control the atmospheric temperatures and not the reverse. The conclusion on this basis is that we are observing a dominantly natural cycle such as has been observed previously in the 20s and thirties of the last Century

I am wondering, what happened to the Antarctic sea ice. It appears to be increasing, not as much as the Arctic ice is declining, but increasing nevertheless. My question is: what kind of models predicted this? Is it really possible that the planet Earth has that kind of bipolar disorder, that the Northern hemisphere exhibits all the features the GHG-dominated models predict, while the Southern Hemisphere is doing exactly the opposite? If the Arctic ice melt is anthropogenic, is the Antarctic ice gain also anthropogenic? Or maybe in the Southern Hemisphere some natural causes are overwhelming the anthropogenic forces? If so, what is the guarantee that the Arctic melt is not also dominated by some natural forces that would have melt the ice even in the absence of C)2 increase? And above all, not only that the Antarctic sea ice is increasing, the continental ice sheets are increasing as well and the temperature trend over most of the continent is flat or declining. Antarctica is a perfect hot-spot for the greenhouse warming because of an extremely dry atmosphere and almost complete isolation from the oceanic influences. If the recent warming was predominantly CO2 driven, it should be the most pronounced there. But it is not, on the contrary.

The only way of dealing with this for scientists was to fabricate some highly implausible and ad hoc assumptions to explain away the inconvenient Antarctic lack of warming and melt, and go full steam ahead with the Arctic scare stories. But, that ain’t gonna work, I am afraid.

The present behavior of Arctic sea ice should be viewed in the context of sea ice around both poles and what is known about polar ice during Pleistocene and Holocene Interglacials. If Arctic sea ice anomalies are diminishing while Antarctic sea ice anomalies increase, a global condition cannot be reasonably recognized without a clear explanation for the difference.

If it can be shown that present sea ice conditions are unremarkable compared with conditions in the recent and ancient past, one must be cautious about rushing to conclusions and making bold statements using words like “unprecedented” and “record-breaking.”

“Scientists discuss” … No they don’t.

Arts studects discuss, theological students, philosophers, etc. discuss. Real scientists create testable hypothesis and then test them against actual data. That is what makes them scientists.

I had the impression this should lead to a more scientific discussion between experts. When Rob van Dorland asks for some references, Prof. Curry gives a link to a blog-post in her comment on November 16, 2012 at 7:49 pm. The paper she refers to is published in PNAS:
http://www.pnas.org/content/109/11/4074.long

This Liu, Curry et al 2012 paper is about the influence of the loss of Arctic ice on winter snowfall in some recent winters. The paper says the following on snow extent:
“A decrease of autumn Arctic sea ice of 1 million km2 corresponds to a significantly above-normal winter snow cover (>3-12%) in large parts of the northern United States, northwestern and central Europe, and northern and central China.”
In the methods-section of the paper it is clear that the dataset from Rutgers University Snow Lab is used. This dataset shows a small positive trend for the complete Northern Hemisphere snowcover during the winter period:
http://climate.rutgers.edu/snowcover/chart_seasonal.php?ui_set=nhland&ui_season=1
The snowcover trend for NH fall is flat and the trend is clearly negative in NH spring, when the incoming energy from the sun is increasing again.

The following statement of Prof. Curry is therefore only relevant for the winter period, when the incoming solar energy is at a minimum:
“Less sea ice in autumn means more snowfall on the continents, which can have a larger impact on albedo.”
Looking at the dataset used for the Liu et al paper, it would be logical that the second part of the statement also applies to the spring period: less snow in spring has a large effect on the albedo. In this case a warming effect.

Prof. Curry’s statement: “The minimum sea ice occurs during a period when the sun is at low elevation, so the direct ice albedo effect isn’t all that large.” also seems rather strange to me. The ice extent is declining in every season and not only when the sun is at low elevation.

It is my opinion that statements that are incorrect should be addressed.

I would like to see a “reply” option on Public Comments, so that Public Commenters can reply to specific public comments.

I would also like to see a reliable method of addressing a comment to one of the Dialogue Participants — at least to the point where the moderator will CC: a Dialogue Participant if a Public Comment is directed to them.

Thank you,

I have two questions for Dr. Curry, which I hope she’ll answer if she deems them good enough:

1) You wrote: “models are inadequate”, ” I don’t have much faith in the climate model projections of sea ice” and “climate models, which have substantial problems in simulating the Arctic climate”.

At the same time you write: “The first issue to debunk is that an ‘ice free’ Arctic is some sort of ‘tipping point.’ A number of recent studies find that in models, the loss of summer sea ice cover is highly reversible.”

To a lay person there seems to be a contradiction there. Why do the tipping point models simulate Arctic climate better than sea ice projection models?

2) How does a cool PDO influence sea ice differently than a warm PDO? Isn’t the word ‘cool’ causing misunderstandings? The definition seems to be US-centric, meaning that the PDO is cool or warm depending on SSTs off the Pacific coast. But looking at this image (left is warm PDO, right is cool PDO). If this image is largely correct there looks to be more warm SSTs near Bering Strait when the PDO is cool than when the PDO is warm.

But let’s say that when the PDO is cool, colder water enters the Arctic through Bering Strait. Is that water really colder compared to the long-term average? Because if I’m not mistaken, the PDO is defined as either cool or warm after “the monthly mean global average SST anomalies
are removed to separate this pattern of variability from any “global warming” signal that may be present in the data” (source.

So, again, how will the PDO add to a slowdown or reversal of the negative trend in Arctic sea ice cover? If it’s not through changes in SST, is it through changes in SAT or changes in atmospheric patterns (Beaufort Gyre, etc)?

I suspect the same question applies to the AMO, but I haven’t looked into it enough.

Arctic sea ice perhaps not the best starting point. As others have observed, it is unwise to overlook the contrasting situation in the Antarctic.

I once had an opportunity to examine a Viking map, tattooed into a piece of leather, clearly showing routes that have rarely been navigable since.

Comments above include a lot of side-assumptions, eg GHGs trapping heat radiating from tropical oceans. If the “standard” model of atmospheric heat transfer (as per Trenberth etc) is correct, then my location (19° 11′ 38″S, 146° 40′ 31″E) would be uninhabitable, rather than as it is, benign – most of the time 🙂

Another theme puzzling me is the speculation about CO2 levels, eg are the Mauna Loa records reliable, representative, is CO2 a well-mixed component of the atmosphere etc.
The Mauna Loa monitoring was originally established for using CO2 as one proxy for detecting changes in volcanic activity, and no doubt useful for that purpose. There are places not far north of here, on the Ring of Fire, where almost neat CO2 continually bubbles up from the sea bed, like out of a can of soft drink, and yes, there are healthy reef flora and fauna, except in the immediate vicinity of fumaroles of course.

It is now possible to buy an accurate 1% CO2 datalogger for as little as $US65. No excuses for speculation any more – get out there and start logging. I have detected “spikes” in CO2 concentration during summer late afternoons, probable evidence of the West Pacific “carbon pipe” (out-gassing) phenomenon. Can’t be any anthropogenic source, occurs in daylight, so not to be confused with the natural increase of +/- 50 ppm that happens after sunset.

Words are important. Words need to mean what they say, especially important words like “ice free”. It’s worse than the absurd phrase “climate change’. Everybody here says “ice free” isn’t descriptive so please quit using it.

It is extremely important that you come up with an accurate phrase that means what it says.

I notice all the proclaimed experts in the discussion unquestionably accept the existence of so-called “carbon” gases in our atmosphere. Surely for a legitimate dialogue you need experts who are willing to question the very foundations of PCAGW.

How for example does carbon dioxide manage to float within the air itself being that it is heavier than air? Anyone who has held a diamond or a piece of graphite can attest that carbon is a solid, not a gas. As a gas it would simply fall to the ground as and so become solid. We see on Mars that the polar caps are made of carbon dioxide in it’s true solid state.

I appreciate that the invited authors are taking the time to respond to comments and each other. The format which groups their comments separately from general users works well, readers can follow a concise debate if they wish without wading through all comments. Separating out off-topic comments is also a good idea, as long as you have the resources to do this.

I’d still be interested in a response to my earlier comment, though. Is the loss of Arctic ice a concern, or is it advantageous, e.g. for transport and resource access? If AGW is occurring and increasing ice loss, should this be in the benefits column?

HarryWiggs, you say “Saying it loudly, a lot, and all over the Internet does NOT make your opinion anything more than what it is.”

I base my opinion on empirical data; or in the case of CAGW the almost complete lack of any empirical data to support the conclusion that CAGW exists. I am quite prepared to accept that my opinion could be wrong. I have been wrong in the past, and will, undoubtedly, be wrong in the future. But my opinion is irrelevant. The only thing that matters is the empirical data I present; or the lack of empirical data that I point to. If you want to criticise what I write, which I welcome, may I request that you concentrate on the empirical data, not the opinion I express.

Dr. Curry’s assessment that there is a greater than 66% likelihood that 50%+/-20% of the recent melt was caused by anthropogenic forcing.

Do we have any data on earlier melts of Arctic sea ice (1920s-1940s)? As I recall, there was a Russian study by Polyakov that showed similar melting to today, as well as a study by Chylek, which showed Greenland temperatures warming at the same rates as today. The temperature record for Illulissat, Greenland, shows this early 20thC warming trend:
http://farm3.static.flickr.com/2620/3797223161_16c1ac5e39_b.jpg

If these data show similar sea ice decline in the 1920-40 period, before there was much anthropogenic CO2 increase, this might help narrow down the anthropogenic portion of the recent decline.

Max

Also from my side I would like to thank Walt, Ron and Judith for the interesting discussion so far!
At this moment I want to address some issues concerning Judith’s interesting paper (Liu et al 2012), also because there are several public comments that are related to the Liu et al study. The paper gives the relationship between the decline of ASI and increasing snowfall in the NH and concludes that:

“The relationships documented here illustrate that the rapid loss of sea ice in summer and delayed recovery of sea ice in autumn modulates not only winter mean statistics (i.e., snow cover and temperature) but also the frequency of occurrence of weather events (i.e., cold air outbreaks). While natural chaotic variability remains a component of midlatitude atmospheric variability, recent loss of Arctic sea ice, with its signature on midlatitude atmospheric circulation, may load the dice in favor of snowier conditions in large parts of northern midlatitudes.”

Suppose that this is a real effect, then a decline in sea ice in summer and autumn would enhance the albedo (mainly in spring since albedo effects in winter are small due to low solar insolation). In turn this would result in cooling the continents around the Arctic. I think there are three possibilities of the effects on the temperature of the Arctic Ocean:

Possibility 1: more snowfall will cool the Arctic Ocean as well. Then, this is a negative feedback counteracting the decline in Sea Ice. Since a strong decline has been observed, shouldn’t there be another stronger process that explains the decline? And if so, what could that be?

Possibility 2: It has no effect on the temperature of the Arctic Sea. In that case I would conclude the mechanisms in the Liu et al. paper is of no importance to the amount of ice in the Arctic Ocean.

Possibility 3: It has a warming effect on the temperature of the Arctic Sea (a positive feedback).

I would be very interested in your view on this issue. Also, I am interested in the opinions of Walt and Ron.

Also, you state in the Liu et al. Paper that the decrease in albedo due to the decline in Arctic Sea Ice is an important effect.

“When highly reflective sea ice is replaced by open water during the ice melting period, there is a substantial solar heat input directly into the ocean, increasing the heat stored in the upper ocean. For example, the cumulative solar heat input in the Beaufort Sea during the ice melting period in 2007 can be a factor of two to five higher than climatology, which is sufficient to warm the upper 5 m of the ocean by 5°C (24). The loss of sea ice in the Canada basin has also been accompanied by the widespread appearance of a near-surface temperature maximum at 25- to 35-m depth due to penetrating solar radiation (25). Warming of the upper ocean retards the recovery of sea ice during the fall freeze-up.”

Could you indicate how this could be compatible with the mechanisms you have been discussing so far?

Finally there is an interesting comment from Jos Hagelaars (and a comparable one from Arjen van Beelen) stating that:

The Liu, Curry et al 2012 paper says the following on snow extent:
“A decrease of autumn Arctic sea ice of 1 million km2 corresponds to a significantly above-normal winter snow cover (>3-12%) in large parts of the northern United States, northwestern and central Europe, and northern and central China.”
The dataset from Rutgers University Snow Lab is used in the paper, that shows a small positive trend for the complete Northern Hemisphere snowcover during the winter period:

http://climate.rutgers.edu/snowcover/chart_seasonal.php?ui_set=nhland&ui_season=1

The snowcover trend for NH fall is flat and the trend is clearly negative in NH spring, when the incoming energy from the sun is increasing again.
The following statement of Prof. Curry is therefore only relevant for the winter period, when the incoming solar energy is at a minimum:
“Less sea ice in autumn means more snowfall on the continents, which can have a larger impact on albedo.”
Looking at the dataset used for the Liu et al paper, it would be logical that the second part of the statement also applies to the spring period: less snow in spring has a large effect on the albedo. In this case a warming effect.
Prof. Curry’s statement: “The minimum sea ice occurs during a period when the sun is at low elevation, so the direct ice albedo effect isn’t all that large.” also seems rather strange to me. The ice extent is declining in every season and not only when the sun is at low elevation.

—END of comment

Again it would be interesting to also hear the opnions of both Ron and Walt on this.
Finally, in this context there is also a comment from Rob Dekker asking Ron to reply:

Probably Dr. Lindsay could comment on this better than I could, but it seems to me that between GCMs and PIOMAS, PIOMAS is forced by NCEP/NCAR observational atmospheric data (at least for hind-casts), and thus has atmospheric effects of snow cover decline accounted for. So it may be no surprise that PIOMAS hind-casts appear to be more accurately simulating the ice volume and extent developments of Arctic sea ice extent.

Couldn’t find that link to the Polyakov study, but here is a paper by Dr. Arnd Bernaerts on the Arctic warming of the early 20th century (EAW):
http://noconsensus.wordpress.com/2009/11/04/past-arctic-warming-also-created-by-currents/

From this paper

“How is the EAW explained today

Although 90 years have passed since the earlier Arctic Warming (EAW) commenced, it seems that the issue is still one of the most puzzling climatic anomalies of the 20th century (Bengtsson, 2004:4055), and there are many questions not answered yet. Instead the matter is often sidelined by regarding it as:
• Natural variability is the most likely cause (Bengtsson, 2004:4045);
• We theorize that the Arctic warming in the 1920s/1930s was due to natural fluctuations internal to the climate system (Johannessen, 2004:341)
• The temperature anomalies are due primarily to natural variability in the weather system (Overland, 2008:81).

Such notions explain too little. The matter becomes even more critical if claimed without sufficient prove, for example: That the past 100 years are significant for the changeover of a climate system dominated by natural forcing to a climate system dominated by anthropogenic influences, as done recently by S. Brönnimann et al., while admitting that “Our understanding of the climate mechanisms operating in the Arctic on different timescales is still limited”. (Brönnimann, 2008: 3, 20) Is it helpful to dramatize the shrinking sea ice during a recent time period, if one is not fully aware of what happened in the early years of the last century?

Actually, few but sufficient air temperature data are available since long time. The paper will use them to show that the early warming was initiated and sustained over two decades by the West Spitsbergen Current, a branch of the Gulf Current in the North Atlantic. Naming the sea as the cause of the Arctic warming from 1919-1939 could help to understand better the current situation in the Arctic.”

End of quote.

Will keep looking for the original Polyakov study.

Max

I propose that writers in this blog ban the word “may” from their texts since the term always implies is opposite “may not” at the same time. Casual readers sometimes overlook the uncertainty present in it. An example from the comments?”:

Neven
2012-11-14 22:31:10

Dr. Meier wrote:

Our expectations for how the ice responds to natural variability is based upon a thicker ice cover, which may no longer be valid.

Bingo…

Neven raises an important point re how much heat does it take to melt the Arctic sea ice. He cites an article by Kwok and Untersteiner citing a number of 1 W m-2. Unfortunately, I am on travel and my laptop is having difficulty opening that article, I can only see the first two pages. So I don’t know exactly where that number comes from.

Heat from the ocean is more effective at melting the ice than heat from the atmosphere. A while back I did some model experiments looking at how much heat from the atmosphere (a perturbation to the downwelling IR flux or surface air temperature) that it would take to melt the sea ice, as a function of different physics included in the sea ice model.

I refer interested readers to the following publications:
Holland, M., J.A. Curry, and J.L. Schramm, 1997: Modeling the thermodynamics of a distribution of sea ice thicknesses. Part II: Ice/ocean interactions. J. Geophys. Res., 102, 23093-23108. http://curry.eas.gatech.edu/currydoc/Holland_JGR102.pdf

Holland, M.M., J.L. Schramm, and J.A. Curry, 1997: Thermodynamic feedback processes in a single-column sea ice/ocean model. Ann. Glaciol., 25, 327-332.

Arbetter, T., J.A. Curry, M.M. Holland, and J. M. Maslanik, 1997: Response of sea ice models to perturbations in surface heat flux. Ann. Glaciol., 25, 193-197.

Holland, M.M. and J.A. Curry, 1999: The role of different physical process in determining the interdecadal variability of Arctic sea ice. J. Climate, 12, 3319-3330. http://curry.eas.gatech.edu/currydoc/Holland_JC12.pdf

Curry, J.A., J.L. Schramm, A. Alam, R. Reeder, T.E. Arbetter, P. Guest, 2002: Evaluation of data sets used to force sea ice models in the Arctic Ocean. J. Geophys Res., 107, art. no 3102. http://curry.eas.gatech.edu/currydoc/Curry_JGR107.pdf

Unfortunately the Journal of Glaciology papers are not available online, I probably have hard copies buried somewhere in my office, I will try to put them online. From the Arbetter et al. paper, which asked how much IR forcing is needed to melt the ice from above, the answer depended on which sea ice model you used, but for a dynamic/thermodynamic sea ice model, the result was tens of W m-2.

Melting the ice from below is much more powerful than melting the ice from above, in terms of W m-2. Note solar radiation penetrates into the mixed layer through open waterm leads and melt ponds and thin ice, so solar acts to melt from below as well as from above. See Holland et al. Part II paper for the details of physical processes of melting the ice from below.

What fascinated me about the big August storm was the potential for deep mixing of the ocean down to the warm thermocline: this could have a much larger impact than any radiative forcing from above on the evolution of the sea ice. So what is going on in the Arctic Ocean is a combination of overall warming from GHG, as well as natural climate variability and weather roulette.

Philip Bradley raises a very interesting point:

“The trend in the Arctic ice minimum extent is -8.2%/decade. While the trend in the maximum extent is -2.78%/decade. This means we are seeing both increasing summer ice melt and increasing winter ice formation (measured by extent).

This IMO points to decreased cloud cover as the primary cause of both, from increased summer insolation warming and increased winter radiative cooling.”

I have been looking at this also. Refer to this diagram: http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/seaice.anomaly.arctic.png
During the period 1997-2006, there is a downward trend in overall sea ice extent, but the seasonal variability in sea ice extent is small and the trend seems equal for summer/fall min versus winter/spring max. Then, since 2007, we see a much larger annual cycle in sea ice extent, driven by the substantially lower summer extent, but but the recent winter extent is greater than in 2005, 2006, 2007.

I suspect also that changing cloud patterns has something to do with this. There is a new paper that looks at the cloud characteristics over the Arctic since 2006, using Clousat and Calipso satellite radar and lidar. http://digitalcommons.unl.edu/usdeptcommercepub/344/. Comparing their Fig 4 (since 2006) with the previous climatologies (Fig 1), it seems that recent summer is lower than climatology and recent winter is higher than climatology. The lower cloud amount in summer is working in the right direction to reduce the sea extent at end of summer. But higher winter cloud amount points in the direction of less sea ice (in winter over the arctic, clouds warm the surface).

So for winter, it seems like ice export or ocean heat transport are the main factors (I haven’t spotted recent measurements of Fram Strait ice export?). It will be VERY interesting to what the max winter sea ice extent is for 2013, I suspect that it might be low as a result of the ocean mixing from the August storm.

A response to Bart Strenger’s comment:

With regards to the impact of increased snowfall on the Arctic Ocean during winter. If there is heavy snowfall during autumn, this will slow down the ice growth process, since snow has an insulating effect. If there is heavy snowfall in late winter/early spring, this can slow down the ice melt process since the sun spends its energy melting the snow before it can start melting the ice.

I wouldn’t quite agree with any of the 3 possibilities put forward by Bart. I suspect that colder, snowier NH continents would primarily have an impact on the hemispheric atmospheric circulations and so influence ice export. And an indirect cooling effect on the Arctic Ocean during winter. So I think the dynamical response in terms of ice export is at least as important as any thermodynamical response.

With regards to the albedo effect of changes to snow and ice cover, to first order this relates to the solar zenith angle (time of year, latitude) and the cloud cover. For low sun elevations (below 55 degrees), cloudiness will reduce the surface reflection of solar radiation relative to clear sky conditions. At higher sun elevations, the reverse is true.

It’s an honour to have been mentioned by Dr. Curry, but it would’ve been even better if she had taken that time to address my two questions.

Something else Dr. Curry wrote, caught my eye:

Philip Bradley raises a very interesting point:

“The trend in the Arctic ice minimum extent is -8.2%/decade. While the trend in the maximum extent is -2.78%/decade. This means we are seeing both increasing summer ice melt and increasing winter ice formation (measured by extent).

This IMO points to decreased cloud cover as the primary cause of both, from increased summer insolation warming and increased winter radiative cooling.”

This might be a very interesting point until one realizes that 1) we’re talking about extent here, whereas volume would be a much more useful measurement to compare the declines in maximum and minimum, and 2) the Arctic Ocean is constrained by land masses, otherwise we might see just as big a maximum decline.

Of course, sea ice experts realize this, which is why you won’t see Dr. Meier or Lindsay bring it up.

Dr. Curry wrote:

“So for winter, it seems like ice export or ocean heat transport are the main factors (I haven’t spotted recent measurements of Fram Strait ice export?). It will be VERY interesting to what the max winter sea ice extent is for 2013, I suspect that it might be low as a result of the ocean mixing from the August storm.”

I find this statement quite puzzling. I can understand the process (lower heat content in deeper waters), but this might lead to more rapid freezing in this area in fall (do we have any indication that this has occured?), and later opening up of this area during spring. Maximum extent is reached WELL outside the regions affected by the storm. These areas are always frozen over near maximum extent!

Autumn is still lingering in the UK the leaves are still falling off the trees,there would need to be a rapid change to support the idea that Arctic Sea Ice loss equals severe winter weather.
The growth in Sea Ice and cold weather seems to favour the Pacific side of the Arctic as it did last year.
http://www.natice.noaa.gov/pub/ims/ims_gif/DATA/cursnow_asiaeurope.gif

Jim Cripwell,
You continue to reference your “catastrophic” AGW (CAGW) term, but fail to notice that the word “Catastrophic” is inherently an opinion, and thus does not have any place in any scientific discussion about Anthropogenic Global Warming (AGW).
In fact, there is not a single scientific paper that defines the term CAGW, and it would be nice if you would identify who first used that term, and in which apparently non-scientific context.

Bart, I am not sure I understand the question. Our PNAS paper addressed wintertime snowfall. In some regions of the NH continents, there is a negative correlation between winter snowfall and autumn sea ice extent. The continental snow is melting earlier in spring. I don’t see a conflict between these?

A response to Neven’s question:

“1) You wrote: “models are inadequate”, ” I don’t have much faith in the climate model projections of sea ice” and “climate models, which have substantial problems in simulating the Arctic climate”.

At the same time you write: “The first issue to debunk is that an ‘ice free’ Arctic is some sort of ‘tipping point.’ A number of recent studies find that in models, the loss of summer sea ice cover is highly reversible.”

To a lay person there seems to be a contradiction there. Why do the tipping point models simulate Arctic climate better than sea ice projection models?”

Climate models are able to simulate the annual cycle of sea ice, i.e. melting in the summer and freezing in the winter. So each winter, the sea ice cover recovers, more or less depending on the weather. No one disagrees with this.

Further, ‘ice free’ is usually taken to mean less than 1 M sq km. This is far from ‘ice free’ in actuality (I find this use of ‘ice free’ to be highly misleading). So what is the point of talking about some sort of ’tipping’ point, when even during summer, the Arctic Ocean is not really projected to be ice free?

The massive cooling during the polar night causes substantial heat loss, which on land results in surface temperatures of -40C and colder. Sea water in the Arctic Ocean freezes at a temp slightly warmer than -2C. The only conceivable way to keep the Arctic Ocean ice free is to bring in much warmer ocean water from lower latitudes. Unless the geography of the Arctic basin dramatically changes, e.g. Alaska disappears, there is no way for this to happen.

I don’t think that anyone would dispute these points.

With regards to your question on PDO. I don’t know how to answer the question in the manner that you have posed it. Large scale atmospheric and ocean circulations vary on multidecadal time scales, which influence circulation regions on shorter time scales as well, which among other things influences sea ice characteristics. Untangling how all this works has received far too little attention IMO.

cwschoneveld
2012-11-17 13:53:04:

“I propose that writers in this blog ban the word “may” from their texts”

in the same comment:
“Dr. Meier wrote:

Our expectations for how the ice responds to natural variability is based upon a thicker ice cover, which may no longer be valid.”

Bingo?

Dr. Curry wrote: “The past 25 years of carbon dioxide increase has contributed to less than 1 W m-2 in radiative forcing. There is no simple calculation to be made of the effect of the sea ice albedo change on the global energy balance. The better comparison is with a change in fractional cloud cover (which is highly variable), which dominates the planetary albedo far more than the arctic sea ice.”

Rob van Dorland wrote: “This seems to be contradictory to Judith’s claim that snowfall (and thus snowcover) has increased. Judith, do you also have a reference to a study that supports this view?”

Though increased snowfall in winter in recent years (is this trend significant?), and decreased snow area in other seasons are not incompatible, this is not what I commented on.

I wanted to point out that, in contrary to the possible small changes in snow area in winter, the total sea ice + snow area has strongly decreased during spring and summer (5-8 million km2). See the links in my previous post. This does lead to a large albedo effect (which might actually feedback on the sea ice extent in summer). This must have very large climate effects, especially in regions previously covered by snow.

I was wondering if Dr. Curry agrees with this, and if she has any data backing up her comment that cloud cover variations (during similar timescales..) dominate the changes in planetary albedo over the observed sea ice + snow area changes in summer. Is there any reliable data showing that there are such large trends in cloud cover?

I see you have adressed part of my second question. Thank you very much for explaining why enhanced mixing by the storm decreases the heat content of the ocean column (I was already aware of this (not of details though), but for other readers it might be much clarifying). However, this has not answered my question. It is a direct comment on what you wrote.

Dr. Curry wrote:

“It will be VERY interesting to what the max winter sea ice extent is for 2013, I suspect that it might be low as a result of the ocean mixing from the August storm.”

Personally, I don’t see how this might strongly influence winter (max) extent, because the area affected by the storm is well outside the region where the max extent is reached. The storm affected areas which are always frozen in winter.

I can understand that the storm would lead to faster freeze up in fall (is there any indication that this has occured? Maybe Neven knows?), maybe thicker ice in this area in winter and later melt during spring. I do however NOT see how this would strongly influence max extent.

Again, when I read Dr. Curry’s comments, I read uncertainty here, uncertainty there, snow is increasing and the ice might restore next year because of the storm. (Exaggerating a bit here of course). The reader might get the impression that these things are the dominant changes currently going on in the Arctic; I think this is not a good representation!

This topic is starting to become more and more interesting, I’d like to thank the authors and hosts for their efforts!

To try to keep this to the topic as originally posted, please show the peer-reviewed “empirical data” you continue to reference, that robustly supports that the loss of the arctic’s ice in 2012 is *not* at all a result of anthropogenic forcings. Also, and to repeat Rob Dekker’s point, your continuing to use the improper term, “CAGW,” tends to lend credence to the idea that no matter how much scientific, refereed data any of us can show you, you might just reject it. to complete this post, let me again say that “CAGW” is *strictly* a construct of those who reject, out of hand, any data that supports the *proper” term, AGW.

I await your reference list, as asked for.

A few comments concerning Judy’s division between the natural variability and GHG forced sea ice extent.
The mean (June to November) Arctic sea ice extent is highly correlated to the mean high Arctic temperature (just look at the data for sea ice extent available from the National Snow and Ice Data Center and the temperature at high Arctic met stations available from e.g. the NASA GISS data). Polyakov et al (Clim Dyn 2009) finds that about 60% of North Atlantic post 1970 warming was due to the multi-decadal variability. Chylek et al (Geophys. Res. Lett. 2010) finds that about a half of the Arctic post 1975 warming was due to the anthropogenic GHGs and about half to the AMO (Atlantic Multi-decadal Oscillation). Mahajan et al (J. Clim. 2011) finds that the GFDL climate model simulations shows the Arctic sea ice extent highly correlated to the AMOC (Atlantic Meridional Overturning Circulation).
All these points towards a large natural variability of the Arctic temperature and sea ice extent. If AMO will continue its 60-70 year cycle (as seen in the 20th century) we can expect the natural climate variability to start counteracting the GHG anthropogenic warming and slow down or stop the Arctic warming during the next few decades. If on the other hand the AMO will continue rising (which would be against hundreds or thousands years of the AMO records – Delworth and Mann 2000, Gray et al, Geophys. Res. Lett. 2004) the Arctic will continue warming as predicted by climate models. However, the climate models (CMIP3) are not able to reproduce the 20th century Arctic temperature variability (Wang et al 2007, J. Clim.), so their prediction of the future Arctic climate is in doubt.

Harry Wiggs, It is my understanding that Marcel intends to introduce a thread on Climate Sensitivity. I suggest that it would make sense to wait until this thread is introduced, and then we can discuss the details you want, on topic, instead of off topic. Please note that the empirical data on global temperatures that is reported on a daily, weekly, monthly and yearly basis is NEVER peer reviewed or refereed.

I can understand that the storm would lead to faster freeze up in fall (is there any indication that this has occured? Maybe Neven knows?)

Arjan, I don’t actually know, but if the summer storm brought up heat from lower layers in the Arctic Ocean, it must have gone somewhere else, because the areas that were most affected by the storm (East Siberian Sea, Laptev Sea), have frozen up relatively fast, as you can see on this regional graphs page. The Beaufort Sea was anomalously warm for most of the melting season, but is now filled with ice, and the Chukchi Sea is also freezing relatively fast.

Given the SSTs in the North Pacific, I wouldn’t be surprised to see the same thing happening as last year: lots of cold and ice on the Pacific side of the Arctic, and a very slow freeze on the Atlantic side of the Arctic (Kara Sea has finally begun to freeze, Barentsz is lagging).

Those of us who are not sure how much to trust GCM’s would have appreciated a summary of the basic physical processes occurring in the Arctic. The average volume of ice in the Arctic has diminished by 310+/-100 km^3/year over 1979-2011 (PIOMAS). Wikipedia says the area of the Arctic Ocean is 14,056,000 km^2 (but the areas defined by Arctic sea ice and the official Arctic Ocean certainly aren’t the same. Dividing gives a rough approximation of 22 mm/yr of decreasing ice thickness. The enthalpy of fusion for ice is 334 kJ/kg times 916.7 kg/m3 for 306,000 kJ/m3 or 306 kJ/m2/mm. Assuming these calculations reasonably and correctly approximate the situation, we need 6800 kJ/m2/yr or 0.21 W/m2 to melt 22 mm/yr of ice.

If one wants to attribute the change in sea ice to a change in anthropogenic GHGs, the next question should be: How is the surface radiative balance changed by GHGs With a standard atmosphere, there is roughly a 1 W/m2 decrease in upward flux at the surface for 2X CO2, but the Arctic is drier, colder and stratified. Of course, only the increase DLR absorbed by sea ice can directly cause melting.

Do traditional feedbacks apply to the Arctic with its stratified atmosphere, temperature inversions and subsistence? WIth surface temperature pegged near freezing during most of the melting season and with little vertical mixing, is water vapor feedback important? The change in energy flux associated with melting is massive: Hudson calculated that the current reduction in sea ice has produced a GLOBAL sea ice albedo feedback of 0.1 W/m2 (from 2-3% of the earth’s surface), but this could be negated by a 2.3% increase in Arctic cloud cover. (Hudson, S. R. (2011), Estimating the global radiative impact of the sea ice-albedo feedback in the Arctic, J. Geophys. Res., 116, D16102.) Unfortunately, melting of sea ice in the Bering Sea and Hudson Bay in June are unlikely to have any impact on whether sea ice at 85 degN melts in late August. Even inside the Arctic Ocean itself, convection is needed to transfer heat from regions melting earlier than normal to the remainng sea ice. Late in the season, convection moves floating sea ice, reducing its effectiveness. Changes in clouds and convection are potentially much larger than radiative forcing. It isn’t obvious why I should believe that GCMs are remotely capable of accurately modeling these convective phenomena. This is particularly true when I read how sea ice in the ECHAM GCM was “tuned” for AR5: Mauritsen, T., et al. (2012), Tuning the climate of a global model, J. Adv. Model. Earth Syst., 4, M00A01, doi:10.1029/2012MS000154. Uncertain albedo causes major difficulties for even modern weather forecasting programs near the rain/snow line for many days after a winter storm.

Under these circumstances, attributing some fraction of the decrease in sea ice to GHGs and another fraction to natural variability appears to represent hubris. Climate science should be able to say with some confidence that the increase in DLR associated with increased GHGs is (or is not) large enough to provide the energy needed to melt roughly 20 mm per year of Arctic sea ice*, but this doesn’t mean these changes were caused by GHGs. After all, the same change in DLR isn’t causing a similar change in Antarctic sea ice. Modelers can tune their GCMs to reproduce the sea ice observed during the 1980s or the 2000s, but fiddling with multiple interacting parameters doesn’t make such models suitable for attribution or prediction. Are the same sea ice parameters used at each pole?

*Even simple linear trends can be misleading. Consider a simplified Arctic Ocean permanently and stably covered with thick ice, 5 meters at the maximum and 3 meters at the minimum; with 2 m of ice melting in the summer matched by 2 m of ice thickening during the winter. Now increase the amount of DLR so that 50 mm of additional ice melts every summer. One would predict ice thickness varying between 2 and 4 meters after 20 years. As the ice gets thinner, (exponentially) more heat will escape through the ice during the winter, the linear trend will diminish and stability will eventually be restored. After 2012’s record melt, it has taken about 1 month longer than in 1980 for the Arctic Ocean to freeze to the shores of Asia and North America. (igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=10&fd=10&fy=1980&sm=11&sd=10&sy=2012) Some areas on the edge of the Arctic Ocean (Kara and Barents seas) are lagging further behind. Instead of thickening for at least six months as in 1980-1, this winter the ice will have about one month less to thicken. However, how much does sea ice actually thicken during the last month of the freezing season when the heat of fusion presumably must travel through more than one meter of ice by conduction and the air above the ice has begun to warm? Do GCMs produce reasonable results for the increase in thickness of first year sea ice with time?

The three professors commenting above have personally struggled with many aspects Arctic climate for decades, figuratively studying many trees in a forest of challenging phenomena. Amateurs (like me) may spend a few hours flying above this forest with some back-of-the-envelop calculations, wondering if the experts are only seeing the part of the forest where scientific progress is being made. (That’s the ultimate in hubris, of course.) Above Professor Curry notes that a radiation transfer problem with clear sky radiative transfer has been solved with the “dirty window”, but she isn’t sure which models incorporate this correction. Unfortunately for my confidence, clear-sky radiation transfer is the “simplest tree in this forest”.

Winter surface ice area/extent has little to say about summer/fall extent in the Arctic Ocean, because the Arctic Ocean completely freezes over during the winter. The areas which add to winter ice extent (Hudson’s Bay, Baltic Sea, Kamchatka Sea, etc.) are guaranteed to melt during the summer/fall. Where they provide some indication as to trends is the date at which they are ice free, which is, indeed earlier.

Anyone pointing to total winter sea ice extent as an indicator of what is happening in the summer/fall is selling you a McGuffin

Dr. Curry wrote:

Clouds are much more powerful radiatively than CO2, so if we are talking about radiative forcing, clouds should be front and center in the discussion.

Again, talking about CO2 causes misunderstandings (see Frank’s comment). It’s not the CO2 over the Arctic that is causing the Arctic sea ice to melt, it’s the accumulated heat (see this graph that shows total heat content) in the Earth system that is most probably causing the sea ice to melt so precipitously.

Dr. Curry also wrote:

Neven, Philip Bradley brought up the issue of clouds, I responded to that point. This response does not imply that I think clouds are the only or even the main driver of the summer sea ice minimum.

If you are going to encourage Philip Bradley (“Philip Bradley raises a very interesting point”), you should also tell him that his methodology isn’t adequate as evidence for his hypothesis. By encouraging him you also tend to make others think you agree with his flawed methodology. Your latest comment seems to imply that you agree with me on this point (or else you would have addressed it, I assume), which means my explanation of volume as a better metric for comparing max and min decadal decline and geographical constraints hasn’t been for nothing.

That’s important to me, because my wife told the geographical constraints and PDO to go to hell last night. 🙂

Regarding the term CAGW I have used it for many years so may or may not have introduced it. The distinction between AGW and CAGW is that between the past and the future impacts, so it is scientific. AGW means human activity has caused at least a major fraction of global warming (and related changes). CAGW means continued human activity will cause dangerous levels of change, many of which have been conjecturally described in the literature. There is in fact an extensive scientific literature on the possible negative impacts to come. As Jim Cripwell points out it is CAGW that matters but CAGW is based on AGW. If AGW is false then CAGW falls, hence AGW is of great interest. But we are all here because of the CAGW speculation. I hope this helps resolve the confusion.

Frank, that was a very thoughtful post. In the early 1990’s, i was working hard on trying to understand the forest, and believe me there wasn’t much to work with. By forest here, i mean clouds, radiation, sea ice, and the feedbacks among them. The culmination of this effort was my review paper
Curry, J.A., D. Randall, and W.B. Rossow, and J.L. Schramm, 1996: Overview of arctic cloud and radiation characteristics. J. Clim., 9, 1731-1764.
http://curry.eas.gatech.edu/currydoc/Curry_JC9.pdf

In the early 90’s, i was one of a small handful of scientists working on these problems. Global climate scientists tended to stop at 60N, 60S.

This work motivated the SHEBA experiment
Uttal, T., Curry, J.A., and 26 others, 2002: Surface Heat Budget of the Arctic Ocean. Bull. Amer. Meteor. Soc., 83, 255-275.
http://curry.eas.gatech.edu/currydoc/Uttal_BAMS83.pdf

SHEBA brought the topic into the mainstream, and there has been a mini-explosion of people now working on these topics. With notable improvements to satellite observations, sea ice models, and cloud microphysics

With the dramatic increase in interest associated with sea ice decline, we haven’t really seen a synthesis of what is going on. In 1996, my article cited about 100 papers, and I really had to dig into some old Russian papers. Now, there are probably a thousand papers to look at. When I retire, maybe I can tackle this one. Or maybe we need a scientific assessment specifically on this topic, with an international team of people.

I agree that we are collectively not yet seeing the forest on this one.

Cooter –
1. The relationship is not linear. Probably the closest is that rate of ice loss relates to temperature and to ice thickness from last year. ie, if the sea currents are warm they could continue to melt ice even if they aren’t getting any warmer, and the thinner the ice is to start with, the faster the ice loss is likely to be.
2. Agree. “Ice free” should mean no ice, but it could reasonably be defined as (eg.) free of all ice that could impede shipping. Whatever, the definition being used needs to be defined, and in the absence of explicit definition it must retain its plain english meaning – “no ice”.
3. The element in the oven is still on so that part of your argument is invalid (the element hasn’t got any hotter for a while and may start getting cooler soon, but it is still pretty warm today). But the failure to look at the Antarctic is disgraceful, and is as you say “a case of scientists turning a blind eye to the data that does not match their working hypothesis.”. Particularly as the IPCC report states clearly that sea ice loss is expected at both poles [AR4 ch.10, “Snow and Ice”].
4. Agree totally. The only way science properly progresses is with theories open to testing and with actual evidence.

I think this is an excellent initiative. However, it is only an important and novel contribution insofar as genuine experts are invited. My impression of Dr. Curry is that she is a premiere example of an expert who voices some skeptical arguments concerning aspects of anthropogenic global warming Her arguments are thoughtful, evidence-based, and grounded in peer-reviewed, published science. I will quickly lose interest in this site if “experts” are invited who do not have a sustained and relatively recent record of publishing peer-reviewed papers in geophysics and climate science. The mere possession of a Ph.D.–and even the title of “Professor”–does not constitute scientific expertise.

Please maintain well-accepted professional standards of scientific expertise in selecting experts to invite.

I have a question about the amount of energy that’s melting Arctic ice as a percentage of the global energy imbalance. Above several estimates have been given about the energy that’s needed to explain the melting of Arctic ice: from a maximum of almost 1 W/m2 to about .25 W/m2. But how much is this energy as a percentage of the global energy imbalance?

Last year Jim Hansen and others wrote an interesting paper about the global energy imbalance:
http://pubs.giss.nasa.gov/docs/2011/2011_Hansen_etal.pdf

They conclude that the current energy imbalance is about .75 W/m2 averaged over the 11 yr period of the solar cycle. They also seem to show in their figure 8 that over the past 10 years about about 2% (.015 W/m2) of that extra energy has been used to melt sea ice, and that the amount of energy melting sea ice seems to be growing quite rapidly. I assume that practically all of this energy is being used to melt Arctic sea ice.

Do I understand this correctly, and what do Walt, Judith and Ron think of this estimate of Hansen et al? And also: by how much can we expect the global energy imbalance to grow over the coming decades and what could that imply for the amount of energy that goes into melting Arctic sea ice? Would the part of the global energy imbalance that goes into melting the sea ice grow or stay the same? I’m also curious to hear what others besides the three guest bloggers think on this.

Reading the discussion so far it seems the question on the potential importance of ice albedo positive feedback hasn’t been properly answered yet. Or maybe I’ve missed it?

At first Judith seemed to say it isn’t an important effect, or maybe even non-existent. Several questions for clarification haven’t been clearly answered, but give the impression that there is some relevant effect after all, although a (negative?) cloud feedback would seem more important to her.

So for further clarification, could all three guest bloggers please answer this question: what are the indications for a positive ice albedo effect and a possibly negative cloud feedback, and what estimates for their magnitude have been found thus far?

Charles

Having thought about this, I think I was premature to introduce this data at this time. It will be far more relevant when we discuss climate sensitivity. For the record I include the CET graph, not because it is global, which it isn’t, but because it is the longest continuous record of directly measured temperatures, going back the the mid-17th century. So, it reinforces the global data. The second graph starts around 1850, and is the longest record of directly measured global temperatures which is available.

Paul Vincelli you write “However, it is only an important and novel contribution insofar as genuine experts.”

Richard Feynman famously remarked “Science is the belief in the ignorance of experts”. On a personal basis, I only trust empirical data. If there is a paper written by anyone, which is backed up by a wealth of empirical data, I really dont care if the author is an expert or not. If the data is good, and the science presented is equally good, then I find the paper to be good science. If the paper is written by the best experts in the field, and the results are based on the output of non-validated models, as is so often the case with CAGW, then, IMHO, the paper is not worth the powder to blow it to hell. I have far more faith in the recordings of daily temperature readings from the AMSU satellite, than I have in the IPCC reports, such as the TAR and AR4. But then, that’s what makes the ball game.

Energy to Melt Glaciers
Regarding the energy levels necessary to melt Earth’s glaciers. I computed the energy to extract Earth out of the Younger Dryas Event at ≈ 11,700 yBP to be ≈ 123 times solar radiation or ≈ 170 kilowatts per square meter. Solar is ≈ 1.368 kilowatts per square meter in space,. This energy level acted for ≈ 150 years at the NGRIP ice core site. That is ≈ 1.13 kw/m^2 per year. See the next paragraph for solar only melting glaciers. The GISP-2 ice core site had lower energy levels lower than 35 kw/m^2 for ≈ 189 years. A level of 0.185 kw/m^2 per year. So it appears that solar could not have been the cause of deglaciation in such a short period. If you examine my work you will discover the cause of glacial cycles over long periods and some sediment cores predict millions of years.

Some of my other research indicates solar energy would have accomplish the same glacial melting ending about 5000 years in the future. See Ill 3.0 in the text book. So this means that solar equivalent is ≈ 1.36 kilowatts per square meter for ≈ 35,000 years vs ≈ 150 years with energy supplied by the external source.

My illustration, Ill 7.7, predicts that Earth is well past, ≈ 5000 years, the point where it would have begun returning to low temperatures leading to glaciation. The controlling factor is the climate servos, two primary, a few secondary and an unknown number as tertiary feedback control systems and a new atmospheric state. These are distributed control systems rather than being in a black box as they are generally understood. Feedback control systems suffer from noise input. Modern Man is a noisy bunch. They also fail and that can be in positive or negative directions of the controlled variable, in this case temperature. In the case at hand, it is an excess of feedback variables, moisture and CO2, which are to be considered.

So, it is obvious that high levels of energy are needed to melt the glaciers in a relative short period as we see in dispersal of the high level glaciation existing from ≈ 27,500 to ≈ 10,000 yBP. I called the Global Glacial Maximum in Antarctica the Quiet Period. It appeared that nothing was happening with the climate, just cold, cold in that period. I later developed a theory for what caused the Quiet Period. It is a shocking revelation. And, may I add, it could happen again starting tomorrow.

That melting from ≈ 30,000 to ≈ 10,000 yBP produced a sea level rise of ≈ 120 meters, 130 meters by some estimates. If the poles continue to melt today, then global warming will have proven itself with disastrous results. The saltwater/freshwater interface of rivers will move upstream a frightening distance world wide

It is not a case of what is the cause of the warming. It is warming and our new task is to work with Nature to mitigate that state as best we can. I think an excellent example of man made environmental disasters is demonstrated by the media production from Florentine Films and the USA Public Broadcasting/Educational television station with call letters WETA, under direction of Ken Burns. WETA is located in the Washington, D. C., USA metro area. Here we see a graphic example of the suffering of humans caused by humans. However, with some excellent work those same humans ended that disaster. The tropical storm(?) Sandy is still another example along with those disasters around the globe which are not weather related. The common thread is, can Modern Man gain control of his unusual talents and mitigate all disasters.

Incidentally, I am selling my research. It took three years to produce and no outside financial support. The series title of my work is:
“ THE ADVENT of MODERN MAN. “
And Volume II title: “ The End of The Last Ice Age. “
Actually, I should have ended it with All Ice Ages. Those of at least the last several million years.
ISBN 9780984918119 Find it at:
http://www.omega-spacepublishingusa.com. And, at Barnes & Nobel when administrative tasks are complete.

Everette L. Wampler

Beyond this however, caution needs to be taken when making connections to global climate sensitivity estimates or to (more recently) synoptic-scale weather feedbacks.

Caution also needs to be taken with being too cautious and be silent about potential worst-case scenarios. Again, Emanuel: “This philosophy makes sense for science, it can be disastrous when applied to risk assessment.”

It must be recognized that while Arctic sea ice loss is a big “tipping point” at an ecosystem, regional climate, and probably socio-political-economic level, that from the global climate system perspective it’s still relatively small potatoes.

Please, define ‘regional’. Does that include the NH land masses? Or does a potential increase in blocking patterns and anomalous snowfall patterns stay restricted to the Arctic Ocean? How about the Greenland Ice Sheet and Arctic glaciers?

If all of that is small potatoes, where does that leave us from the risk assessment POV?

Chris – to shamelessly self-promote and repeat a comment I made earlier – if you look at a “climate shift” in 2001-2002 and the trends in 1) TLT and 2) arctic sea ice, the total energy trend is not shifted; less TLT warming but more arctic sea ice (PIOMAS volume) loss. Perhaps this helps explain Judith’s 3 items (as Rob van Dorland has asked her to clarify).

Neven,

I can’t imagine that replacing, say, a 60% albedo surface over much of the surface at high latitudes with a 20% albedo surface will have no impact on the general circulation (and thus have far-field impacts, perhaps on snow cover or blocking patterns, which will obviously propagate into some sort of mid-latitude response). However, these are very much research-level questions, and the links to changes in future weather patterns are pretty much speculative at best. Everyone likes to talk about the links to sea ice nowadays whenever a blocking pattern or extreme event happens, but even ignoring the issue of natural variability, there is currently no consensus on how wave heights or blocking frequency might change in a warming climate. Elizabeth Barnes and Dennis Hartmann at UWashington, for example, have some recent results from the CMIP5 models that seem to suggest the opposite of the Francis et al. hypothesis. I discussed some of the things going on in a blog post following up on Hurricane Sandy.

From a dynamics perspective, I can’t imagine what a “worst case” scenario might be, or whether it will be that big of a deal outside the atmospheric science community. It could be, but it’s difficult to envision just what that might be or whether Kerry’s precautionary advice even applies. From a global climate sensitivity perspective though, and especially with respect to science fiction views on methane catastrophes or runaway greenhouses, there’s no basis for thinking about the Arctic region as the difference between manageable and catastrophic levels of climate change. More sensitive models for example, don’t inflate climate sensitivity because of more ice loss, but usually because of more low cloud dissipation.

There is much more to worry about in the Arctic though with respect to biodiversity, the natives to the Arctic region, and even national security (e.g., increased access to exploration and the hunt for potential natural resources has already led to Arctic nations posting overlapping and disputed claims of territory).

In regards to the stability of the Arctic sea ice, I would agree with Ron that it may be more surprising that it has been so stable over our years of observations, both in our modern satellite record (since 1979) and earlier. Given that the ice quite thin (overall on average 2.5 meters or so, with multiyear being 3-4 meters), it doesn’t take much forcing, relatively speaking, to melt completely during summer (1 W/m2 or less).

So how unique is the current decline in a historical context, to the best of our knowledge? It’s a difficult question to answer, and I don’t think say too much with high confidence. And neither Judith, Ron, nor I are paleoclimatologists. It is a question we get asked frequently at NSIDC, so I’ve looked into research a bit, so I can give you my best understanding of the current knowledge.

Before I summarize my understanding, first, I would say that the more interesting issue that if the Arctic has been ice-free in the past, what would it mean for our interpretation of the current decline? There are (at least) two ways of looking at it:

1. If the Arctic has been ice-free during summer in the past, obviously it was due to natural forcing, so the current decline could also be due to natural forcing.
2. If the Arctic has been ice-free during summer in the past due to natural forcing then anthropogenic forcing of a similar magnitude will have a similar effect.

The first view is not implausible on its face, but it is simplistic because it doesn’t consider that the same result could be due to the same causes. Lightning starts forest fires, but that doesn’t rule out that a forest fire may be due to human actions.

The second view is much more useful in my view because it has potential predictive value at least for the equilibrium state of the ice cover under a given forcing. For example, in the Holocene (~8000 ybp – see below), the Arctic Ocean likely had ice-free or near ice-free summers and temperatures were similar to, maybe still a bit higher, than Arctic temperatures in recent years. Thus, the decline we’re seeing is entirely expected and we would expect to see it continue to near-zero summer extent in the coming years. The timing is still uncertain but it changes things from a “if the Arctic loses summer sea ice” to “when the Arctic loses summer sea ice”.

Now to summarize my understanding of the pre-satellite historical record:First, pre-1979, we have information from sea ice charts created by various countries. These were (and still are) produced primarily to support ships in the area. They were done primarily manually (and still have a strong manual component in terms of integration of different satellite sources) using ship observations, aerial reconnaissance and, starting in the 1960s, early satellite data. The data are not complete and often a fair amount of guess-work, albeit guess work by experienced analysts. The focus was not on making a consistent climate time series, merely doing the best chart possible at any given time. Since about 1953, the coverage has been reasonably complete – enough to create monthly fields. This was original done by John Walsh and colleagues at the University of Illinois. This has been stitched together with the satellite record by people at UK Hadley Centre, but not in a particularly consistent manner.

Forgive me for a brief shameless promotion, we’ve just published a homogenized version of this product that is consistent our NSIDC satellite record: http://www.the-cryosphere.net/6/1359/2012/tc-6-1359-2012.html (open access), which includes background on these and other sources of 20th century Arctic sea ice extent.Before the 1950s, the 1930s are often mentioned as a warm period. However, this is primarily in the Atlantic region, where observations were more common. Ice charts from the Denmark (http://nsidc.org/data/docs/noaa/g02203-dmi/) and Russia, indicate some periods of low summer ice, but on a more regional scale than we see now.Before that, other than sporadic, very limited records, we reliant on paleo records (sediment cores, ice-drifted debris, etc.). This information is as one might expect extremely sparse, both in space and time. A synthesis by Kinnard et al. (2011) indicates the current decline is unique in the last 1450 years. A review of longer records (Polyak et al, 2010) indicate that the last time the Arctic had little or no summer ice was during the Holocene Thermal Maximum, ~8000 ybp.

Happy Thanksgiving to all the U.S. readers!

Kinnard, C., C.M. Zdanowicz, D.A. Fisher, E. Isaksson, A. de Vernal, and L.G. Thompson, 2011. Reconstructed changes in Arctic sea ice over the past 1,450 years, Nature, 479, 509-512, doi: 10.1038/nature10581.

Meier, W.N., J. Stroeve, A. Barrett, and F. Fetterer, 2012. A simple approach to providing a more consistent Arctic sea ice extent time series from the 1950s to the present, The Cryosphere, 6, 1359-1368, doi:10.5194/tc-6-1359-2012.

Polyak, L., and several others, 2010. History of sea ice in the Arctic, Quaternary Science Reviews, 29, 1757-1778, doi:10.1016/j.quascirev.2010.02.010.

Similar to Lennart van der Linde, Neven, and various other commenters, I am surprised how much the experts (specifically Dr. Curry) address uncertainty in modeling “natural variability” (AMO, clouds, etc) while barely touching on any feedbacks that could explain the difference between climate model projections and reality.

In general, the emphasis by Dr. Curry on uncertainties over explaining long term trend in the Arctic ice decline seems rather misplaced. On a linear trend over the past 30 years, the standard deviation of Arctic sea ice is something line 500 k km^2, and on a quadratic down trend it is even less (something like 300 k km^2 IIRC). At the same time, the models (CMIP3/CMIP4/CMIP5) are pretty good in reproducing this uncertainty, but fail miserably on the long term trend (underestimated by 20-40 years).

So, why is Dr. Curry talking so much about uncertainties (such as the influence of clouds) while the real problem seems to be that models underestimate the long term (AGW) trend ?

Since ice extent modeling shows more variability than ice volume modeling, there seems to be a consensus among the experts that ice volume is the best way to assess multi-annual developments in Arctic sea ice, when it comes to separating natural variability and long term anthropogenic trend. Dr. Lindsay even explicitly mentioned that modeling with ice volume shows a much earlier divergence that modeling extent :

The CCSM3 model, for example, shows a clear separation in ice volume between the control and the A1B scenario as early as 1985, 10 to 15 years before the separation in ice extent (Schweiger et al 2011).

Indeed it is clear that ice volume development has even less “natural variability” than ice extent :
http://psc.apl.washington.edu/wordpress/wp-content/uploads/schweiger/ice_volume/BPIOMASIceVolumeAnomalyCurrentV2.png?%3C?php%20echo%20time%28%29%20?

This multi-decadal development is amazingly consistent, and relentlessly downward. It is hard to find any “natural variability” due to “uncertainty” in clouds, in this graph.

Which brings us back to feedbacks, negative and positive.
If the negative feedback of faster ice growth in fall (which is very nicely explained by Frank) indeed would be significant, than should we not see a ‘leveling off’ of this volume anomaly graph ?
Then why is it that it seems that instead of leveling off, ice volume is declining even faster (about 2X) than before the major extent losses occurred (since 2005).

It seems to me that ice extent and ice volume graphs both indicate that positive feedbacks in the Arctic are still stronger than negative feedbacks at this point, and one may ask when that would change. Could it be that currently Arctic sea ice volume is simply not stable, and will continue to decline until the Arctic warms up substantially ?

The next comment is from Gavin Schmidt, copied from RealClimate.org:

“So . . . what is the bottom line on the attribution of the recent sea ice melt? My assessment is that it is likely (>66% likelihood) that there is 50-50 split between natural variability and anthropogenic forcing, with +/-20% range. Why such a ‘wishy washy’ statement with large error bars? Well, observations are ambiguous, models are inadequate, and our understanding of the complex interactions of the climate system is incomplete.”

This statement is meaningless. The number you are trying to estimate is uncertain and so has to be characterised by some kind of distribution (which is to be estimated). A statement of likelihood is a quantification of the area under that curve between some limits (see http://www.realclimate.org/index.php/archives/2012/01/the-ar4-attribution-statement/ for examples). This means that the the likelihood of any single value (i.e. 50% or whatever) is actually zero. Rather, statements need to be couched as follows: the probability of the number being greater than 50% is XXX, or of lying between this range is YYY etc.

When will we be allowed to employ avatars? They’re helpful.
[CD: they are now implemented]

Dr. Meier said in his comment on ‘November 22, 2012 at 2:00 am’ that there is paleoclimatic evidence “that the last time the Arctic had little or no summer ice was during the Holocene Thermal Maximum, ~8000 ybp”.
That would make the near future Arctic ice free summer – not an “if”, but a matter of ‘when” according to Dr. Meier – pretty unique on the human time scale.

Logically there is a relation for long term variations in the amount of sea ice with temperature. Looking at some temperature reconstructions e.g. Vinther 2009, Richard Alley’s GISP2 ice core data or this well-known Wikipedia picture, it seems that the temperature is dropping since about ~8000 years ago. ‘Something’ changed this long term temperature decline in the Arctic area around 1900 as is visible in the graph of Kaufman 2009. As they put it:
“The cooling trend was reversed during the 20th century, with four of the five warmest decades of our 2000-year-long reconstruction occurring between 1950 and 2000.”
When on longer timescales the Arctic gets warmer, the amount of ice decreases. Or the other way around. The multi-decade warming trend in the Arctic is about 0.47 °C/decade (UAH since 1979): even more warming = even less ice. Seems very logic to me.

What is the opinion of the experts about this long term relation of temperature in the Arctic and the amount of ice?

In my last post, I expressed Modern Man’s ability to solve problems.
Perhaps I should elucidate what I meant by Modern Man’s unusual talents.
And, that is the ability to think in terms of:
“ Abstruse Abstract Thought. “

My dictionary describes abstruse as meaning: “ Remote from ordinary minds, difficult to be comprehended or understand, profound, recondite. . Hidden from mental perception by ordinary mentality. “ And I add: “ Combining broad and detailed analysis of abstract objects to elicit new concepts without reference to existing concepts. The concept of ideas for bringing order from nothingness. “ Conceiving something new not having a pedigree.

My dictionary describes abstract as meaning: “ Considered or thought of in itself; treated apart from any particular object. ” And, I add: “ A mental process using symbolic objects to arrive at meaningful conclusions. The mental processes where complex ideas are represented by symbolic objects manipulated within the mind to give real world meaning to a process or relationship. “

I am sure you have other thoughts on this concept.

Thank you,

Everette L. Wampler

@Rob Dekker, Nov 23th

I did address my open question – without any form of prejudice – to Walt Meier for the simple reason that he introduced the topic in his opening post. In another post, he listed some references. I’m not trying to debunk any of them, as you suggest in a rather insulting manner. I merely wanted to point out that they seem to contradict with the US-CCSP study. In my turn, I’m tempted to suggest that you debunk that study because it doesn’t confirm with ‘your beliefs’. But I won’t. Let’s keep Climate Dialogue a platform for open minded people. 

Bert Amesz,

I am sorry that my tone felt “insulting” to you. It was not intended as such.
I was disappointed however that TWICE in this blog post your brought up the claim that :

between 750 and 1500 AD parts of the Arctic were open, at east during some summers, while during LIA sea ice extent was excessive.

Where you use page 176 of the US-CCSP, 2009 report as your evidence, but so far you did not engage in a scientific dialogue about this statement.

I suggested that :

I think that the authors do not present enough evidence to claim that the Northwest passage was open during some summers over that period.

What do you think ?

My main argument was that the 5 species of bowhead whales dated to around 1000 B.P. in the Beaufort sea is not enough evidence to conclude that the NW passage was open. In fact, the graphs from the US-CCSP 2009 report suggest that bowhead whale bones have been found through all three areas of the Arctic dated over the past 5 thousand years, with a small “west” (towards the Beaufort) concentration during 1000 B.P. and a small “eastern” (towards Baffin Bay) concentration 250 years later, with no significant change in bowhead whale bone finds in the Central Canadian Archipelago corridor over the past 6000 years.

Secondly, the scientific papers mentioned as the original source of the data are Dyke et al 1996 and Savelle et al 2000.
Neither one of these studies supports the US-CCSP conclusion of an open NW passage around 1000 AD. In fact, Dyke et al 1996 states :
http://arctic.synergiesprairies.ca/arctic/index.php/arctic/article/view/1200/1225

At 1 ka B.P., the bowhead range was similar to that at 2 ka B.P.; the range appears to have expanded in places and to have retracted in others (Fig 19). The pattern of occupation of Prince Regent Inlet seems to have switched: earlier, whales seem to have occupied the east side but not the west; later, the west side but not the east.

which suggests regional climate changes that shifted sea ice back and forth (east and west) during that time, but no suggestiond of an “open” NW passage at all.
Lavelle et al 2000 barely mentions 1000 AD as special, apart from this remark :

Although the range apparently expanded slightly at approximately 1000 B.P., this occurred during an otherwise overall reduction in the bowhead range relative to the 5500-2500 expansion.

http://arctic.synergiesprairies.ca/arctic/index.php/arctic/article/view/871

which also suggests regional changes in an otherwise cooling pan-Arctic climate.
So the US-CCSP 2009 page 176 conclusions seems not sustained by the original data it quotes, and that is why I suggested that they do not have enough evidence to sustain their statement.

Thirdly, bowhead whales are amazing marine mammals, perfectly adjusted to Arctic sea ice, and I would not be surprised if they have been crossing the NW passage continously over the entire Holocene, through the LIA and MWP and any other regional or pan-Arctic climate change periods too.

Lavelle et al 2000 makes another statement that is interesting in this respect :

Regarding predation, note that although we have recorded thousands of bowhead bones and tens of beluga, narwhal, and walrus bones in the CAA, we have yet to record one bone of a killer whale, the only known predator of bowhead whales besides humans.

After all, sea ice is their nich, where the hide from their predators (killer whale) and reach feeding areas where no other whale can get to. They are quite literally the living “ice breakers” of the Arctic, and over the past 10 million years even evolved an immensely strong head bone which allows them to break through sea ice of 2 feet thick. The Arctic is their home, and noone else comes even close.

So, I think they have been traveling through the NW passage over the entire Holocene period.

What do you think ?

Rob and Bert,

Thirdly, bowhead whales are amazing marine mammals, perfectly adjusted to Arctic sea ice, and I would not be surprised if they have been crossing the NW passage continously over the entire Holocene, through the LIA and MWP and any other regional or pan-Arctic climate change periods too.

Indeed. Bowhead whales probably don’t need a completely open NWP to make it from one end to the other, or intermingle in between. All they need is leads and small polynyas to breathe.

In 5 of the last 6 years the NWP has been completely open, with ships, yachts, catamarans and even kayaks going through. Nowadays there are boats making the passage without reinforced hulls! The last two melting seasons saw several expeditions going through the NWP and the Northern Sea Route in one season, effectively circumnavigating the Arctic sea ice pack and North Pole.

There is very little evidence that this has happened since the Holocene Climatic Optimum, when 8% more solar radiation (+40 W/m2) reached the Northern Hemisphere in summer (that is not the case in the current Rapid Ice Loss Event). The only evidence, ie remnants of bowhead whales, that can be and is drummed up by Bert Amesz is weak at best.

Rob Dekker (and Neven)

Rob, apologies accepted and thanks for your comprehensive reaction. Again: I wasn’t ‘claiming’ anything, just pointing out some puzzling discrepancies between the various studies. But you convinced me that studies on the basis of bowhead whales only, lead to inconclusive results. I’m not a scientific researcher, just a simple – but unbiased – hydrogeologist who tries to understand what’s happening in the Arctic. To me it’s clear now that the current decline, in extent and thickness, is unprecedented in recent history.

But I have another remark, now concerning LIA. I’ve read the Polyak paper and I saw the graph of Kinnard (2011). Only his graph, because the paper is behind a pay wall. Kinnard concludes a relatively low extent during LIA, while Polyak shows the opposite (which looks logical to me). Moreover, Kinnard’s current decline seems dazzling compared to Polyak. Hence, quite different results. Is this also due to the proxies they use?

P.S. Here is a non-paywall version of Kinnard et al 2011 :
http://www.geotop.ca/pdf/devernalA/Kinnard_et_al_nature_2011.pdf

Rob Dekker

Thanks for the copy of Kinnard et al 2011. Be assured that I understand the difference in ‘Arctic’ and ‘Nordic Seas’ ice extent. What I mean is the following. In his paper (2010, fig 12), Polyak tries to match the ice extent in the Nordic Seas (Svalbard, Scandinavia) with the findings of Kinnard (2008) regarding Arctic ice extent over the period 1870-2003. He concludes ‘a close match, except for an obvious discrepancy in the early 20th century ‘. So far, so good.

Now have a look at another period: LIA. From Polyak (Fig 12) one may conclude that, in Nordic Seas, the ice cover was relatively extensive. Which seems the opposite of the findings of Kinnard (2011, Fig 3a) regarding late-summer Arctic extent. According to Kinnard (2011), ice coverage during LIA was even lower than during the Medieval Warm Period. He concludes that the low ice extent during LIA (and the Dark Ages Cold Period) was due to an increased northward advection from warm Atlantic water into the Arctic. Nevertheless: while Polyak was looking for a match between Nordic Seas and the Arctic for a more recent period, it appears to me that such a match doesn’t exist for LIA. I suppose that an increased advection of warm Atlantic water would have affected the ice extent in the Nordic Seas as well? Unfortunately, Kinnard (2011) doesn’t compare his findings with the ones of Polyak (2010).

Then there is another striking difference between the two graphs: the Polyak paper (fig 12) shows a more gradual decline (with ups and downs) since early 19th century (end of LIA), while according to Kinnard (2011, fig 3c) the decline starts more than a century later, mid 20th century. Hence the word ‘dazzling’ in my previous post.

So while the bowhead whales have been discarded, I’m still stuck with the fact that two – I assume – eminent research groups come up with different findings with regard to an important topic: natural variability.

Bert, maybe you should ask Kinnard and Macias-Fauria what it is exactly they did. For which regions? For which parts of the year (max, min, whole year)? You also might want to check eventual differences between Kinnard 2008 and Kinnard 2011. And be careful not to mix up Polyak and Polyakov.

Polyak was looking for a match

(…)

his findings with the ones of Polyak (2010)

Polyak et al. didn’t look, there are no findings. In their paper Polyak et al. describe the history of Arctic sea ice, and towards the end compare the findings of two papers to show “the pan-Arctic nature of the modern ice loss”.

I don’t know what it is you would like to get at, but I’m amazed at how much info you seem to get out of eyeballing a couple of graphs and reading the subscript. Maybe you could teach me that. 🙂

What you also might find interesting from that final paragraph is the following:

The remarkable modern warming and associated reduction in sea-ice extent are especially anomalous because orbitally-driven summer insolation in the Arctic has been decreasing steadily since its maximum at 11ka, and is now near its minimum in the precession cycle

Which is why some people here asked why accumulated heat due to GHGs could not be responsible for more than 100% of the recent rapid ice loss.

@Neven – Nov 25th

As to your last remark, I found an interesting paper from Spielhagen et al (2011) regarding the temperature of Atlantic Water in Fram Stait during the past 2000 years. He reports a rapid and unprecedented temperature increase since 1850. Here the link for (the paywalled) paper.

http://www.sciencemag.org/content/331/6016/450.abstract

Here another (paywalled) paper from Cronin et al (2012) on Arctic Ocean warming during the last glacial cycle.

http://www.nature.com/ngeo/journal/v5/n9/full/ngeo1557.html

PS: ‘One picture is worth a thousands of words’, is a saying.

Bert, do you have a copy of the Spielhagen paper?

@ Rob Dekker

I googled on ‘heat flux fram strait’ and found this paper from Schauer et al (2004):

http://ftpout.npolar.no/oan/backup/Undervisning/UNIS/U.Schauer_Fahrbach2003JC001823.pdf

During 1997-2000 they measured the heat flux in Fram Strait and report a strong increase in the West Spitsbergen Channel. The net increase amounts to some 25 TW and its due to two equally important factors: stronger flow and higher water temperatures. In conclusion they suggest ‘ that a warming signal from the late 1990s is presently spreading in the interior Arctic Ocean’.

Let’s assume that the net increase of 25 TW is uniformly divided over the Arctic north of 80°. This would result in 5 W per square meter which is more than the radiative effect of GHG’s. So your conclusion of ‘100% AGW’ is a little premature. But I agree with your conclusion that Arctic sea ice is being attacked from top (atmospheric conditions) and bottom (warm ocean currents).

It would be nice to know whether the Schauer-measurements had a follow up. I would think that permanent monitoring would be useful.

Formula’s are ok but I do not succeed in producing a proper table. How can we have a proper discussion then? For those who want to know the proper figures, please send me an email at andries.rosema@ears.nl

Ron, Walt and Judith,

While writing a summary of the discussion so far, the following question came to my mind (you all already made remarks related to this, but I would like to focus more):
since you all indicate it is very difficult to really prove a significant contribution of GHGs (or global warming) to the decline in Arctic Sea Ice, I wonder to what extent you agree with the reasoning that observed natural variability since 1950 cannot explain the decline as observed in past decades (see for example in Notz (2012) who concludes that: ‘We find that the available observations are sufficient to virtually exclude internal variability and self acceleration as an explanation for the observed long-term trend’). This reasoning leads to the conclusion that observations alone are sufficient to prove there must be an anthropogenic factor in the current decline of Arctic sea ice.

So the question is: To what extent do you agree with this reasoning? (maybe also taking into consideration the two studies mentioned by Rob that the current decline is unprecedented in the past 1450 years and the study from Spielhagen (2011) that shows a 2 C increase in temperature of Atlantic Water entering the Arctic Ocean during the 20th century, unprecedented in the past 2000 year.)

The links that Mr. Cripwell pointed to are respositories of data which have been peer reviewed and the algorithms for processing new data coming down from satellites have been peer reviewed as well as having multiple levels of peer review for the construction and validation of the instruments, satellite tracking and more.

That the Arctic refreeze is covering the surface with a thin layer of new ice is no surprise. One simply needs to have followed any discussion about first year ice increasingly dominating the Arctic ice pack as the volume falls rapidly or peeked in at any of those sites mentioned by Mr. Cripwell.

Some things are obvious

As obviously, the variation of maximum ice extent in the Arctic is an indicator of conditions outside the Arctic Ocean basin, places such as Hudson’s Bay. That ice will always melt in the summer, so it has little to no effect on minimum ice extent.

Bert Amesz,
Thank you for your response.

I think that if your assessment of 5 W/m^2 increase in ocean heat flux is correct, then it would likely explain the dramatic multi-decadal decline in Arctic sea ice volume (some 75% at the September minimum over the past 3 decades) as registered by the PIOMAS model :
http://img.photobucket.com/albums/v224/Chiloe/12_Climate/sea_ice_PIOMAS_min.png

Note that this PIOMAS volume overview is still right on track to hit “ground zero” volume around 2020, and there seems to be no “negative feedback” pushing back at the accelerated rate of ice volume loss.

Unfortunately, from a scientific point of view, it is not so easy to make that claim, nor to blame increased ocean heat flux for this decline. Allow me to explain :

For starters, it may be good to realize that from basic physics we know that ocean heat flux affects thick ice (MYI) especially in winter much more than thin ice (FYI). The thicker the ice, the more it insulates, and thus the less bottom-freeze occurs during winter, but without ocean heat flux it will still grow a bit each winter. In fact, without any ocean heat flux the Arctic ocean would freeze solid all the way to the bottom eventually. Think about it. However, if there is any ocean heat flux, even if the atmosphere is below zero, thick ice can still be bottom-melting while thin ice is growing. So ocean heat flux reduces thickening of MYI, and shortens MYI freezing period as well. A double whammy as they say here in the US. Thus, if ocean heat flux increases (over the decades), the MYI gets thinned the most, which affects ice volume, but not ice extent. Since this (ice volume reduced, MYI reduced, while ice extent is still hanging in there) is what we have seen over the past 3-4 decades, as witnessed by the PIOMAS volume results, I believe that ocean heat flux may have played an important role in “preparing” the Arctic sea ice for what now seems to be an imminent collapse.

Secondly, it is not certain that all heat that gets injected underneath Arctic sea ice through the West Spitzbergen current actually will contribute to ocean heat flux that melts MYI from beneath. The ocean underneath sea ice in the Arctic is strongly stratified, which can prevent deep ocean heat from reaching the ice cover. In fact, the paper you quote from Cronin et al (2012) on Arctic Ocean warming during the last glacial cycle suggests that during the last glacial cycle, this stratification was so strong that virtually none of the medium depth heat inserted from the Atlantic made it to the surface. Thus, ocean heat flux must have been very low, and basic physics thus suggests that sea ice must have been very thick during the last glacial period.

When trying to model this stratification, and the resulting ocean heat flux, the problem appears to be inherently non-linear. In fact, modeling ocean heat flux from below the halocline in a stratified Arctic has been an immense theoretical challenge which has not been solved yet, and may never be. The main problem is that the heat flux is dictated by turbulence and thus the Navier Stokes equations, which in 3D do not even have a mathematical solution, even if you have the boundary conditions.

Needless to say, modeling ocean heat flux has been a real challenge, and a more empirical approach may be more useful in the case of Arctic sea ocean heat flux. This is why I asked Dr. Lindsay to clarify how ocean heat flux is incorporated in the PIOMAS model.

In absence of clarification on PIOMAS, a “must read” paper is Polyakov et al 2010 (“Arctic Ocean Warming Contributes to Reduced Polar Ice Cap”), which spends a great deal on the issue of ocean heat flux.

Most sites show only the abstract, and charge for the paper itself, but the great people from Woods Hole Oceanographic Institution (WHOI) provided not just the Ice Tethered Profiler (ITP) buoys that supplied data for this paper, but were also kind enough to share the paper with all of us :

https://darchive.mblwhoilibrary.org/bitstream/handle/1912/4345/2010jpo4339.1.pdf?sequence=1

One quote from the paper, which seem applicable in the context of your assessment (5 W/m^2), I think you are in the ballpark of what analytical models suggests, although you maybe a bit on the high side. From Polyakov et al :

Estimates of upward heat flux from the AW yield values ranging from 1.3 W m^2 (analytical model, Rudels et al. 1996), to 2.1 W m^2 (1D mixed layer model, Steele and Boyd 1998), to 3-4 Wm^2 (parameterization for the bulk heat transfer coefficient from the mixed layer temperature, Krisfield and Perovich 2005), to as high as 4-6 W m^2 (heat budget estimates for the eastern Nansen Basin, Walsh et al. 2007).

Another quote, one of the most comprehensive statements from this paper, cutting through all the theoretical deliberations, is this one :

Until recently, maximum (~2-3 C) AW temperatures have been found in the Nansen Basin, while in the Canadian Basin the AW temperature has remained near 0.58C (Fig. 2a). This decrease of AW temperature with distance from its Arctic entrance region implies that AW heat must be lost along the AW spreading pathways.

which suggests that at least a great deal of the increasing AW heat makes it to the bottom of the Arctic sea ice, thus indeed contributing to ocean heat flux. Especially check figure 2 in the paper, since it suggests that even as recently as the past decade, warming AW has increased the temperature of the deep water under the Arctic basin (where the thick MYI ice hangs out )quite dramatically, while there does not seem to be much of increase in the temperature of the out flowing ocean waters. Thus, much of this heat may have been ‘lost’ to the surface and thus melting thick ice…

Also, Polyakov et al’s own modeling (which may very well be an underestimate) suggests that 0.5 W/m^2 increase in ocean heat flux over the past 50 years, which they then used in a 3D model and found that it causes a 28-35 cm of ice thickness reduction, which is something like 30% of the ice volume decline.

Unless we understand how PIOMAS models ocean heat flux, we may not know if that assessment is realistic or not.

@Bob Brand

My remark referred to the ‘100%’ only, not to AGW as such. As I stated, Schauer reports an increased heat flux due to (on a 50/50 basis): (i) increase of flow velocity and (ii) higher water temperatures. You can’t attribute both effects for 100% to humans. I think that a significant part of the increased heat flux is caused by natural variability in AMO and/or AMOC.

Nice post, Dr Curry

What I miss in the discussion on polar ice melt, is the following.
Sea water is salt, contains sufficiently to decrease the freezing pint by 5 to 6 degrees Celsius.
Refreezing requires temperatures well below zero, so you will get ice derived from salt water. Although a certain amount of the salt will be expelled and not incorporated into the ice, this ice is also likely to thaw earlier than ice, composed of snow, which does not contain any salt. This ice will thaw at 0 degrees, and freeze at 0.

As far as I know (and I do not know everything), this has not been addressed in the various studies about thawing of the Arctic.

Please, your opinion.

Bert Amesz said

Schauer reports an increased heat flux due to (on a 50/50 basis): (i) increase of flow velocity and (ii) higher water temperatures. You can’t attribute both effects for 100% to humans.

Of course you can’t. Schauer analyzed a 3 year period (1997-2000), which is decidedly in the short term variability noise.
Why do you think this study is relevant for human attribution to multi-decadal Arctic sea ice decline ?

I think that a significant part of the increased heat flux is caused by natural variability in AMO and/or AMOC

Which evidence do you have to sustain that opinion ?
According to Day et al 2011, AMO/AMOC (as well as PDO, AO, NAO) can explain AT MOST 30% of the observed multi-decadal Arctic sea ice decline, with the remaining 70 % anthropogenic.
And they use the extremely conservative assumption that influence of anthropogenic global warming on Northern Atlantic SSTs is only linearly increasing over the past century. Which is certainly an underestimate of anthropogenic GHG influence.

In my contribution of 29 November I have argued, using radiation balance calculations, that increased rates of melting in the arctic are predominantly due to an increase of cloud cover. With Meteosat we observe a 10% increase of cloud cover since 1982. This observation is consistent with the increase of rainfall reported for the same period by the IPCC 4th Assessment Report.

My contribution was however difficult the read because the table collapsed. Therefore (with thanks to Theo Wolters) here is a link to the correct text and table: http://www.nlslash.nl/climatetheory/seaicemelting.doc

The theory put forward predicts a simultaneous negative radiative forcing and decrease in surface temperature in the Atlantic. This is confirmed by our Meteosat surface radiation temperature data, also since 1982.

I would appreciate feedback from the experts and public on this view.

Bert Amesz said :

I think that a significant part of the increased heat flux is caused by natural variability in AMO and/or AMOC
…
Mahajan et al 2011: AMOC no correlation with summer ice extent, but the strengthened AMOC could have contributed to decline of winter ice.

Thank you for mentioning Mahajan et al 2011, the findings of which were confirmed by Day et al 2011. It’s pretty clear that both studies place the influence of warming Atlantic waters on Arctic sea ice in the Northern Atlantic marginal seas (Nordic Seas and West Greenland). Of course, Arctic sea ice does not know if that warming is caused by the “natural variability” of AMO, or simply because the entire Northern Atlantic is warming up with the rest of the planet. So even that 30% “long term variability” will have an anthropogenic factor.

Either way, if “natural variability” of AMO/AMOC only correlates well with winter sea ice decline, one may wonder what caused the jaw-dropping summer sea ice decline over the past 3 decades, no ?

I would like to consider the role of clouds in the Arctic Sea Ice decline a little bit more.

We probably agree that there is a huge annual cycle in the Arctic in solar insolation, temperature (hence IR fluxes), cloudiness, latent and sensible heat fluxes. There are many feedback loops between these entities. Yet, we observe a steep decline in sea ice volume already starting in the 80s. IMO this suggests that there are first order forcing mechanisms (perhaps only one dominant), which trigger the observed decline, possibly via positive feedbacks (e.g ice albedo).

As we agreed that most of the melt (2/3) is from below and the remaining (1/3) is caused by the atmosphere, the dominant forcing mechanism is the increase of ocean heat content, in particular the increase in the upper layers. (we may discuss the reason for this, but a good candidate is anthropogenic forcing).

Considering the smaller (1/3) effect of changes in the atmosphere on the decline:
Judith stated in her contribution (at the 18th of November): ‘Clouds are much more powerful radiatively than CO2, so if we are talking about radiative forcing, clouds should be front and center in the discussion.’ However, as stated in her guest blog there is a strong seasonal cycle in the effects of clouds, warming in the winter and cooling in the summer.

So, my first remark is that there will be months where the CO2 (surface) forcing is more important than cloud forcing as the sign of the latter reverses during the year.

More importantly, is there agreement that latent heat changes related to the decline in sea ice extent is much larger than (radiative surface) forcings due to cloud and CO2 changes?

Thirdly, might there be a negative feedback (in summer and autumn) of enhanced latent heat release, hence more (low) clouds, implying less forcing from above. If so, it will be difficult to believe that changes in cloudiness is a main driver of the decline.

I would be very pleased if you (all) can comment on these points.

@ Rob Dekker

One may argue as follows. In wintertime, the increased AMOC/AMO affects the volume (and extent) of multiyear sea ice. See also NSIDC:

http://nsidc.org/images/arcticseaicenews/20111004_Figure6.png

So at spring, the starting conditions deteriorate gradually. The spring sea ice extent is increasingly dominated by one year ice. Subsequently, this thin ice disappears rapidly due to atmospheric conditions in spring and summer. This would imply an indirect relation between AMOC/AMO and summer ice extent. The AMO, often seen as fingerprint of AMOC , is a natural climate variability. Also during the fifties, the AMO was positive and sea ice extent was low. But I agree with you: in addition to the natural AMO-variability, there is an overall warming trend in North Atlantic waters.

Jim Cripwell, allow me to be blunt: Until such time as you drop the clearly-partisan terms “warmist” (though I’m happy to wear the tag, inasmuch as it is WARMING) and the utterly inaccurate and non-scientific term “CAGW,” which is NEVER used outside the disinformation camps. nothing you say can be taken as anything but a diatribe borne not of scientific rigor, but of partisan and frankly, unsubstantiated claptrap. It does not become the style of an honest broker, not of a true skeptic.

Harry Wiggs, Fair enough. As I have explained before I am perfectly willing to use any terms that anyone finds acceptable. If you will let me know what you would use as a substitute for “warmist”, and “CAGW”, I would be delighted to use them. With respect to “CAGW”, “AGW” does not describe what I am talking about. AGW exists, but it’s effect, from what little empirical data we have, is negligible. I agree that AGW is real, but I do not accept that what I call “CAGW” is real.

Thanks Ron for your reply. I thought there was consensus about the 2/3 ocean and 1/3 atmosphere issue in relation to ice melt. Walt brought this up referring to Steel at al.

I think you made a very relevant statement: ’Because they [clouds] are so important I would not be surprised if they are found to be at least part of the source of the ice melt, but because cloud temperature and properties may change due to changing surface properties, sorting out cause and effect could be difficult.’

Therefore let’s consider first order feedbacks in the Arctic due to ice decline (without pointing to the initial cause):
1) ice – albedo feedback:
less ice -> more SW absorption by the ocean -> higher ocean temperatures -> more ice melt
2) Latent Heat – Cloudfeedback
Less ice -> more LH -> more (low) clouds -> positive or negative feedback depending on surface albedo and season
3) Snowcover – Arctic temperature feedback
Less ice -> higher temperatures in the Arctic -> Less snow cover over continent surrounding the Arctic –> higher temperatures in the Arctic region (positive feedback). Or it might as well be a negative feedback if higher temperatures in the Arctic result in more snow cover.

I would appreciate if we could make a list of first order feedbacks in the Arctic region and rank them in terms of sign and strength. I also invite Judith and Walt to give ideas and comments on this.

Dr Lindsay said

I don’t agree that 2/3 of the melt comes from the ocean. How do we know that?

to which Rob van Dorland replied :

I thought there was consensus about the 2/3 ocean and 1/3 atmosphere issue in relation to ice melt. Walt brought this up referring to Steel at al.

I think there was a misunderstanding here. Steele et al 2010 refers to the diurnal cycle of Arctic sea ice melt, whereas I think Dr Lindsay refers to the multi-decadal loss of volume recorded by PIOMAS.

The two (estimates of ocean heat melt versus atmospheric heat melt) may indeed be completely unrelated.
After all, bottom-melt in the ice margin during the summer melt season mainly originates from ocean warming due to the change in albedo in the ice margin (ice to dark ocean), and thus both bottom and top melt originates from the atmosphere (and Arctic summer sun).

In contrast, for multi-decadal volume loss, ocean heat (and bottom melt) typically refers to heat flux through the cold halocline in the Arctic, caused by influx of Pacific and Atlantic water underneath the ice pack. This ocean heat flux affects mostly thick multi-year ice during winter (since heat flux through thick is in winter is much smaller than through thin ice), and thus is of a completely different nature than the diurnal cycle bottom melt in the ice margin in summer.

To address Dr. Lindsay’s question “How do we know that?” regarding bottom melt due to increased ocean heat flux from Atlantic and Pacific waters, the best scientific estimate that I have found is (once again) Polyakov et al 2010 :
https://darchive.mblwhoilibrary.org/bitstream/handle/1912/4345/2010jpo4339.1.pdf?sequence=1

They suggest that over the past couple of decades, ocean heat flux (due to Atlantic water temperature increase) has increased 0.5 W/m^2 under the ice, meanwhile noting that ocean heat flux through the halocline is mostly determined by turbulence, which is notoriously non-linear. Also, they note significant geographical differences of ocean heat flux :

there is a wide spread of flux magnitudes depending on geographical location. For example, Padman and Dillon (1987) estimated heat fluxes above the AW core (320-430 m) using microstructure profiles from the Beaufort Sea to be in the range 0.02-0.1 W/m^2. Heat fluxes in the Canada Basin at the base of the mixed layer and below ice were estimated as 0.3-1.2 and 0.2 W/m^2 ;over the Chukchi Borderlands the estimates were higher, 2.1-3.7 and 3.5 W m^2 (Shaw et al. 2009)….etc etc

Still, when inserting their estimate of an overall 0.5 W/m^2 increase in ocean heat flux in their 3D model, they found that it causes a 28-35 cm of ice thickness reduction, which is something like 30% of the observed ice volume decline.

Dr. Lindsay, can you please comment on these ocean heat flux findings by Polyakov et al 2010 ? Specifically, does the PIOMAS model confirm or deny their findings, and is their estimate (of a multi-decadal volume loss attributed to increased ocean heat flux due to warmer AW influx) in the ballpark of the PIOMAS ocean heat flux estimates ?

It would be REALLY interesting to understand PIOMAS’ findings/modeling of ocean heat flux over the past decades a bit better, and if anyone has performed any experiments with PIOMAS in varying ocean heat flux, to determine how sensitive overall volume decline is to increased Atlantic ocean water influx..

In the first place I would like to thank the individual scientists that are engaged here discussing the topics from the grounding of their specialties. To have you discuss from the position of the passion you have for your work in open forum is actually a rare access for the public versus the craft of science writing which distills down the science for the public pallet.

In your discussions, there seems to be a recurring theme of the limits of data based on the introduction of satellites in the late 70s and other limitations. I would suggest there is a vast trove of data that I have not seen referred to and this is the data from all the military traffic in the arctic by the Russians, the Americans and any other country with nuclear submarines, as these boats and crews would have a very strong interest in sea ice conditions above them and would have recorded huge amounts of data as they plied their various courses at varying times of the year.

Another point is the factor of changing salinity and its affects on ice formation and retention. As referenced in one of the abstracts I looked at.

http://www.nature.com/nature/journal/v481/n7379/fig_tab/nature10705_F1.html
http://www.nature.com/nature/journal/v481/n7379/full/nature10705.html

If arctic warming is a fact then the water flow of the feeder continents of Asia the North America through permafrost melting and all the other sources of runoff which could include glacial melt and the potential of changing rain volumes across the continents would seem likely to impact on ice formation and retention.

Then there could be the interactions and feedbacks from the weather. The cyclonic event of august 2012 was just one of many recorded cyclonic and anti-cyclonic events recorded over time. These events are related to the high pressure ridges and positions of the jet stream as it changes over the course of a year. A storm can have significant impact on the breakup of sea ice depending on the strength of the the storms systems at critical times in the melting phase of summer.
Regrettably I can not find the original abstract that referred to specific years of intense activity of these storm periods but this piece infers a direction.

http://link.springer.com/article/10.1007%2FBF01030491?LI=true

The impacts in the general ecology of the Arctic Ocean of the loss of sea ice can also be a venue to be taken account of. What impacts there are can be inferred from such events as drowning polar bears unable to reach ice flows for seal hunting as the distance to the ice is too far and to other aspects of the total ecology that perhaps can feed into ice formation and melting regimes, even on a microbial level. Life forms shape ecologies as they are shaped by them. What they are in the Arctic is likely subtle but likely not without impact.

In the dialogues going forward, the ability of the scientific community that focuses on the Arctic to communicate what is known and the possible futures will likely need to find new methods that could contribute to the public and policy makers awareness. One such project that is quite impressive is the Vulcan Project.

http://vulcan.project.asu.edu/

To create such granular visualizations of as many aspects of the complexity of the Arctic by the use of video is something that would, I believe, be a fantastic way to engage the vested interests of all the partners.

The summary can best be described as “all sound and fury; signifying nothing” (Shakespeare, Macbeth). Three confirmed, ardent, warmists write what they think, and the summary is supposed to mean something? I dont think so.

We now have a few more months of data with respect to both Arctic and Antarctic sea ice extent. The Antarctic has been setting maximum extent values for the date as the summer progressed, and looks likely to start the refreeze a little early, with the minimum extent being the second highest value since records began. I know there is a pal reviewed article which “explains” why CAGW has no effect on Antarctic sea ice. But then, the warmists have a pal reviewed article on any aspect of CAGW, where the empirical daya does not support the religious belief that CAGW is real. Then the warmists do their best to see that counter articles never get published; just look at what happened to Livingston and Penn, Roy Spencer, and Anastassia Makerieva.

Meanwhile, Arctic sea ice extent made a spectacular recovery from the minumum set in September 2102. The latest data shows that Arctic sea ice extent is just within the 2 sigma limit of the average of precious years. Further, the temperature north of 80 is significantly below average. This could well signify a late start to the melt of that ice which needs to melt if the Arctic is to be ice free next summer, or at least set another record for the minimum extent.

I presume that, sometime, there will be a pal reviewed article which explains why CAGW only affects Arctic sea ice extent in the summer. While the empirical data shows the dramatic loss of summer sea ice, it shows that winter sea ice extent is far less affected. I wonder why. But I am sure someone will invent a good reason, and get it published.

All in all, not a very promising start to this blog.