A few people have already remarked on some pretty surprising numbers in Arctic sea ice extent this year (the New York Times has also noticed). The minimum extent is usually in early to mid September, but this year, conditions by Aug 9 had already beaten all previous record minima. Given that there is at least a few more weeks of melting to go, it looks like the record set in 2005 will be unequivocally surpassed. It could be interesting to follow especially in light of model predictions discussed previously.
There are a number of places to go to get Arctic sea ice information. Cryosphere Today has good anomaly plots. The Naval Sea ice center has a few different algorithms (different ways of processing the data) that give some sense of the observational uncertainty, and the National Snow and Ice Data Center give monthly updates. All of them show pretty much the same thing.
Just to give a sense of how dramatic the changes have been over the last 28 years, the figures below show the minimum ice extent in September 1979, and the situation today (Aug 9, 2007).
The reduction is around 1.2 million square km of ice, a little bit larger than the size of California and Texas combined.
Update: As noted by Andy Revkin below, some of the discussion is about ice extent and some is about ice area. The Cryosphere Today numbers are for area. The difference is whether you count ‘leads’ (the small amounts of water between ice floes) as being ice or water – for the area calculation they are not included with the ice, for the extent calculation they are.
Update: From the comments: NSIDC will now be tracking this on a weekly basis.
Tired of waiting for NSIDC to report on their ice data? Get Google Earth and pick up your KML feeds from: http://nsidc.org/data/google_earth/
I see only one thing to say in view of this year’s ice extent: yikes!
The current record retreat in Arctic ice during the summer melt comes with reports of another endangered specie: cultural icon Santa Claus will relocate toy manufacturing and distribution operations to the Antarctic continent, along with his fleet of pole-adapted aeronautical reindeer. Due to increased average shipping distances, the traditional gift-giving will be required to be rescheduled to Boxing Day.
A tipping point, indeed.
Ah oui monsieur Phillippe! But who is it that you are saying Yikes too? I admit Ive only sent one Yikes email to a politician this week about this, which is slack of me I know. Ill try and do better!
If I sent a YikesMail every time I posted here, I may get a better result!
I limit my Yikes e-mailing to politicians due to my immigration status, which, however legal, does not allow me to vote. I guess that perhaps I should send more yikesing, since all the mail they send me to court my non-vote indicates that they probably don’t know anyway…
Does anynody have a link on the met office Polar ice long term animation presentation? Once seen on their main page….
“Yikes” would imply some sort of deep seated surprise. The only surprise here is that weather is a chaotic system and there’s no way to predict when conditions will come together and force a systematic state change. The presence of a catastrophic boundary has been obvious for years — and the only question has been when it would be encountered.
I’m not sure we have encountered that boundary, though it sure looks likely. We wait to see what sort of recovery the ice makes this winter. If the recovery’s more than a half million below previous years then barring several major explosive eruptions the ice cap is history. If the recovery is fairly complete then there will probably be several years of vacillation first. (Climatologists
need to spend more time studying ecological collapse and less time studying physics. This is a chaotic system they are trying to model, not billiard balls.)
Weather is chaotic; climate isn’t.
http://illconsidered.blogspot.com/2006/03/chaotic-systems-are-not-predictable.html
Re #257 [The only surprise here is that weather is a chaotic system and there’s no way to predict when conditions will come together and force a systematic state change.]
Actually it’s not a chaotic system.
A standard definition of chaos is given in
Davaney, R., Introduction to Chaotic Dynamical Systems 2nd Ed,
Addison-Wesley, 1989:
A function f:R -> R exhibits deterministic chaos if it
satisfies three properties:
1) sensitivity to initial conditions arbitrarily close to every
point x, there is a point y with f^n(x) and f^n(y) iterating far apart.
2) dense periodic points arbitrarily close to every point x, there is a
point y with f^m(y) = y for some m.
3) mixing for every pair of intervals I and J, for some k f^k(J)
and I overlap.
This means that if “chaos” is used in the technical sense, a chaotic system must be closed (and deterministic), while Earth’s climatic system is thermodynamically open. A better term in this case is a “dissipative system”: one that feeds off a low-entropy source of energy (in this case the sun) to self-organise, giving off higher-entropy energy in the process. Such systems do indeed often show the kind of “systematic state change” you talk about. However, the general direction of that change in response to certain kinds of external influence (in this case adding greenhouse gases) can be quite predictable, while the exact path taken may not be.
Re 258″ “Weather is chaotic; climate isn’t.”
Weather is chaotic, and therefore climate has chaotic boundary conditions. That is, you can’t predict when a catastrophe will happen.
Re 259: “Introduction to Chaotic Dynamical Systems”
Did I say anything about thermodynamics, or any sort of dynamics? I’m a programmer. Please explain in terms if information lost while transmitting a message.
Thus, closure only means something to me as a mathematical property, and thermodynamics is nothing more than a nifty description of our current lack of understanding. (There is no such thing as a “closed system” as you are describing them, and therefore no ‘real’ reason to consider them. Your ‘definition’ has no content.)
Now… The sensitivity of weather to minor perturbations is such that there is no determination of the result from some supposed initial conditions, but there are periodic behaviors. Weather systems behave chaotically, and the outline of their behavior can be fairly modeled as you have stated. I agree that the general shape of climate can be predictable — but not timings. (I don’t know if climate is properly chaotic or not. The timeframe for deciding this is much longer than the time I expect to be alive.)
I would like to know, if there exist any studies on the impact of ice-free Arctic into the climate of the Northern Hemisphere. Pointers appreciated!
Is some research group modelling ice-free North Pole? Is there reasearch going on incorporating that condition into the global climate models?
My intuitive feeling is that the melten Arctic will significantly increase the number of extreme weather conditions.
And just how accurate are current passive microwave methods, in terms of their ability to successfully detect and characterize sea ice? (This question is neither meant to be naive, nor, rhetorical, it is meant to provoke a serious discussion informed by science and engineering considerations).
RE: #47 – For geopolitical or military reasons, would one or more great powers ever consider purposely dissipating sea ice? Of course they would. To think otherwise is to be in denial of human nature. As ugly as that may be.
The ice by Severnaya Zemlya is thinning noticeably over the last few days, with what looks like leads opening between the islands closer to the coast.
http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/arctic.jpg
It’s looking a bit cloudy over some of the exposed ocean, but there is still room for the sun to shine through.
http://en.allmetsat.com/images/composite_ssec_c.php
It’s a tense moment, eh!
http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/current.365.jpg
No net change in extent over the last day or so – but we still have to make it through the September null before it turns around.
What will tomorrow bring? A pause, or a dive?
RE #261 [(There is no such thing as a “closed system” as you are describing them, and therefore no ‘real’ reason to consider them. Your ‘definition’ has no content.)
Now… The sensitivity of weather to minor perturbations is such that there is no determination of the result from some supposed initial conditions, but there are periodic behaviors. Weather systems behave chaotically]
If the definition I quoted had no content, you wouldn’t be able to say that there were no such things. Agreed, there are none – but some systems, e.g. a double pendulum, are near enough closed to be usefully treated as chaotic systems in the sense of the definition I quoted. Weather and climate are not. What do you mean by “Weather systems behave chaotically”? Just sensitive dependence on initial conditions together with periodic behaviours?
[[There is no such thing as a “closed system” as you are describing them, and therefore no ‘real’ reason to consider them. ]]
You just debunked reductionist science in general.
Re 266: “Just sensitive dependence on initial conditions together with periodic behaviours?”
First, there is no such thing as initial conditions. A chaotic system is one which can respond differently at different times to the same stimulus. The Natural Number group is not chaotic, a cat is chaotic. It may be that all natural systems are chaotic, though I am not proposing this as a fact.
While I agree that dynamics can usefully describe the behavior of some chaotic systems, I do not agree that dynamics defines them. (Basically, I disagree with your philosophical stance here that our descriptions have some sort of fundamental meaning.)
Sphere–well, given that climate is defined as the average weather conditions persisting over time, the very idea of predicting events is outside the scope of “climate studies”. Weather does indeed often exhibit characteristics of chaos. Climate has been remarkably stable over the past 10000 years–coincident with the development of human civilization.
Re #268 [Basically, I disagree with your philosophical stance here that our descriptions have some sort of fundamental meaning.]
That’s not my philosophical stance. I’m just trying to discover what you mean by “chaotic”. There’s an everyday meaning of this word: without any discernible order, a complete mess. You don’t seem to mean that. There’s a mathematical meaning for it, which I guessed you thought you meant, but which doesn’t fit either weather or climate – hence my original post. You don’t mean that. I’ve suggested a meaning. You don’t mean that. But you won’t say what you do mean. Maybe you don’t know?
Look at the time scale.
The solar system is chaotic, over the extremely long time scale.
http://adsabs.harvard.edu/abs/2007DDA….38.1101L
But Velikovsky is still wrong.
Re 267: “You just debunked reductionist science in general.”
To the extent one believes there is some sort of Truth and believes in Science as a religion, yes. If one is very careful, and always clear that science never proves anything, but only tends to confirm or tends to deny — no.
Re 270: “There’s a mathematical meaning for it…”
You provided a thermodynamic meaning, not a mathematical one.
I’m objecting to your hijacking the word and ignoring the common meaning while asserting this as some sort of ‘real’ meaning. The common meaning of chaotic is “unpredictable, and unexpectable.”
It’s very common for a certain brand of Scientismist to hijack words and then make claims about other people’s understandings based upon these mis-applied meanings to previously otherwise well defined words. This is what I think you are doing here.
Given the well established meaning of ‘chaotic’, a chaotic system is one for which you cannot predict a consistent response to a given stimulus. (I would speculate that this always implies the system has internal state…but this is mere speculation.)
I’m perfectly willing to grant that thermodynamics provides a very useful description of certain chaotic systems. I am not willing to let you (or other Thermodynamic Scientismists) hijack the word chaotic.
Re 271: “The solar system is chaotic, over the extremely long time scale.
http://adsabs.harvard.edu/abs/2007DDA….38.1101L ”
I’m a bit unclear on what the abstract was trying to say, but it seems to me they were saying that the Solar system has dynamic strange attractors. Since our ability to predict orbits decays with distance in time this system at least appears chaotic to us. (Given the bleeding of atmospheres and the like, it’s a pretty safe bet that the Solar system is chaotic period.)
Another week.
Another million square km.
The curve seems to be leveling, but no signs of an abrupt bottom like last year. At least not so far. (How low can you go?)
My turn for a YIKES! The news media has picked this up almost not at all. We’re going to be talking about next spring’s poor recovery before they even notice.
Yes, the sea ice area is 2.99 million kilometers, and the sea ice extent is 4.78 square kilometers as of August 28.
Which shows that the remaining ice is in bad shape.
They are now predicting the ice to be gone by 2030, which I consider a very conservative statement.
If 2007 was the tipping point, then it will be far sooner.
Sphere (#275) wrote:
The chart I usually look at doesn’t show it leveling off…
Northern Hemisphere Sea Ice Anomaly
http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/current.anom.jpg
… but this chart shows some leveling off:
Northern Hemisphere Sea Ice Anomaly Timeseries
http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/sea.ice.anomaly.timeseries.jpg
There is a pretty good write-up here:
Link from the same article, this does a pretty good motion image day-by-day:
Arctic Sea-Ice Monitor by AMSR-E
http://www.ijis.iarc.uaf.edu/cgi-bin/seaice-monitor.cgi?lang=e
Gives you a good sense of how thin the concentration is.
John Wegner, 13 August 2007:
“I’ve looked at the MODIS satellite images a little closer (and with some places having less cloud cover.) There still isn’t a way through the NorthWest Passage without an icebreaker. Maybe in another week or two.”
Maybe now John?
From Cryosphere Today’s latest movie:
http://members.optusnet.com.au/anon10/CryosphereToday.23.8.07.jpg
And the Terra/MODIS composite for same day:
http://neo.sci.gsfc.nasa.gov/RenderData?si=600438&cs=rgb&format=JPEG
(Yes, it shows a touch more brashy ice than CT resolves.)
How soon before the North-East Passage opens too (arrow)?
Timothy,
I think Shere is referring to this graph when he talks about bottoming out. http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/current.365.jpg
Last year the whole of September was fairly flat, but there is no sign of a flat period starting yet this year.
… although this one does:
http://nsidc.org/news/press/2007_seaiceminimum/images/20070828_timeseries.png
RE #273 [You provided a thermodynamic meaning, not a mathematical one.] You’re just mistaken there: the definition refers only to functions from the real numbers to the real numbers: pure mathematics. Thermodynamics is relevant in deciding where it’s useful to apply the mathematical definition to the physical world: it doesn’t usefully apply to systems where thermodynamic exchange with the external environment is important to the dynamics you’re interested in – like Earth’s climate. However, I’m quite happy now that you’ve said what you mean by “chaotic”.
[[Re 267: “You just debunked reductionist science in general.”
To the extent one believes there is some sort of Truth and believes in Science as a religion, yes.]]
You miss my point. You say there’s no reason to study closed systems since closed systems don’t exist in nature. That’s like saying there’s no need to study ideal gases or frictionless surfaces. It’s the statement of a scientific illiterate.
The sea ice concetrations for 2007 versus 2005 for the August 27 NSIDC update are even startling than the ice extent reduction. sea ice update
I’ve been busy and popped into NSIDC and Cryosphere today after a while away.
I posted in #110 on 14/8/07 that it was premature to make projections for coming
years based on the ice situation as of 14/8/07. I now think that position is much
less tenable.
I’ve just seen the state of the perennial ice and ice thickness as implied
in this graph from NSIDC:
http://nsidc.org/news/press/2007_seaiceminimum/images/20070828_concentration.png
(As I now find Mauri Pelto has linked to – but it’s so important it bears repeating)
Also figure 4, the animated gif on the NSIDC website – the reduction of 5 year (red)
perennial ice is staggering.
It really is looking to me like there’s been such an impact on perennial ice that
this is pretty likely to impact the overall trend. (But I stress I am only an amateur
reader of climate science.) Like others above I’m now thinking of Hansen’s alarm about
the rapid nature of ice sheet loss.
This has left me shaken to the core, an ice free Summer Arctic Ocean before 2020 does
not sound unreasonable to me right now. :(
Re: 284: This has left me shaken to the core, an ice free Summer Arctic Ocean before 2020 does
not sound unreasonable to me right now. :(”
How about before 2010? (Unless you count glacial outflows.)
Right now I’d not predict that outright — but if we start next spring’s melt a million km short I think it’s a pretty good bet.
It’s clear the existing regime has collapsed. The question now is whether there’s going to be several years of wild swings or a direct transition to a new meta-stable state. If the ice doesn’t recover this winter there’ll be little room for a swing — too much excess energy. (Barring explosive volcanism.)
Re 282: “You miss my point. You say there’s no reason to study closed systems since closed systems don’t exist in nature. That’s like saying there’s no need to study ideal gases or frictionless surfaces. It’s the statement of a scientific illiterate.”
Hmmm…
Perhaps it would help if you quoted me exactly rather than misrepresent what I said.
There are no circles either. They are just useful fictions — exactly like ideal gases, frictionless surfaces, and closed systems. Being useful fictions does not make them ‘real’. Any definition which attempts to reify closed systems is a definition with no content — it represents nothing. If chaotic systems can be said to exist then they cannot be properly defined as an instance of something which does not exist — they may, however, be partially modeled by a representation of something which does not exist.
I think you missed some classes on the experimental method and the philosophy of science, child.
Question on climate models.
Are there any models out there which have in any way predicted the ice dieback of this summer? They don’t have to have gotten the year right. All that I want to know is if any model has predicted that at some point in the near future there would be a one year anomalous minimum double any preceding year. Are all our models basically meaningless?
(I’m basically asking if all existing models assume that the underlying numeric formula are unchanging. Are there any which attempt a general systems analysis, with changing dynamics?)
Re #287
I have been predicting a rapid die back but have not published anything. My ideas are based on a famous book in its time called “Climate through the Ages” by C.E.P. Brooks. Once the ice starts retreating there is a positive feedback which speeds the melt up.
The problem is that the whole of science is hung up on linear trends. For instance see: http://nsidc.org/news/press/2007_seaiceminimum/images/20070810_SeptTrend.gif
Plot the ice extent for 2007 on that diagram and you can see that, the rather than the straight line shown, the best fit is a parabola going to zero in a couple of years.
You can argue about whether the climate is chaotic, but it is a non-linear dynamical system with negative and positive feedbacks. So long as the negative feedbacks dominate you have a stable system such as that over the last 10,000 years. But if the positive feedbacks dominate then you have a rapidly changing system just as we had 10,000 years ago at the end of the Younger Dryas. The positive feedbacks cause the system to runaway until it gets into a new state where the negative feedbacks again dominate. The new state is then stable, and it will sit there for perhaps another 10,000 years.
But the next rapid change is only a few years/months away!
“Sphere Says:
29 August 2007 at 17:29
Re: 284: This has left me shaken to the core, an ice free Summer Arctic Ocean before 2020 does
not sound unreasonable to me right now. :(”
How about before 2010? (Unless you count glacial outflows.)
Right now I’d not predict that outright — but if we start next spring’s melt a million km short I think it’s a pretty good bet.
It’s clear the existing regime has collapsed. The question now is whether there’s going to be several years of wild swings or a direct transition to a new meta-stable state. If the ice doesn’t recover this winter there’ll be little room for a swing — too much excess energy. (Barring explosive volcanism.)”
I agree, another summer season like this and barring an unusual winter freeze season it’s hard to see much ice this time next year! It would seem to require a change from the regime prior to this year’s for the ice to last as long as 2020?
________Why not 2010?_________
Firstly I’d rather stick to 30 year periods, but a 30 year period will not do
this process justice, so the minimum I’ll go for is decadal. However note I
didn’t say 2027 (2 decades from now) – from my reading that sounds like it could
be an overshoot – I don’t think it at all reasonable to expect a linear continuation
of the trend offset downwards by this year’s figures.
It’s almost Sept 2007, that gives 2 melt seasons after this for the Arctic to be
ice free. And the actual area, rather than the anomaly doesn’t seem to me to
support “2010 ice free in September”:
http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/current.area.jpg
Bear in mind that what NSIDC observed and I bought attention to in post #110,
the unlucky combination of circumstances, was just that. It may impact future years,
but expecting it again and again seems unrealistic.
An ice free Arctic in 2 melt seasons seems very unlikely to me. But adding on
another 10 melt seasons and bearing in mind what seems now to be a likelihood
of non-linear response, 2020 seems feasible.
If people go around claiming “ice-free summer by 2010” they’re likely to have gaggles
of the usual tiresome contrarists squawking about “alarmism”. I don’t think we’ll do
anything about CO2 emissions, but if I’m wrong (as I hope I am) then more delay and
ill-informed doubt is not what we need now.
Alastair, please. That was published last year and there’s an entire thread here at RC about it. “The whole of science” isn’t “hung up on linear trends.” You’re in the mainstream.
You may be thinking of this thread, which points out that linear extrapolations are more optimistic about sea ice than climate model runs and every model includes some runs that show periods of unusually rapid decline (which get blurred if only the average rather than the extreme cases is looked at). This looks at model runs that look much like this week’s Arctic sea ice.
https://www.realclimate.org/index.php/archives/2007/01/arctic-sea-ice-decline-in-the-21st-century/#more-391
Even if the ice does recover this winter, it may not mean much, as it takes far less energy to melt freshly created thin first year ice, than it does to melt thick multiyear ice.
Also of interest, on Cryosphere today, if you look at the ice lost for the individual area, rather than the whole picture, it appears that the high arctic (north pole) area shows that melting has come to an end. Now I expect the areas around the ice edge to continue to melt for a few more weeks until the sea cools down, and also as we get rid of some isolated pockets of ice in the Fox Basin of Canada.
The North Pole Web Cam #4 on the Polar Sterm has reacehd rougly 89 North. Im not sure if the latest images show melt ponds or open leads that are beginning to skim over with ice. But it appears that the Polarstern is having an easy go at it, getting to the North Pole, looking at the distance they covered each day.
Sphere,
There is a risk that this discussion could descend into sophistry. Weather certainly exhibits characteristics of chaos. Climate deals with long-term averages of weather. It is constrained by the conserved quantities–energy, momentum, angular momentum…. And there is no evidence that for small perturbations of the climate, you get chaotic behavior. Perhaps we need to be careful here in our terminology.
Also, I would characterize your dismissal of the study of closed systems as a bit rash. While a circle may be merely a useful construct, a Sphere evidently does exist ;-) and can either concentrate on clarifying points or on scoring them.
re 293 and sophistry…
===========
More than a bit off-topic, to say the least, but for some odd reason, call it a weird itch at the back of my consciousness, I am reminded more and more that “Sphere” is also the title of a sci-fi novel.
Not that I expect this post to make it to the discussion, mind you…very off-topic, indeed.
Re 283: Thanks Mauri for the link. It cleared up a question that was starting to bother me. (See way down on the cited page)
My question was: Was this rapid melting a meteorological or a climate phenomenon?
The answer seems to be that it was mainly a meteorological one. A durable high pressure area was set up in May over the Arctic ocean with very clear skies, and an increased amount of solar energy was received at the surface. Then there was strong feed-back due to lower albedo (pools of water over ice).
The causation issue with respect to climate change seems approximately the same as with the multi-year droughts and record heat waves. That is, the causation mechanism remains uncertain.
(My judgment on this issue tends to be that global warming has, and will have, most certainly impacts on all these spectacular phenomena – but that the impacts are unknowable in any single case. For instance, it is impossible to know if the N.O. Superdome was in fact endangered or if it was saved from collapse by the impacts of global warming on hurricane Katrina. We have to live with that kind of uncertainty, accept some additional risk. The weather statistics have changed.)
This Arctic melting case has an interesting component. There is a lot of “memory” involved. What will be the fate of the extra energy that was received and captured in the surface waters? How will it be mixed, circulated and dissipated?. Still, if the melting will expand next summer depends mostly on the weather, although the starting level will probably be favorable for a further reduction.
Since 2002, I have said 2020 was to be the time of a blue arctic ocean in the summer. With this years results, I am updating that to 2011. Overall, I agree with Walt Bennett. It seems to me that it all started when we lost the overall ice thickness and mass between 1989 and 2000. This loss occurred without the benefit of the albedo feedback. The mechanism for this loss has not been explained or understood to my knowledge. Now with the albedo feedback kicking in, the ice is quickly melting. How warm will the Arctic Ocean become in summers now? How much will Greenland melting accelerate? What is the name of the next climate age? [May I suggest Ohshitocene — as in we sure didn’t see this coming!]
Re 288: “You can argue about whether the climate is chaotic, but it is a non-linear dynamical system with negative and positive feedbacks. So long as the negative feedbacks dominate you have a stable system such as that over the last 10,000 years. But if the positive feedbacks dominate then you have a rapidly changing system just as we had 10,000 years ago at the end of the Younger Dryas. The positive feedbacks cause the system to runaway until it gets into a new state where the negative feedbacks again dominate. The new state is then stable, and it will sit there for perhaps another 10,000 years.
But the next rapid change is only a few years/months away!”
Yes — if the rapid change isn’t in progress; which we won’t know for a while. I haven’t argued whether climate is chaotic or not — except to speculate that all natural systems are chaotic.
Do you propose that the climate can be accurately modeled by some unchanging numeric equation of some finite degree? That is: Does the nature of climate remain unchanging over time?
(P.S. Can we reclaim negative and positive feedback from society at large, or are we stuck with negative feedback being a negative?)
“Paul Crutzen, a Nobel Prize winner for his work on the ozone layer,
has set out to give the future a new name.
The Anthropocene Epoch: In Favor
“In a stirring 2000 paper, … proposing to use the term ‘anthropocene’ for the current geological epoch.” He repeated his proposal to a much larger audience in a January 2002 Nature article.
“This is a well-worded and precise proposal….
“… his name is only a proposal, which would have to be ratified by the world’s major scientific organizations first….”
Links to the papers and more info in the full article, which I recommend:
http://geology.about.com/od/geotime_dating/a/anthropocene.htm
Craig Dillon (#296) wrote:
According to Cryosphere Today, we were below 3 million km sea-ice area on Tuesday, more than 25% below the previous all-time low of 4 million set in 2005. I’d would try and relax: at this rate we have another five years of summer sea-ice. 2013.
Re 292: “Even if the ice does recover this winter, it may not mean much, as it takes far less energy to melt freshly created thin first year ice, than it does to melt thick multiyear ice.”
I’m vaguely aware of that — but I don’t think we’ve got a good measure of ice volume and mass about as a talking point. I’ll just say that if the ice recovers this winter I’ll wait to see the next summer’s melt before making a guess on when the Arctic will be ice free.
If the ice falls 1 million km short I’m betting on two more melt cycles. If not then I’m not ready to bet.