Ice-free Arctic

[Daniel B.]

We're trying to project a description of a complex phenomenon well beyond our data, and should be cautious in whatever we predict.

That is exactly why makes this so darn scary. No one really knows but everyone is pretending that it will all be just fine without knowing is that is true or not. Given the uncertainty in both the date and possible consequences of a blue ocean event, there should be an all out effort to:

1. Invest in all resources necesary to find out if, when and how the BOE is happening
2. Prepare humanity for it.

I wish I was exaggerating and being alarmist but all the evidence I find tells me I'm not. An ice free Arctic is the biggest threat humanity has ever faced.

Actually, I think you are.  While they is a real probability that an ice-free Arctic could happen soon (<10 years), there is a similar probability that it will not happen for a long time.  I understand that you want to alert others to this possibility, but I think you are over-estimating how soon it will occur.  Secondly, your last statement is conjecture.  The consequences of an ice-free Arctic are not known.  Saying that it is “the biggest threat that humanity has ever faced,” sure sounds alarmist.

[gerontocrat]

It seems that during a -AMO and -PDO cycle the most we can hope for is a hiatus in global temps.  That might or might not translate to a hiatus in Arctic temperatures.  The trend can indeed slow down for a few years or even decades. Sadly GHG's ensure the temperatures won't cool enough to reverse the trend.

Temperatures as in Atmospheric Temperatures.

But CO2 ppm will continue to increase. More heat is trapped.
Where does it go? The oceans.
Does some of that extra heat go North (and South) eventually? Yes.
Is that process slower than movement of heat through the atmosphere? yes.

So Choose - death of Arctic Ice by a 1,000 cuts or by 100 cuts?

[Archimid]

Actually, I think you are.

Of course you do. I already gave you the scientific explanation for that. It is too scary for you to handle, so your brain plays tricks on you to protect your emotional health leading you to denial.

I know this is mean but I believe that is the scientific truth that explains how you can ignore the dangers. Is either that or ignorance. You may or may not have that excuse, Dr. Curry doesn't. I know is real because even as I'm saying this I don't want to believe it is true. In fact I spend a good portion of my day trying to avoid thinking of this and pretending it is not happening. This whole global warming nightmare has a very detrimental effect on my health. I perfectly understand why anyone would try to avoid it like a plague.

While they is a real probability that an ice-free Arctic could happen soon (<10 years), there is a similar probability that it will not happen for a long time.

Define long time. If by long time you mean 20 years, then I have no option but to agree, but only because the trends must be acknowledged.  The probability that it happens much later than 2037 is much lower than for it to happen sooner because of the small detail of global warming.

The consequences of an ice-free Arctic are not known. Saying that it is “the biggest threat that humanity has ever faced,” sure sounds alarmist.

No, the consequences are not known but that uncertainty is reason for extreme concern not complacency. This is extremely concerning by virtue of the centrality of the Arctic to the North Hemisphere, the longevity of the Arctic sea ice in the current state of the climate system and the surface area of the Arctic relative to the world. Arctic sea ice is a BIG deal(and thats a huge understatement). That alone, without knowing anything else is reason for extreme concern.

To estimate the danger the key is understand that the Arctic is ice stored during winter that the climate system uses in the summer normalizing the climate throughout the year.

If the ice is not there during the summer solstice the Arctic will receive more power from the sun than the tropics. It will get EXTREMELY warm. It doesn't get warm now only because there is ice left to be consumed by the sun. I think a good way to visualize it is with the n80 temperatures graph:



Why the large variability in winter but extremely low variability in summer? The summer temperatures near the surface are dictated by the ice. If the ice wasn't there the N80 temps graph would not have the top shaved off, instead it would follow a shape similar of solar insolation above 80:



Do you think that can destabilize the climate? If you say no please give some evidence or at least a reasonable explanation, because that would be an extraordinary claim. 

That is reason to sound the loudest alarms possible so that everyone can get ready and maybe, if we try hard enough, even stop it.

I sound alarmist because that is the only possible way to tell you the truth. I can lie to you and not alarm you, but that would be a lie. A comfortable lie for both you and me, but literally deadly lie. The truth is that there is climate emergency going on. It is likely that extreme weather keep rising until a BOE, then after the BOE the Earth will be a different planet at least until a new form of climate stability is reached.

[Daniel B.]

No, the extraordinary claim is that the climate will be destabilized.  You are confusing truth with your own opinion.  Not that it could never happen.  It is not impossible.  I do not disagree with your predictions, just because they are scary.  No, I disagree because they are unsupported by the evidence.  Your ad hominem attacks will not sway my thinking.  Only scientific evidence will do so.  As many other posters have stated, the Arctic is a chaotic system with many factors affecting the outcomes, and consequently, making predictions are quite difficult.  Therefore, making the extraordinary claim of an ice-free Arctic in less than a decade, requires extraordinary evidence to go against the current trend.

I know many people fear uncertainty.  That is quite normal.  We become comfortable in our knowing.  However, claiming that uncertainty means danger, is a rather pessimistic view.  By your posts, I believe you are truly alarmed.  However, you have yet to convince that I should be alarmed.

[Sleepy]

Only scientific evidence will do so.

A thought is 98% unconscious. We can only understand what our brains allow us to understand. If you belive in science, you've gotta belive that too.

[El Cid]

We could have possibly lost much of the summer ice during the Holocene climatic optimum and the Eemian. Still, it did not lead to runaway instabilization of the global climate. An ice-free (summer/autumn) Arctic would definitely lead to much warmer temperatures in the Arctic and possibly warmer temperatures in the (NH) midlatitudes. It could possibly lead to atmospheric circulation changes which could enforce or weaken the warming. There are just too many unknowns but Daniel B is right: there is no scientific evidence that such a new state would lead to collapse and chaos.
Climate models projecting 2100 temperatures show 3-6 degree warming in NH midlatitudes. That includes all the emissions till 2100 AND an ice-free Arctic. So if the Arctic goes ice-free by 2030 (and it might) I think the scientific opinion is that NH midlatitude temps will be 1-3 degrees above baseline, which is a lot but definitely won't lead to collapse.

Mind you, I believe that climate change is a very serious long term (meaning 50-100 yrs) threat but I do not think that it will cause systemic problems in 10-20 years even if we lose summer ice and I think that opinion is unsupported by facts.

[Avalonian]

Fascinating as this thread is, there's obviously little common ground in what will actually happen. My gut feeling is very similar to Archimid's take on it, but clearly other people don't see it the same way, no matter how much that might jar with my sense of reality. So, can we actually find things to agree on, and build up from there?

I can see there's not going to be universal agreement on likely timing until a BOE actually happens. But if the next two or three years sees another massive record (in the style of 2012 but more severe still), will that be enough to accept that the decline is not slowling down? Conversely, hat outcomes in the next 3 years would be enough to change your mind, from either side?
     Personally, I fully expect to see a new record soon, but I'm also willing to believe that some negative feedbacks are now slowing the process. If it reverts to around 2013-4 levels and stays there over the next few years, I might even start to wonder about that Gompertz fit... 

Now let's assume that we do get an ice-free Arctic at some point (unspecified, but not too far away). Are we all agreed that this means that Arctic summer temperatures will rise dramatically above zero, to an extent depending on the length of time the ice is gone for?

[Gray-Wolf]

Now let's assume that we do get an ice-free Arctic at some point (unspecified, but not too far away). Are we all agreed that this means that Arctic summer temperatures will rise dramatically above zero, to an extent depending on the length of time the ice is gone for?

I think there are a number of us waiting to see the DMI 80N plot leave the 'latent heat of fusion' dictated summer temps to its first spike well above freezing ( 2 or 3 degrees?) as the floes thin out and open water exceeds ice covered?

I think the very end of 2016 melt season did see ( with the eye of the faithful) an uptick at the very end of the 'pinned to zero' period of the DMI 80N

With low peripheries again there is a chance for an early break up of the central pack and then drift may lead to an open central polar area and again a chance to see 'lift off' of the temp line on the DMI plot!

[El Cid]

Now let's assume that we do get an ice-free Arctic at some point (unspecified, but not too far away). Are we all agreed that this means that Arctic summer temperatures will rise dramatically above zero, to an extent depending on the length of time the ice is gone for?

I think that when we lose all summer ice it will happen in August/September, not before, so Arctic temperatures will not be able to go up (due to the coming polar night and the still cold sea), instead they will just refuse to go significantly down in Sep-Oct, and and stay very high until Nov/Dec whent the Arctic will refreeze.

I believe that for Arcic temperatures to go up significantly by August, we would need to see a comlete meltout by June, but that is probably not on the cards for now.

So my take is that Sep-Nov  (Dec) temps will be drastically higher but not summer temps. What changes this will cause to atmospheric circulation - i have no idea...

[crandles]

Actually, I think you are.

Of course you do. I already gave you the scientific explanation for that. It is too scary for you to handle, so your brain plays tricks on you to protect your emotional health leading you to denial.

You do realise we can easily counter that you are concentrating of this subject, so you have vested interest in it being important, 'so your brain plays tricks on you to protect your emotional health leading you to denial'?
So this doesn't work well as an argument.

.

Define long time. If by long time you mean 20 years, then I have no option but to agree, but only because the trends must be acknowledged.  The probability that it happens much later than 2037 is much lower than for it to happen sooner because of the small detail of global warming.

We agree that global warming is not a small detail, but for such a statement you need some quantification which appears completely missing in this paragraph.

The trends are the least of the science. Far more important there are lots of models with lots of different dates for ice free but generally 2040 or later. Given this poor agreement, they might all be saying later than reality. However, they are all showing a slow down in the rate of loss as ice free is approached. When there is such solid agreement on that, and the data also shows it then this looks fairly solid but still not 100% so there is still a small chance of being in next 10 years but I would suggest the vast majority of the probability is for 2037 or later.

Above personal opinion means little. It does however appear to recognise more of the science than you are doing.

[crandles]

Now let's assume that we do get an ice-free Arctic at some point (unspecified, but not too far away). Are we all agreed that this means that Arctic summer temperatures will rise dramatically above zero, to an extent depending on the length of time the ice is gone for?

No. The first time it happens will probably be before the trend reaches ice free.

I agree with El Cid that it will be late in the season. Some extra warmth will be accumulated by mid September. However during fall this will get vented to the atmosphere. (less ice less insulation)

I go further than El Cid:
The delay before ice forms solidly enough to collect snow won't be very long but this is the snow season. The lack of snow cover on the ice means less insulation so more heat continues to be lost to the atmosphere during winter. The result is thicker ice cover than usual for that period at the next maximum. Less snow might cause slightly earlier melt onset but these effects dissipate and things will probably be back on trend about 2 years after the ice free event.

(Probably - I admit some risk of weird weather effects meaning it doesn't play out as models suggest.)

[Alexander555]

Now let's assume that we do get an ice-free Arctic at some point (unspecified, but not too far away). Are we all agreed that this means that Arctic summer temperatures will rise dramatically above zero, to an extent depending on the length of time the ice is gone for?

No. The first time it happens will probably be before the trend reaches ice free.

I agree with El Cid that it will be late in the season. Some extra warmth will be accumulated by mid September. However during fall this will get vented to the atmosphere. (less ice less insulation)

I go further than El Cid:
The delay before ice forms solidly enough to collect snow won't be very long but this is the snow season. The lack of snow cover on the ice means less insulation so more heat continues to be lost to the atmosphere during winter. The result is thicker ice cover than usual for that period at the next maximum. Less snow might cause slightly earlier melt onset but these effects dissipate and things will probably be back on trend about 2 years after the ice free event.

(Probably - I admit some risk of weird weather effects meaning it doesn't play out as models suggest.)

The problem is that what you explain here ,is what is happening already for a long time. The area with less or no  ice is already getting bigger for dozens of years. So it looks like the heat that is escaping is warming the Arctic. It's not making the ice thicker, we lost 75 % of the summer ice. Probably because the ocean stores more and more heat every year. And in some way you could say that it will only release that heat if the air above it is colder. If that point moves further north, it will hold that heat until it's further north.

[gerontocrat]

Now let's assume that we do get an ice-free Arctic at some point (unspecified, but not too far away). Are we all agreed that this means that Arctic summer temperatures will rise dramatically above zero, to an extent depending on the length of time the ice is gone for?

No. The first time it happens will probably be before the trend reaches ice free.

I agree with El Cid that it will be late in the season. Some extra warmth will be accumulated by mid September. However during fall this will get vented to the atmosphere. (less ice less insulation)

I was looking at albedo and Archimid pointed me to a lovely piece of work by Tealight and Nico Sun.

In the 2012 Arctic spring, sea ice was comparatively high, very high, compared with other years in the 2010's. Then relatively late the big collapse came.

In 2016, sea ice loss started really early, but then became sort of average, with sea ice minimum hundreds of thousands of km2 greater than 2012.

And despite that 2016 cumulative albedo warming potential for the year is higher by far than 2012.

i.e.A low winter maxima and early melt is the most effective way to heat up the Arctic ocean.
A late melting don't cut it.

[Sleepy]

I was looking at albedo and Archimid pointed me to a lovely piece of work by Tealight and Nico Sun.

Don't want to interfere here, but Nico Sun is Tealight.

[gerontocrat]

I was looking at albedo and Archimid pointed me to a lovely piece of work by Tealight and Nico Sun.

Don't want to interfere here, but Nico Sun is Tealight.

The original graph said it was by tealight and calcs by nico sun. Ho hum9

[Daniel B.]

Fascinating as this thread is, there's obviously little common ground in what will actually happen. My gut feeling is very similar to Archimid's take on it, but clearly other people don't see it the same way, no matter how much that might jar with my sense of reality. So, can we actually find things to agree on, and build up from there?

I can see there's not going to be universal agreement on likely timing until a BOE actually happens. But if the next two or three years sees another massive record (in the style of 2012 but more severe still), will that be enough to accept that the decline is not slowling down? Conversely, hat outcomes in the next 3 years would be enough to change your mind, from either side?
     Personally, I fully expect to see a new record soon, but I'm also willing to believe that some negative feedbacks are now slowing the process. If it reverts to around 2013-4 levels and stays there over the next few years, I might even start to wonder about that Gompertz fit... 

Now let's assume that we do get an ice-free Arctic at some point (unspecified, but not too far away). Are we all agreed that this means that Arctic summer temperatures will rise dramatically above zero, to an extent depending on the length of time the ice is gone for?

I think we can all agree that the Arctic sea ice is shrinking.  Beyond that, I am not sure.  We have vastly different views as to how fast, and when the Arctic might become ice-free.  Remember, ice-free is defined as less than 1 million sq. km, not zero.  The Arctic has been rather chaotic the last few years, more so than usual, making it very hard to establish the current trend, let alone anything long term.  Few scientists are claiming an ice-free state in less than 20 years.  However, even those that are estimating a longer time frame cannot rule out a shorter term option.  Neither can they rule out a much longer option.  Thomas, no has published a probability curve regarding when the Arctic might become ice-free.  But like everything that has random variables, there is the probability that it could happen sooner or later than expected, especially when trying to pinpoint a condition that has not occurred in eons.  I will agree that a near record low in a year or two would enough to accept that the decline is not slowing down.  Contrarily, another year or two (maybe three) should be enough for most to accept that it is slowing down.  As always, more years of data will lead to better analyses.

Your last question about Arctic summer temperatures rising dramatically seems to lack agreement here also.  As many posters have shown, summer temperatures in the Arctic have not deviated much (compared to winter temps.), despite losing a significant portion of the ice surface to open water.  It would seem reasonable to assume that more heat would be absorbed by the darker water, and I am surprised that summer temperatures have not risen.  However, that is but one factor influencing the Arctic, and may be overridden by others. 

[Alexander555]

Daniel, for what reason are you saying it's slowing down ? Where do you see it slowing down ? The temperature anomolies are getting bigger and bigger.

[crandles]

Daniel, for what reason are you saying it's slowing down ? Where do you see it slowing down ? The temperature anomolies are getting bigger and bigger.

I'm guessing
"Contrarily, another year or two (maybe three) should be enough for most to accept that it is slowing down."
was referring to arctic sea ice volume.

Maybe he found my posts at #220 and #228 convincing.

[Daniel B.]

Daniel, for what reason are you saying it's slowing down ? Where do you see it slowing down ? The temperature anomolies are getting bigger and bigger.

Not temperature, but sea ice decline as show here and the graphs presented by crandles.

https://climate.nasa.gov/vital-signs/arctic-sea-ice/

[Alexander555]

I see, we have to wait and see where it goes. But the last year i have been paying attention to it a little more than normal. And so far almost the entire winter the temperature was 3 to 6 degree C above average at the arctic. With sometimes vast areas 20 to 30 degree C above average. And these anomalies are getting bigger , that means a less cold arctic . Only the last few weeks it was colder than average.

Yesterday we had something interesting in the news. About the fishing industry. All the spicies are moving north. The spicies they were catching in the south of Europe. They are now in their nets in the North-Sea, like ink-fish . And the fish from the North-Sea moves further to the north. That is all about temperature. Something big is going on.

[Archimid]

Now let's assume that we do get an ice-free Arctic at some point (unspecified, but not too far away). Are we all agreed that this means that Arctic summer temperatures will rise dramatically above zero, to an extent depending on the length of time the ice is gone for?

Some extra warmth will be accumulated by mid September. However during fall this will get vented to the atmosphere. (less ice less insulation)

i.e.A low winter maxima and early melt is the most effective way to heat up the Arctic ocean.
A late melting don't cut it.

I posted the comment chain because I believe this discussion is the crux of the matter. I agree with everything said, but I think that based on what you all agreed the conclusion is that after the first BOE there is a significant change. Here is why I think that:


Let's define the first Blue Ocean Event (BOE) as the moment in time there is 0 ice in the Arctic Ocean. Let's take the average September minimum volume as the day of the first BOE.  Let's suppose that ice growth begins on the same day of the BOE. Lets assume the same conditions of the last 11 years.

If starting from 0, the ice grows the same way it grew for the last 11 years, then we can expect an average maximum of 17.69 x 1000km3. The average losses during the same time are of  18.09 x 1000km3. The data points to no recovery after a BOE.

 Of course things are not so simple. I'm going to check my assumptions.

Let's suppose that ice growth begins on the same day of the BOE.

That's a very bad assumption because:

Some extra warmth will be accumulated by mid September. However during fall this will get vented to the atmosphere. (less ice less insulation)

 If that heat is vented to the atmosphere, the atmosphere warms delaying ice formation in an amount roughly proportional to the heat accumulated during the year and transferred into the atmosphere. The earlier the BOE happens the more heat is accumulated. Pushing the start of ice formation back, likely resulting in a lower maximum.

But it is not just the warmth accumulated in the Arctic Ocean what delays the date. If the current behavior of the atmospheric currents worsen, then warm air intrusions from the NH will likely further delay things.

Then there are waves. Ice will not form while there are large waves. I'm not sure of any mechanism to control waves other than the ice front. As the ice front grows the waves are pushed back allowing for more ice growth. Maybe ridiculous amounts of snow might do it.

I think it is the very least the beginning of ice formation will be delayed.

Lets assume the same conditions of the last 11 years.

That is also a bad assumption because there will be negative albedo changes, wavier jets streams coupling with humidity changes in the atmosphere above the Arctic Ocean, and changes in the currents of the Atlantic ocean. The conditions will probably be more different than the difference of the last 11 years from the time before.

If starting from 0, the ice grows the same way it grew for the last 11 years, then we can expect an average maximum of 17.69 x 1000km3.

Horrible assumption. In the Arctic ice begets ice. That means that when it starts growing the area of ice growth will be extremely small and requires exponential growth to match the speed at which the ice grew for the last 11 years. There is no physical problem with that except that it requires a cold atmosphere and a cold ocean.  From what we have seen in the last three freezing season there is not enough endogenous cold in the arctic to maintain the cold temperatures needed to generate large amounts of ice even now. After a several BOEs it will be much worse.

The average losses during the same time are of  18.09 x 1000km3.

A stupid assumption because there can't be less than 0 ice! You can't lose ice that is not there in the first place. Therefore negative ice must be converted to heat that is added to the system.

A late melting don't cut it.

Not the first time around, but after the first BOE, the next time around there will be a lot less ice, which will melt much sooner, pushing the date of the BOE earlier and earlier.

  The only way to stop it and reverse it is to remove heat from the system.

[Alexander555]

I don't know if it fits the entire planet. But over here in Europe, the higher summer temperatures last a lot longer into the end of the year. It looks like the summer is expanding. But you don't have that difference at the end of the winter. There are a few  extremer cold peaks ( if you filter the data out) that happen later than average. But i think that's the result of that weaker Vortex. Spring itself moved 15 days forward in 20 years.

[Tealight]

I was looking at albedo and Archimid pointed me to a lovely piece of work by Tealight and Nico Sun.

Don't want to interfere here, but Nico Sun is Tealight.

How do you know? I might have different personalities that work together on a project 

I haven't read the whole discussion in detail, but I agree that with gerontocrat that spring ice cover isn't very indicative for the September minimum. Even better than the total cumulative albedo is looking at the regional distribution. 2012 had a high albedo in spring in the periphery, but in the central Arctic albedo was very low over a very widespread area. Other record low years typically have 1-2 dominant areas of low albedo and then the heat just prevents refreezing in autumn.



I'm not joining the discussion of what we define as ice-free, because it depends on your application. It's different for shipping, polar bears, ocean heat accumulation, ....


Btw: For the start of the 2018 melt season I calculated the cumulative AWP for years back to 1988. I fear before 1988 the much larger pole hole would violate data consistency. Some years are very interesting. 1990 has the record low albedo in the east Siberian sea, beating even 2007. 1998 had very low albedo on the entire american side and really high on the asian side.

https://sites.google.com/site/cryospherecomputing/warming-potential

[Alexander555]

Just to make sure that what i think is right. Because the term albedo is new for me. But it represents the capacity of an object  to reflect the sunlight, correct ? So a low albedo means the sunlight is not reflected, but used to melt the ice or to warm  the water. So that makes me think that all the red colour you see, reflects the bad condition of the ice, or the snow on top of it. Is that correct ?

[oren]

Archimid, thin ice gains volume much faster than thick ice during the cold of midwinter. That is a missing assumption in your argument.
Daniel B., I suggest to look at more graphs than just the headline trend of extent at minimum. For area we already had a near record low in 2016. Also look at volume at max and min. Look at the percent of old ice in the arctic. I think you may find that ice internals still point down when the main graph you are focusing on seems to be in a "hiatus". Just like ocean heat content continued building up when the deniers were claiming global warming has stopped based on the headline graph of global surface temps.

[oren]

Just to make sure that what i think is right. Because the term albedo is new for me. But it represents the capacity of an object  to reflect the sunlight, correct ? So a low albedo means the sunlight is not reflected, but used to melt the ice or to warm  the water. So that makes me think that all the red colour you see, reflects the bad condition of the ice, or the snow on top of it. Is that correct ?

When Tealight/Nico Sun calculates albedo-warming potential (very useful btw), he is dealing with the effect of open water albedo (compared to ice/snow-covered ice), multiplied by the insolation level on that date in that location. As far as I understand, he is not dealing with the state of the ice, the state of its snow cover, or the state of clear sky vs. cloudiness.

[Archimid]

thin ice gains volume much faster than thick ice during the cold of midwinter.

That is only true if the temperature holds constant for both thin and thick ice. But winter temperatures are not holding constant, they are rising significantly.

My argument is that the growth rate depicted here



is limited by temperatures



Yes it is completely true that thin ice thickens faster but  growth is limited by the temperatures. Since it is a fair assumption that Arctic temperatures after a BOE will be much higher than the past, then it is fair to assume that there will be much less ice growth.

[Sleepy]

I was looking at albedo and Archimid pointed me to a lovely piece of work by Tealight and Nico Sun.

Don't want to interfere here, but Nico Sun is Tealight.

How do you know? I might have different personalities that work together on a project 

Only one person responded to my email when I asked about using your graphs.

[Shared Humanity]

I was looking at albedo and Archimid pointed me to a lovely piece of work by Tealight and Nico Sun.

Don't want to interfere here, but Nico Sun is Tealight.

How do you know? I might have different personalities that work together on a project 

Only one person responded to my email when I asked about using your graphs.

That is the one with the more agreeable nature. You don't want to mess with the other one. 

[Tor Bejnar]

One aspect of BOE not mentioned recently (last couple of days in this thread) is ocean water column mixing due to storms in ice-free areas.  As soon as the central Arctic gets well mixed, it won't freeze over during the winter, I've read in ASI threads.  Whether this happens the first winter after a BOE or not until the 20th one, is far beyond me.  This was seen in the Arctic during the early Eocene (~50 ma = 50 million years ago) when crocodiles basked on Arctic shores.  (The Earth was also ~8 degrees warmer then.  I don't think there is strong evidence of this happening since then.  Lots has changed since then:  Panama strait closure amongst them)

Of course, even a little mixing would delay the onset of freezing some, and the more delay of freezing, the more potential for more mixing.

[Tealight]

Just to make sure that what i think is right. Because the term albedo is new for me. But it represents the capacity of an object  to reflect the sunlight, correct ? So a low albedo means the sunlight is not reflected, but used to melt the ice or to warm  the water. So that makes me think that all the red colour you see, reflects the bad condition of the ice, or the snow on top of it. Is that correct ?

Yes i use the standard red=hot, blue=cold heat-map. Or in this case low albedo=hot=red, high albedo=cold=blue.

Just to make sure that what i think is right. Because the term albedo is new for me. But it represents the capacity of an object  to reflect the sunlight, correct ? So a low albedo means the sunlight is not reflected, but used to melt the ice or to warm  the water. So that makes me think that all the red colour you see, reflects the bad condition of the ice, or the snow on top of it. Is that correct ?

When Tealight/Nico Sun calculates albedo-warming potential (very useful btw), he is dealing with the effect of open water albedo (compared to ice/snow-covered ice), multiplied by the insolation level on that date in that location. As far as I understand, he is not dealing with the state of the ice, the state of its snow cover, or the state of clear sky vs. cloudiness.

Well, I indirectly consider the state of sea ice. The NSIDC data (NASA Team algorithm) is very sensitive to melt ponds. During peak summer a fully ice covered grid cell might show as 60% sea ice concentration due to melt ponds. My AWP algorithm then treats the cell as 40% ocean. It wouldn't work as well with AMSR2 data.

[Iceismylife]

...
  Therefore, making the extraordinary claim of an ice-free Arctic in less than a decade, requires extraordinary evidence to go against the current trend.

...

The current trend is the four lowest maximum ice volume years are the last four years.

The current trend is ice free very soon.

[Archimid]

We could have possibly lost much of the summer ice during the Holocene climatic optimum and the Eemian. Still, it did not lead to runaway instabilization of the global climate.

Let me tell you my theories about that. First some context.



I agree 100% that the Arctic didn't have ice during the Eemian 135,000 years ago. There is plenty of evidence supporting that. It lasted for thousands of years, enough to develop a flora and fauna comparable to today's tropics. It was great, except that the human population of the time was about 20,000 individuals.

The Eemian started when Milankovitch cycles aligned and kicked started the CO2 cycle. The temperatures climbed for 10,000 years, melting the Arctic past the Holocene average, and remaining much warmer for 5,000 years. When Milankovitch cycles subsided things cooled down until eventually the Arctic returned. After the return of the Arctic and with it winter, glaciation commenced for the next 100,000 years.

The Holocene started in the same way the Eemian started. The sun kicked start the CO2 cycle and the world warmed for 8,000 years but then the Younger Dryas happened. Regardless of why it happened, global temperatures were held back for a thousand years, lowering CO2 and wasting time in sync with milankovitch cycles.

By the time maximum CO2+insolation was reached it was too late. The Arctic sea ice cap was saved. The ice would teeter in and out of existence during the Holocene Thermal Maximum but the gigantic ice sheets of the Northern Hemisphere protected it.

 4,000 years later the time the maximum isolation+CO2 is over. Humans are left with a warm planet that's not as warm as the Eemian and but much warmer than the glacial times. Just perfect to grow food using agriculture over most of the planet. Over the next 5000 years temperatures remain relatively constant even when milankovitch cycles slowly decay overtime. CO2 product of human burning of trees and agriculture keeps the planet sufficiently warm.

Then 200 years ago humans discovered fossil fuels. Ancient Solar energy stored in dead ancient forests. In 200 years we took CO2 from 280 ppm to 410ppm. During the Eemian it only reached 300ppm. Now temperatures are responding to CO2 and the CO2 cycle has been kickstarted again. Soon the Arctic will be gone and it won't be back until CO2 is reduced.

[Daniel B.]

...
  Therefore, making the extraordinary claim of an ice-free Arctic in less than a decade, requires extraordinary evidence to go against the current trend.

...

The current trend is the four lowest maximum ice volume years are the last four years.

The current trend is ice free very soon.

We are referring to ice-free at minimum, not maximum.  It will be a very long time (if ever) before the Arctic will be ice-free during the winter.

[Archimid]

 If we assume a BOE will happen soon and it doesn't, we are at least better prepared for when it does.

 If we assume a BOE won't happen soon and it does civilization as we know it is finished.

[oren]

Here is some data to help ground the discussion. As ice grows more dispersed over time, I think area can often give better clues than extent.
A. It's a long gradual process, and human civilization will not end tomorrow one way or the other. (But wait 50 years and we're probably doomed.  But that's a different thread).
B. I don't think there's a slowdown. Cherry-pick the June 1st graph for difficulty with the slowdown notion. Remember June is peak insolation month, and area is the parameter affecting albedo.
C. As the trend is down, AND volatility to the downside is increasing, as stated up-thread I would be highly surprised if a BOE (<1M km2 extent (or area?) at min) doesn't show up by 2030.

Edit: the chart was wrong due to the pole hole issue, and was removed. Posted later.

[crandles]

Cherry-pick the June 1st graph for difficulty with the slowdown notion. Remember June is peak insolation month, and area is the parameter affecting albedo.

Area at June 1 well constrained by land hence it is practically a straight horizontal trend?

Practically a straight horizontal trend doesn't seem like a good argument for a near term BOE.

[oren]

Cherry-pick the June 1st graph for difficulty with the slowdown notion. Remember June is peak insolation month, and area is the parameter affecting albedo.

Area at June 1 well constrained by land hence it is practically a straight horizontal trend?

Practically a straight horizontal trend doesn't seem like a good argument for a near term BOE.

True, but I mentioned it, with its significant record low of 2016, as an argument against a gompertz-style slowdown with a 2007 inflection point.
If 2016 with its terrible winter had a 2012-like summer weather, we might already have been halfway to a BOE right then.

[crandles]

Hmm probably got part of my last answer wrong.

Would prefer a straight line to fit below.

CT area data stopped in early 2016, so haven't got last two years in this which would probably change direction of curve.

Edit:
Wondering if this means there are step changes in NSIDC area data as the pole hole has reduced in size which you haven't taken into account. That issue existed with their monthly data.

Edit 2: I see you recalled this before me.

[oren]

Hmm probably got part of my last answer wrong.

Would prefer a straight line to fit below.

CT area data stopped in early 2016, so haven't got last two years in this which would probably change direction of curve.

Looking at your chart, it seems I got my data wrong. As I couldn't find an immediate NSIDC area graph, I took ths NSIDC regional spreadsheet and summed the area of all the regions. Now I recall the pole hole issue, which changed size at some point. I will check the data later tonight.

[crandles]

If 2016 with its terrible winter had a 2012-like summer weather, we might already have been halfway to a BOE right then.

I tend to think this about 2017 rather than 2016 because of its low maximum volume. A record equalling melt volume would have taken us down to 1336km^3 for Sept so that would be getting fairly close to virtually ice free.

This 'if' didn't happen once (surprisingly getting a below average melt volume when the trend appears to be upward). Hard to get a feel of whether likely to happen soon from just one instance. However, this does make me nervous that my arguing for slowing down trend may be wrong.

[Daniel B.]

Here is some data to help ground the discussion. As ice grows more dispersed over time, I think area can often give better clues that extent.
A. It's a long gradual process, and human civilization will not end tomorrow one way or the other. (But wait 50 years and we're probably doomed.  But that's a different thread).
B. I don't think there's a slowdown. Cherry-pick the June 1st graph for difficulty with the slowdown notion. Remember June is peak insolation month, and area is the parameter affecting albedo.
C. As the trend is down, AND volatility to the downside is increasing, as stated up-thread I would be highly surprised if a BOE (<1M km2 extent (or area?) at min) doesn't show up by 2030.

I think area would give better clues than extent, if we have better area measurements.  Area and extent correlate quite well during the colder months, when a significant fraction of the Arctic is ice covered (area always being smaller).  However, in the summer months, melt ponds form on the ice, fooling the satellite sensors into thinking it is open water.  Extent measurements will not be affected as much, as these ponds will not cover enough ice to rate them as open water. 

This graph shows how the correlation between area and extent breaks down in the summer:

[oren]

Here is the real chart from NSIDC area, including the pole hole counted as full concentration. Luckily, it is now of negligible size.
Anyway, the June 1st curve still defies a statistical slowdown, in my view.

[oren]

I think area would give better clues than extent, if we have better area measurements.  Area and extent correlate quite well during the colder months, when a significant fraction of the Arctic is ice covered (area always being smaller).  However, in the summer months, melt ponds form on the ice, fooling the satellite sensors into thinking it is open water.  Extent measurements will not be affected as much, as these ponds will not cover enough ice to rate them as open water. 

This graph shows how the correlation between area and extent breaks down in the summer:

I believe August doesn't have lots of melt ponds, as a large fraction of them usually drain through the ice by then. I see this interesting chart as showing something different (in combination with melt pond and rain "noise") - the growing dispersion of the ice during late summer, as the pack becomes weaker and more mobile. One of those "ice internals" showing that the Arctic is not in a stable hiatus but on a downward trend.
In any case, it would be interesting to view this chart all the way to the present.

[Tor Bejnar]

... view this chart all the way to the present.

I want to see that chart "into the future".   
Wait; that's speculation;  wrong thread.

[Iceismylife]

...

We are referring to ice-free at minimum, not maximum.  It will be a very long time (if ever) before the Arctic will be ice-free during the winter.

First.  How long can we sustain the trend of the last four years being the lowest extent at maximum before the minimum reaches zero?  The minimum is a fraction and a function of the maximum.  The relation is complicated and not linear. But if you keep dropping maximum then the minimum should follow.
Second.  There has been some math done on this forum that shows that if you go ice free you stay ice free in the arctic ocean in the winter.  But you can see it in these two photos. The first one is in infrared the second is in visible light.
https://go.nasa.gov/2GeYlR6
https://go.nasa.gov/2GfvfB0

The cloud tops are cooler (darker) than the ice.  Open water will have less heat loss do to radiation than ice covered water.  It will have a lot more heat loss do to evaporation.  But with mixing from waves and other sources there is plenty of heat there to keep the water open after it goes ice free.

(And with those clouds going from one side of the arctic ocean to the other when they hit a rise on the other side they will drop a fraction of their water as snow. But that is for another thread)

[Daniel B.]

Considering that the minimum has not declined over the past four years, it is hard to say.  The minimum is not a direct function of the maximum.  If it were, the past four years would be the lowest, and not 2012 and 2007.  The ultra low minimum in 2012, was surrounded by relatively high maxima both prior to and following the summer minimum (high for the past fifteen years).  While intuitively, it would makes sense that the minimum would follow a falling maximum, in reality, that is not occurring.

I disagree that an ice-free Arctic in summer will lead to an ice-free winter.  Open water will lose more heat, and it will refreeze.  Additionally, the ice adjacent to the islands will spur further ice growth seaward.  The cloud albedo is only part of the radiation equation; the clouds themselves will absorb solar radiation and retaliate heat skyward, in the same manner that they absorb terrestrial heat and reradiate earthward.

[Iceismylife]

...

I disagree that an ice-free Arctic in summer will lead to an ice-free winter.  Open water will lose more heat,

I agree with you up to here.

and it will refreeze.

I disagree with you here.  Open water is subject to physical processes that ice covered water isn't.  It is subject to mixing by wave action.  The arctic ocean has a low salinity top layer that limits heat available to reject to the atmosphere.  Mix that layer in with the warmer lower water and there is plenty of heat to keep it from refreezing.

  Additionally, the ice adjacent to the islands will spur further ice growth seaward.

I agree with you.  Butt and this is a big one.  Generate enough bottom water and you reverse the surface flow past Greenland. That would pull the heat from the equator into the arctic ocean.

The cloud albedo is only part of the radiation equation; the clouds themselves will absorb solar radiation and retaliate heat skyward, in the same manner that they absorb terrestrial heat and reradiate earthward.

The gulf stream will melt all the ice that forms next to the islands.  Once ice free it will stay ice free. IMO.

[be cause]

Ice -free ? .. I give it 6 months (again ).

[Daniel B.]

The gulf stream will melt all the ice that forms next to the islands.  Once ice free it will stay ice free. IMO.

Currently, the glue stream does not make its way through the Canadian archipelago.  After passing Iceland, it flows into the Arctic on the Russian side.  By the time it reached the islands, it has circumvented the Arctic and cooled substantially.