When will the Arctic Go Ice Free?

[DrTskoul]

Thanks..

[binntho]

Very interesting article. Some points I stumbled on:

Of the 0.71 W/m2 of globally averaged heating, 0.21 W/m2 is estimated to have already occurred between 1979 and 2016.

So of the 25 years that AGW is to be sped up, slightly more than 8 have already been added?

The cloud-free, ice-free Arctic scenario results in a global radiative heating of 2.2 W/m2 compared with the 1979 baseline state, which is 3 times more than the 0.71 W/m2 baseline estimate derived above for unchanged clouds. The completely overcast ice-free Arctic scenario results in a global radiative heating of 0.37 W/m2, which is approximately half as large as the 0.71 W/m2 baseline estimate (Figure 2b).

Changes in cloud cover are indeed are very large unknown, but I get the impression that cloud cover has increased over the recent years, and that this is to be explained by more open water than before. So this would seem to be an important negative feedback, i.e. the more open water, the more cloud cover.

[Archimid]

That it, unsecured, but at least instantly available.

[Shared Humanity]

Very interesting article. Some points I stumbled on:

Of the 0.71 W/m2 of globally averaged heating, 0.21 W/m2 is estimated to have already occurred between 1979 and 2016.

So of the 25 years that AGW is to be sped up, slightly more than 8 have already been added?

The cloud-free, ice-free Arctic scenario results in a global radiative heating of 2.2 W/m2 compared with the 1979 baseline state, which is 3 times more than the 0.71 W/m2 baseline estimate derived above for unchanged clouds. The completely overcast ice-free Arctic scenario results in a global radiative heating of 0.37 W/m2, which is approximately half as large as the 0.71 W/m2 baseline estimate (Figure 2b).

Changes in cloud cover are indeed are very large unknown, but I get the impression that cloud cover has increased over the recent years, and that this is to be explained by more open water than before. So this would seem to be an important negative feedback, i.e. the more open water, the more cloud cover.

H2O is a powerful greenhouse gas and increased clouds in the fall and early winter in the Arctic have slowed down freeze by preventing heat from venting into space.

[Archimid]

Clouds. During the Arctic summer, while there is ice and the ocean is cold, clouds are our friends. During Winter, when there is no light to reflect, clouds are our enemies. If it was summer and there was no ice and the ocean was warm, clouds would also be our enemies during summer.

[petm]

Depends also on the type of cloud.

https://www.nsf.gov/news/special_reports/clouds/question.jsp

[binntho]

The type of cloud is absolutely important, and predicting changes to cloud cover is one of the most difficult aspects of any attempt to quantify future warming.

And as petm says, clouds in winter can keep the lid on the temperature. So all in all, a very tricky subject!

[Peter Ellis]

Not free. I followed your link, had to log in ( give information) and then directly request access to the paper. Haven't received it yet.

Here is the abstract of the paper:

During recent decades, there has been dramatic Arctic sea ice retreat. This has reduced the top‐of‐atmosphere albedo, adding more solar energy to the climate system. There is substantial uncertainty regarding how much ice retreat and associated solar heating will occur in the future. This is relevant to future climate projections, including the timescale for reaching global warming stabilization targets. Here we use satellite observations to estimate the amount of solar energy that would be added in the worst‐case scenario of a complete disappearance of Arctic sea ice throughout the sunlit part of the year. Assuming constant cloudiness, we calculate a global radiative heating of 0.71 W/m2 relative to the 1979 baseline state. This is equivalent to the effect of one trillion tons of CO2 emissions. These results suggest that the additional heating due to complete Arctic sea ice loss would hasten global warming by an estimated 25 years.

From here:

https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1029/2019GL082914

... or just use Sci-Hub to find papers <cough>

Looking at it...  this is a self evidently ridiculous paper.  It's a simple albedo calculation, based on the complete absence of ice from March through to September.

From their methods:

In the calculations of albedo and radiative heating presented here, we use data from the National Aeronautics and Space Administration's CERES Terra SSF Edition 4 monthly averaged 1 × 1-degree product, between March 2000 and October 2016, available online (https://ceres.larc.nasa.gov/order_data.php). The Arctic Ocean is defined here as the land-free area poleward of 60◦N. Due to issues concerning polar night, we only consider the months of March to September of each year.

i.e. they account for extra incoming energy during an ice-free summer, but do not account for any extra outgoing energy during the winter.  This is effectively equivalent to assuming that the Arctic freezes over as normal each winter (i.e. insulating the surface and restricting heat loss) and then the ice magically disappears overnight some time in March.

It's completely unphysical and the only utility it has is giving a ballpark figure for HALF the energy equation resulting from an ice-free Arctic. Don't waste any time trying to interpret this one any further than that.

[Klondike Kat]

I guess that is like trying to calculate the earth’s energy budget by focusing solely on incoming solar energy during the daytime hours.

[Archimid]

Might as well post it here too. From the paper now bashed by the usual suspects:

[Yuha]

It's completely unphysical and the only utility it has is giving a ballpark figure for HALF the energy equation resulting from an ice-free Arctic. Don't waste any time trying to interpret this one any further than that.

The paper actually says that (though its hidden in the discussion section):

The present study focuses on the additional radiative heating from the complete loss of Arctic sea ice, but it does not estimate the amount of global warming that would be associated with this level of ice loss.

[petm]

Looking at it...  this is a self evidently ridiculous paper.  It's a simple albedo calculation, based on the complete absence of ice from March through to September.

Totally. You should be able to do a similarly simple calculation of the increased outgoing radiation assuming, say, an ice-free arctic all winter and also get it published in Geophysical Research Letters. Simple.

Never mind that the climate suggested by an ice-free winter arctic would make such a cooling effect moot.

And never mind that the authors make the most conservative possible assumption about cloud cover.

And never mind that you could just use the curves reported in Fig. 3 to calculate increased forcing for whatever shorter period you liked.

It's obviously a totally irrelevant publication. Move along.

[Archimid]

i.e. they account for extra incoming energy during an ice-free summer, but do not account for any extra outgoing energy during the winter.

If they did account for "extra outgoing energy" they would find that 25 years of earlier warming is an underestimation.

[KiwiGriff]

Visible light warms the ocean as far as it can penertrate. In blue water this can be to 50 meters or more.
The ocean radiates energy in the infrared from a surface skin of only a few microns depth.
As long as it is still cold enought to ice over the surface film in winter once the sun is gone and insulate the water belowe the winter radiation  is moot .
The deeper ocean will still hold most the energy it has absorbed over an ice free summer .
 The amount of energy radiated from the ocean to space will not change signifactly until the ice is gone year around .
We will be lucky to still be able to observe the arctic if it does get that warm.

[binntho]

Visible light warms the ocean as far as it can penertrate. In blue water this can be to 50 meters or more.

Sunlight penetrates up to 1000 meters in ocean waters, depending on circumstances, but almost all of the red and infrared wavelengths are absorbed more or less as soon as they hit the water (which is why water is blue - same applies to ice). I wasn't able to find numerical values but I suspect that the vast majority of heat input is grabbed in the top one or two meters.

The ocean radiates energy in the infrared from a surface skin of only a few microns depth.
As long as it is still cold enought to ice over the surface film in winter once the sun is gone and insulate the water belowe the winter radiation  is moot .

Water becomes denser as it cools, and will sink if there is less dense water below it, so a cooling surface becomes a heat sink for much more than just the surface skin.

Additionally, ocean water is densest at point of freezing (because of the salt) and will not freeze until the water column is sufficiently cold to such a depth that surface waters stop sinking fast enough to avoid freezing.

All in all this means that the top layers grab most of the heat from the sun, and once the sun goes down, the top layers keep on losing heat as long as they can. Once the surface starts freezing over, a very substantial majority of the accumulated solar heat input will have been lost.

A huge negative feedback in other words, and one that grows in tandem with less summer ice and higher SSTs.

[Klondike Kat]

Yes, open water in winter is a huge negative feedback, as substantial heat will be lost to the atmosphere.  Hence, a year-long ice-free Arctic will not contribute as much to global temperatures as a seasonal one. 

Open water will facilitate an Arctic re-freeze, and at a faster rate.  Just compare 2012 to 2013; 2013 had 50% more ice at minimum compared to 2012 (NSIDC extent), but they were almost equal by years' end, due to the higher rate of freeze in 2012.  The same occurred between 2006 and 7.

https://nsidc.org/arcticseaicenews/charctic-interactive-sea-ice-graph/

[Archimid]

The ocean doe not vent the heat out to space. In vents it out to the atmosphere. The more heat in the ocean the warmer the atmosphere gets. The warmer it gets the less ice is created, regardless of how much heat the ocean vents.

Clouds and a more humid atmosphere reinforce that positive feedback.

[Klondike Kat]

The ocean doe not vent the heat out to space. In vents it out to the atmosphere. The more heat in the ocean the warmer the atmosphere gets. The warmer it gets the less ice is created, regardless of how much heat the ocean vents.

Clouds and a more humid atmosphere reinforce that positive feedback.

Without a significant cloud cover, the heat vented to the atmosphere will be lost to space. 

Ice formation is a direct result of water temperature only.  Any contribution from the atmosphere will be in how it affects the water temperature.

[Sterks]

Yes, open water in winter is a huge negative feedback, as substantial heat will be lost to the atmosphere.  Hence, a year-long ice-free Arctic will not contribute as much to global temperatures as a seasonal one. 

Open water will facilitate an Arctic re-freeze, and at a faster rate.  Just compare 2012 to 2013; 2013 had 50% more ice at minimum compared to 2012 (NSIDC extent), but they were almost equal by years' end, due to the higher rate of freeze in 2012.  The same occurred between 2006 and 7.

https://nsidc.org/arcticseaicenews/charctic-interactive-sea-ice-graph/

So why did it fail between 2016 and 2017?

[binntho]

Yes, open water in winter is a huge negative feedback, as substantial heat will be lost to the atmosphere.  Hence, a year-long ice-free Arctic will not contribute as much to global temperatures as a seasonal one. 

Open water will facilitate an Arctic re-freeze, and at a faster rate.  Just compare 2012 to 2013; 2013 had 50% more ice at minimum compared to 2012 (NSIDC extent), but they were almost equal by years' end, due to the higher rate of freeze in 2012.  The same occurred between 2006 and 7.

https://nsidc.org/arcticseaicenews/charctic-interactive-sea-ice-graph/

So why did it fail between 2016 and 2017?

Looking at volume, (e.g. here https://sites.google.com/site/pettitclimategraphs/sea-ice-volume#asivmaaappolar) one can see the deep dips of 2007 and 2012 with sharp rebounds, particularly after 2012. But 2016 isn't such an obvious dip and a rather weak rebound.

Extent also shows some rebound but not really that much.

So other factors are clearly at play, could it be increased cloud cover during winter?

[Archimid]

Ice formation is a direct result of water temperature only.  Any contribution from the atmosphere will be in how it affects the water temperature.

So freezing degrees days is a useless metric. If we just use sst we can have a better estimate of ice formation? 

Nonsense.

[DrTskoul]

The ocean doe not vent the heat out to space. In vents it out to the atmosphere. The more heat in the ocean the warmer the atmosphere gets. The warmer it gets the less ice is created, regardless of how much heat the ocean vents.

Clouds and a more humid atmosphere reinforce that positive feedback.

 

If there are no clouds the heat eventually is released in space. That is why you can reach freezing temperatures in the desert in an open sky at night.

Clouds though  keep essentially a blanket over the ocean and reduce its  rate of cooling.

[Archimid]

yep, a cloudless, dry atmosphere would indeed quickly cool the Arctic. I just don’t know why KK expects a cloudless, dry atmosphere with an open warm ocean right underneath it and with the hemisphere bombarding the Arctic with converted heat and even more humidity.

[Klondike Kat]

Ice formation is a direct result of water temperature only.  Any contribution from the atmosphere will be in how it affects the water temperature.

So freezing degrees days is a useless metric. If we just use sst we can have a better estimate of ice formation? 

Nonsense.

freezing degree days is useful for estimating ice thickness and growth of sea ice, not its initial formation.  Water still freezes at temperatures below its freezing point (0C for fresh water, lower for salt water).

[Archimid]

 When the ice thickens new ice forms. 

The thicker the ice the least heat flux to the atmosphere. Thinner ice allows more heat flux into the atmosphere which warms the atmosphere. A warmer atmosphere prevents ice formation. Simple.

[Klondike Kat]

When the ice thickens new ice forms. 

The thicker the ice the least heat flux to the atmosphere. Thinner ice allows more heat flux into the atmosphere which warms the atmosphere. A warmer atmosphere prevents ice formation. Simple.

So that the less ice present, the more heat the flows into the atmosphere, and is lost to space, cooling the Arctic.

[Archimid]

The Arctic is not the moon. Heat has to pass trough the atmosphere, warming it. A warmer atmosphere means less freezing and less freezing means a warmer atmosphere.

How much the atmosphere warms depends on the greenhouse gases it is comprised of. Water is the biggest greenhouse gas. An open water Arctic ocean WILL increase its water content, which will hold more heat in an already warmer Arctic.

By the way extent or area growth IS NOT ice formation. They are both 2 dimensional measure of surface area. Ice is 3D.

[Shared Humanity]

...area growth IS NOT ice formation.

 

[petm]

...area growth IS NOT ice formation.

 

Area grows due to dispersion as well, as witnessed in recent days, in which case it is likely accompanied by melt not formation.

[Archimid]

It is not. Area growth is merely an increase of the area covered by the ice. A 2D abstraction. A useful 2D abstraction.

It is a terrible mistake to conflate area or extent growth/reduction with ice freezing/melting. 

[binntho]

How much the atmosphere warms depends on the greenhouse gases it is comprised of. Water is the biggest greenhouse gas. An open water Arctic ocean WILL increase its water content, which will hold more heat in an already warmer Arctic.

I somehow feel a confusion here between heat content of the various gases and their greenhouse effect.

Greenhouse gases don't really warm up as such (or rather, they do not have a specific heat different from other non-greenhouse gases), but they do re-radiate heat, including back to the surface. Clouds on the other hand (with water in it's liquid form) do accumulate heat, and will store it or re-radiate as they themselves cool down.

Cloudy weather also indicates a higher level of water vapor, increasing the local greenhouse effect.

But the two things - cloud effect and greenhouse effect - are very different.

[binntho]

Been wondering whether the heat capacity of the atmosphere above the Arctic ocean would be able to accommodate all the energy lost when the top few meters of ocean cools. Turns out that, with a bit of clouds to help, it isn't that unreasonable for the atmosphere to catch the released energy.

But clouds would certainly be a very big help, as well as any water wapor going around. In dry air, a hell of a lot of heat will radiate up through the atmosphere (or be re-radiated) and end up in space.

[Klondike Kat]

Been wondering whether the heat capacity of the atmosphere above the Arctic ocean would be able to accommodate all the energy lost when the top few meters of ocean cools. Turns out that, with a bit of clouds to help, it isn't that unreasonable for the atmosphere to catch the released energy.

But clouds would certainly be a very big help, as well as any water wapor going around. In dry air, a hell of a lot of heat will radiate up through the atmosphere (or be re-radiated) and end up in space.

Not to mention that cloud cover will significantly reduce incoming radiation.  Clouds (or lack thereof) work both ways in the overall heat transfer process. 

[binntho]

Been wondering whether the heat capacity of the atmosphere above the Arctic ocean would be able to accommodate all the energy lost when the top few meters of ocean cools. Turns out that, with a bit of clouds to help, it isn't that unreasonable for the atmosphere to catch the released energy.

But clouds would certainly be a very big help, as well as any water wapor going around. In dry air, a hell of a lot of heat will radiate up through the atmosphere (or be re-radiated) and end up in space.

Yes, but for the purpose of the current discussion, incoming radiation is not really an issue since we are talking about heat loss after summer minimum.

Not to mention that cloud cover will significantly reduce incoming radiation.  Clouds (or lack thereof) work both ways in the overall heat transfer process.

[Renerpho]

The question of this thread has been answered. DMI has the arctic ice free today. 

https://web.archive.org/web/20190804121548/http://ocean.dmi.dk/arctic/icedrift_anim/index.php

[Stephan]

It is time for the monthly update of my extrapolation when the extent [Extent], volume [Volumen] and thickness [Dicke] will reach zero. The extrapolation occured linearly and by a logarithmic function; the latter one almost constantly resulting in earlier times (valid for volume and thickness, not for extent). July value now includes 2019.
Extent value, volume and thickness for July 2019 lie below the long-term linear trend lines. These anomalies are significant, therefore the "BOE numbers" decreased by 1 year (thickness), 3 years (extent) and 5 years (volume) compared to last year.

Please note that this is not a forecast but a trend!
See attached table.

[DrTskoul]

It is time for the monthly update of my extrapolation when the extent [Extent], volume [Volumen] and thickness [Dicke] will reach zero. The extrapolation occured linearly and by a logarithmic function; the latter one almost constantly resulting in earlier times (valid for volume and thickness, not for extent). July value now includes 2019.
Extent value, volume and thickness for July 2019 lie below the long-term linear trend lines. These anomalies are significant, therefore the "BOE numbers" decreased by 1 year (thickness), 3 years (extent) and 5 years (volume) compared to last year.

Please note that this is not a forecast but a trend!
See attached table.

So some time between 2027 and 2073, with the most likely timeframes between 2035 and 2060

Dicke... really ?

[Klondike Kat]

It is time for the monthly update of my extrapolation when the extent [Extent], volume [Volumen] and thickness [Dicke] will reach zero. The extrapolation occured linearly and by a logarithmic function; the latter one almost constantly resulting in earlier times (valid for volume and thickness, not for extent). July value now includes 2019.
Extent value, volume and thickness for July 2019 lie below the long-term linear trend lines. These anomalies are significant, therefore the "BOE numbers" decreased by 1 year (thickness), 3 years (extent) and 5 years (volume) compared to last year.

Please note that this is not a forecast but a trend!
See attached table.

So some time between 2027 and 2073, with the most likely timeframes between 2035 and 2060

Dicke... really ?

Yes, I am a little surprised that thickness is decreasing faster than extent.  If the ice was melting uniformly, one would expect volume decreasing faster than extent than thickness (3 dim. > 2 dim. > 2 dim.).

[philopek]

Yes, I am a little surprised that thickness is decreasing faster than extent.

Since 1cm of ice can still cover the same area and extent can even be higher due to fragmentation and dispersion, it makes totally sence that Volume & Thickness shrink first.

Just image that thin ice will break up more and cause higher extent than a solid (thick) ice layer of the same area.

[KiwiGriff]

 
My comment above got many attempts to shoot it down in flames.
I was guilty of grossly over simplifying due mostly to a comment made on a cell phone and my laziness. My expectation was that the audience would know enough to read what I was implying not what I said.
Science folks, the art of being less wrong.
I maintain I was less wrong than some of the replies.

We have claims that a feed back,more  heat escaping from the open water  over winter due to low extent, has resulted in a change in the long term trend of decline.
This seems to me to be based on cognitive bias, a selective use of the available data and a curve fitting exercise.
Noise.
Using the effects of a few weather events to add an extra term into the curve of decline is questionable. Such ideas strongly reminds me of the so called "pause" in warming that certain sectors pushed after 1998's excursion from the long term trend in global mean surface temperatures.
Ocean heat content below the ice has continued its rise. Atmospheric temperatures above the ice have continued to increase. Volume,thickness and area has continued to fall.
None of these metrics support a change to the shape of the long term trend.

At some point in the next decade or two we will see another weather event that results in a BOE.
Extent  will then revert back to the trend in the following years. After a single BOE has happened once it will happen again with increasing frequency.
Such events will  hasten a blue ocean over summer becoming a permanent state. 

FWIW.
DK?
I know just enough to know I don't know enough to know how much I don't know.

[Stephan]

Yes, I am a little surprised that thickness is decreasing faster than extent.

Since 1cm of ice can still cover the same area and extent can even be higher due to fragmentation and dispersion, it makes totally sence that Volume & Thickness shrink first.

Just image that thin ice will break up more and cause higher extent than a solid (thick) ice layer of the same area.

Here are some 10y averages of extent, volume and thickness for July:
1979-1988:  10,16 M km²   -    21,42 k km³   -   2,11 m
2010-2019:    7,98 M km²   -    10,13 k km³   -   1,27 m
loss:             21,5 %         -      52,7 %         -   39,9 %
It is obvious that an extrapolation gives earlier results for a BOE for volume, followed by thickness, last is extent.

[RikW]

Yeah, though I guess volume won't reach zero much earlier than extent

[Stephan]

If the thinning of the ice sheet continues, then - close to the end - the area will relatively decrease much faster than the volume, so all three measures will reach zero at the same time.

[philopek]

Yeah, though I guess volume won't reach zero much earlier than extent

haha... it's all meant for the current state, current extent range, current area range etc.

BTW if we consider that a big part of the ice is under water, we could technically (sat-measurement-wise) indeed see kind of close to zero ice extent, while volume can be higher than close to zero, but that's kind of niggling, of course in general it's what you hint at with this

[Sterks]

Isn't the ESS way of vanishing this year a good example of what you guys are discussing? Thickness approaches zero earlier than extent...

[Klondike Kat]

Unless of course the ice forms an infinitely thin layer.  Then volume could theoretically reach zero decades before the extent.  Although thickness would reach zero at the same time as volume. 

[philopek]

Isn't the ESS way of vanishing this year a good example of what you guys are discussing? Thickness approaches zero earlier than extent...

Yes, approaches, but once zero is almost reached, as stated above, extent will be zero but there will be still ice below the 15% threshold and that would account for existing matter of fact volume and the greater part of that rest-ice will be below sea-level, hence invisible by sensors but still there and accounting for real-live-volume, existing, no matter what anyone calculates.

Of course that will be quite a small number and very short lived, hence a kind of irrelevant transition phase toward zero zero zero.

[crandles]

If the thinning of the ice sheet continues, then - close to the end - the area will relatively decrease much faster than the volume, so all three measures will reach zero at the same time.

True.

Of course this doesn't rule out area rate of decline slowing down and to meet zero at the same time, thickness and volume decline slowing down much more.

Here are some 10y averages of extent, volume and thickness for July:
1979-1988:  10,16 M km²   -    21,42 k km³   -   2,11 m
2010-2019:    7,98 M km²   -    10,13 k km³   -   1,27 m
loss:             21,5 %         -      52,7 %         -   39,9 %

Care to add figures for 2000-2009 and 1990-1999 so we can see how loss%
from 2000-2009 to 2010-2019
compares with loss% figures
from 1990-1999 to 2000-2009
?

We have claims that a feed back,more  heat escaping from the open water  over winter due to low extent, has resulted in a change in the long term trend of decline.
This seems to me to be based on cognitive bias, a selective use of the available data and a curve fitting exercise.
Noise.
Using the effects of a few weather events to add an extra term into the curve of decline is questionable. Such ideas strongly reminds me of the so called "pause" in warming that certain sectors pushed after 1998's excursion from the long term trend in global mean surface temperatures.
Ocean heat content below the ice has continued its rise. Atmospheric temperatures above the ice have continued to increase. Volume,thickness and area has continued to fall.
None of these metrics support a change to the shape of the long term trend.

There is more to it than the 'heat loss depending on thin ice' portrayed here.
Thick MYI ceases making it around Beaufort gyre. The area of such thick MYI plummeted over 2000-2012 period. When thick MYI disappears, the thickness above that which can be grown in a winter does grow back. When FYI melts it largely grows back each winter.

Is this 'selective use of the available data' or is it a more detailed study of the situation rather than a gut reaction that things are going to get worse at increasingly rapid rate?

Noise? That may well still be possible rather than the data showing it is statistically significant. If it isn't statistically significant, I think the data is still suggestive. Gompertz curves have been suggested around here since at least 2011 and the data seems to be showing that has been a much better model than linear/exponential/other curves heading towards steeper curves.

Then there are the climate models pretty well all of which show a declining rate of loss of extent as zero ice is approached.

[Stephan]

There you go, crandles:

10y averages of extent, volume and thickness for July:
1979-1988:  10,16 M km²   -    21,42 k km³   -   2,11 m
1990-1999:    9,60 M km²   -    19,08 k km³   -   1,99 m
2000-2009:    8,87 M km²   -    15,36 k km²   -   1,73 m
2010-2019:    7,98 M km²   -    10,13 k km³   -   1,27 m

decadal decreases:
1990s vs 79/88:    5,5 %     -     11,1 %         -      5,7 %
2000s vs 1990s:    7,6 %     -     19,5 %         -    13,1 %
2010s vs 2000s:  10,0 %     -     34,0 %         -    26,2 %

My interpretation:
1. The decrease rate of all measures increases from decade to decade. Acceleration of Arctic Sea Ice melt.
2. The decrease in volume increases faster than the decrease in extent.
3. The decrease in thickness is accelerating in the 2010s → probably the loss of most of the MYI from 2007 on. More than a doubling from decade to decade.

[Glen Koehler]

X2 curvilinear regression.  R2 = 99.99% 
"1985" = midyear of 1979-1988, etc.

Year          Model estimate of Arctic Sea Ice volume at September minimum.
1985   21.41
1995   19.10
2005   15.34
2015   10.14
2025   3.49
2029   0.42
2030   "-0.37"

Conclusion:  If current volume loss trend continues, then around 2030-32, the September minimum will have virtually no Arctic Sea Ice.  Thus, all ice in following spring will be FYI from preceding winter.

But a regression curve extended beyond the data range can give overly aggressive prediction for rate of change.  A straight line regression has fewer assumptions, is more conservative and robust, Occams Razor etc.  Wipneus' straight line trend shows ASI volume hitting zero in 2032.

https://14adebb0-a-62cb3a1a-s-sites.googlegroups.com/site/arctischepinguin/home/piomas/grf/piomas-trnd7.png?attachauth=ANoY7crKtav-HYAq4lRVrck3FefppueqqYumZJyyPRWqKeDnVn1i8vijnas0L9wGuj90u1IgAAvG2UeObcryXJ77z5qU9jM1uHAoHO1XyoMAE8O49bSCo6hk-I7e-Z9KtLVBcMeJ_CBthasCWK-XUOGsyZEHNPREv6-IYzu2jSl9DWNUa2dqU71ccpWI5fz9YjX1PLecXkAIY2goroSLvb6G3JYTZ8hN837VDsLhctORifdxsAQxnjD8wAJ-C8ENPMFbXVsgh03e&attredirects=0

(Requires second click to download graph)

    The Wipneus graph shows 2019 being just about matching the midline estimate, slightly higher  than 2012.  By 2020 the midline estimate matches 2012, then trend reaches zero in 2032.  Lots of variation around the midline estimate of course, but seeing 2019 land right about where projected lends credence to regression validity and to the idea that very soon every year is likely to have less ice at minimum than 2012.   

Curved regression applied to Thickness shows zero at 2033. 
Extent curved regression does not reach zero until 2070. 
But no volume = no ice for Extent.

Thus on current trajectory in about 11-13 years (2030-32) human-caused climate change may have so altered the Arctic Sea Ice as to cause fundamental functional change to a keystone physical component of the Earth's climate system.

Meanwhile -- pundits, politicians and economists discuss the fine points of other issues as if they matter more than the planetary life support system.  Other issues are important, but human civilization relies on a supportive climate system, so not destroying that must take precedence.