When will the Arctic Go Ice Free?

[gerontocrat]

I have been doing my spreadsheet housekeeping and found a lot of stuff about BoE all over the place. Having put it into better order and updated the data here are some thoughts. (all output derived from data from the NSIDC and PIOMAS standard files.

What is the definition of a BOE? I sort of think it is described as 1 million km2 or less of Arctic sea ice extent for the whole month of September.

BUT. As the melting season progresses sea ice concentration reduces - always.
The Central Arctic region (CAB) at maximum has a concentration of nearly 100%. At minimum, after losing about 25% of its sea ice extent, concentration drops to around 70%.

The Beaufort sea at maximum has a concentration of nearly 100%. At minimum, after losing about 75% of its sea ice extent, concentration drops to around 35%.

An Arctic Ocean with just 1 million km2 sea ice extent is probably to have at most a sea ice area of half that, i.e. a max of 500,000 km2.

So on a definition for the BOE of a maximum extent of 1 million km2, we are talking about just half a million maximum area of sea ice in the 7 central seas with an area of 8.9 million km2, i.e. 94% or more open water.

But, that seems to be the consensus. Ho hum.

click images to enlarge

[gerontocrat]

But what i wanted to show was

Reductio Ad Absurdum - a method of proving the falsity of a premise by showing that its logical consequence is absurd or contradictory:

I have taken the NSIDC sea ice area and PIOMAS sea volume averages for September and combined them to produce the graph of average thickness. The graph uses actuals to 2020 and then projections based on the long-term linear trends. I have done this for the entire Arctic and the 7 central seas of the High Arctic.

In percentage terms the annual reduction in volume is far higher than that of sea ice area. hence the reducing average thickness.

Reduction Ad Absurdum is achieved in around a decade. To avoid absurdity, volume reduction must reduce substantially. On the other hand sea ice area and /or volume reduction could speed up and a BOE happen earlier.

I also looked at October data for the High Arctic for 2 reasons. In the last 2 years October sea ice area has reduced to record lows, and the scientists seem to tell us that late refreezing is likely to accelerate as AGW + Polar Amplification proceeds.

The result is Absurdity is reached just a couple of years later. i.e. a 2 months ice-free Arctic

CONCLUSION ?: Somethings gotta give, What, where and when?

click images to enlarge

[Glen Koehler]

<snip>
Reduction Ad Absurdum is achieved in around a decade. To avoid absurdity, volume reduction must reduce substantially. On the other hand sea ice area and /or volume reduction could speed up and a BOE happen earlier.

I also looked at October data for the High Arctic for 2 reasons. In the last 2 years October sea ice area has reduced to record lows, and the scientists seem to tell us that late refreezing is likely to accelerate as AGW + Polar Amplification proceeds.

The result is Absurdity is reached just a couple of years later. i.e. a 2 months ice-free Arctic

CONCLUSION ?: Somethings gotta give, What, where and when?

      If I understand that the question is how to resolve incompatible trends for Area vs. Volume, then my response is: Area gives more than Volume.  (Almost) no Volume dictates (almost) no Area.  And the Volume trend looks robust.
   
      "Almost" because I think trends will break down for the final residual ice Volume and Area in protected inlets etc. along the CAB-CAA border.  But that is pocket change in comparison to the bigger issue, which is "How long can the open Arctic Ocean continue to sustain these loss trends and retain more than a small residual of ice in protected areas?"

      Mathematically the trend lines must meet at zero.  So one trend line has to bend to meet the other.  One mathematical option is for the Volume line to bend upwards so that it does not hit zero before the Area line.  And as noted above, to some degree that may happen.  But I don't think it will happen very much.  Thus, while Volume in September is not likely to reach 0.000 in September, I think it is likely to get so close that it might as well be zero in terms of the effects of ASI on albedo and weather, ecosystem function, ship passage etc.

      So if the Volume line does not bend up to meet the Area line, then the Area line must bend down to meet the Volume line as they both near zero.  I think Area decline will accelerate as the ice that remains becomes progressively easier to melt due to qualitative changes in the ice at the micro level (salinity, density) and also at the macro ice pack level (more fracturing, mobility, export, open water albedo decline, warm air mass incursion, storminess, wave action, Atlantification etc.)

      The Absurdity will be resolved and disappear, as of course mathematically it must.  The real Absurdity is that we would allow such a drastic change to a critically important component of the planet's climate regulatory system.  The irrefutable fact that worries me the most is that the energy required to warm a gram of water by 1 degree C is 1/80 the energy required to melt a gram of ice.  The ice serves as a huge buffer to absorb warming energy being added to the system year after year.  The excess energy going into heating the Arctic Ocean will be more effective at doing so once the ice is gone.

      The observation trends are clear that September loses its ice first, with August and October not far behind.  So by that logic, I raise the stakes by calling Gerontocrat's bet for two months of ice-free Arctic absurdity by the early 2030s, and raise the bet to 3-months of an essentially ice-free Arctic by 2035 or earlier. 

      The CMIP3 to CMIP5 to CMIP6 globals models show a trend of projecting earlier dates for 1M km2 Sept Extent.  But they still don't call for even 1 month of 1M km2 "BOE" much before 2050.  See attached graphs below, from Stroeve et al. (2012), Geophys. Res. Lett., 39, L16502, doi:10.1029/2012GL052676 and SIMIP Community (2020). Geophysical Research Letters, 47, e2019GL086749. https://doi.org/10.1029/2019GL086749.  The blue line in first graph is CMIP3, the red line CMIP5.  In the bottom graph, the brown-gold line for CMIP6 is same RCP4.5 scenario as the red line in the CMIP5 graph.

      On one hand it's "What the heck do I know?"  On the other hand, it's PIOMAS, HYCOM, NSIDC, Wipneus, Gerontocrat, Stephan, and ASIF showing numbers that clearly indicate trends running far ahead of expert projections, and which show no sign of reversing course.  It's like the Arctic modelers are saying "Who ya gonna believe -- Me or your lyin' eyes?"  For me, the Eyes have it.  It is not the math that is Absurd, it is the reality it describes, and the fact that we are creating that reality.  Or as Shakespeare might have said, "The Absurdity, dear Brutus is not in our math, But in ourselves."

[Tor Bejnar]

2 meter-thick ice in the Beaufort Sea in May is more likely to melt out during the summer than 2 meter-thick ice in May that drifts near the North Pole during the summer.

CAB sea ice is more resilient than ice in the surrounding seas - it is farther north and the melting season starts later; northern ice continues to thicken (bottom freezing) after southern ice extent starts decreasing (bottom and top melting). The ice-melt charts (showing rates of volume and extent loss over the years) are usually made for the entire Arctic, so they include  southern areas.  Nearing the BoE, most remaining ice will likely be in the CAB.

Therefore, I suggest volume loss will slow down more than will extent/area loss as we approach a BoE, but both will slow.

[Glen Koehler]

      Good point.  I would have more faith in latitudinal delay if the final ice was centered around the North Pole.  The current HYCOM thickness model shows bulk of the thick ice being well south of 80N in Beaufort and along ESS shore.  PIOMAS thickness is less southerly but somewhat similar to HYCOM, and even CS2SMOS which shows thick ice overall at higher latitude still does not have it centered around 90N.
      Apologies to whomever I copied the attached graph from that shows the expected location of the final 1M km2 ASI Extent.  Current distribution may be a one-year exception, but in March 2021 the thick ice is centered well south of where the final 1M km2 of ice graph says it should be centered.
     While true that higher melt rate in southern locations has affected the Vol., Ext. etc. ice loss trends, in terms of Sept. minimum we are already in the era when almost all the ice remaining ice at end of melt season is in the CAB.  Gerontocrat's "River Valley" Area graphs for individual seas https://forum.arctic-sea-ice.net/index.php/topic,3426.msg303130.html#msg303130  shows that Barents, Greenland, Hudson Bay have been melting out for quite a while.  And that except for the Beaufort Sea and the CAB, the other high Arctic seas (ESS, Laptev, Kara) are already reaching or nearly reaching zero Area late in the melting season.
     Other Gerontocrat Volume graphs make the same point for Chukchi, CAA, and show even the Beaufort getting near zero.  That leaves the CAB as the last wall, and its late-season Volume is also less than what it used to be. 
     If getting down to the CAB as the final stand is going to slow down the trends, that should become apparent in a slowing of the trend lines over the next few years because we are already into the 'CAB last stand' era.  The later the Volume trend starts to slow the harder it has to bend away from zero.  The current straight-line trend has it reaching zero in 2033.

[Stephan]

I may recall the discussion we had some months ago which I had called "trend of the trend lines". All months between July and November show a strong convergence (that means the slope of the "late" values extent and area is getting steeper year by year which means that the BOE date shifts to earlier times in comparison to more or less stable slopes of the "early" values volume and thickness which do not affect the predicted BOE date). I will follow this month by month in my monthly posting of the BOE numbers.

Unlike the summer/autumn months there is a divergence for the months Jan to May...

[crandles]

      Mathematically the trend lines must meet at zero.  So one trend line has to bend to meet the other.  One mathematical option is for the Volume line to bend upwards so that it does not hit zero before the Area line.  And as noted above, to some degree that may happen.  But I don't think it will happen very much.  Thus, while Volume in September is not likely to reach 0.000 in September, I think it is likely to get so close that it might as well be zero in terms of the effects of ASI on albedo and weather, ecosystem function, ship passage etc.

      So if the Volume line does not bend up to meet the Area line, then the Area line must bend down to meet the Volume line as they both near zero.

False dichotomy: They can both bend up (or down)

And the models tend to show they do both bend up as zero is approached.

[gerontocrat]

In the end, AGW wins, even in winter

[crandles]

In the end, AGW wins, even in winter

So sure it shouldn't be:

If GW continues at pace for long enough, yes GW will eventually win.

St Laurence is fairly low latitude. One example is nice, but if it picks lowest latitude place and doesn't show evidence of this moving to higher latitude at a rapid rate ... then why shouldn't I think this will be harder to achieve at higher latitudes that have months longer of darkness and seems to be taking its time to move much further north.

Concentration trends seem to affect wide bands in summer months and only narrow bands in winter.
https://nsidc.org/data/seaice_index/compare_trends

Bands of blue represent 40 years of Northward progress and may look like it will take much less than 40 years in September but that is assuming it continues at same rate. But we have had a fast rate of decline 2000ish to 2010ish and now the rate of retreat seems to be getting slower and that is what models suggest will happen.

[The Walrus]

      Mathematically the trend lines must meet at zero.  So one trend line has to bend to meet the other.  One mathematical option is for the Volume line to bend upwards so that it does not hit zero before the Area line.  And as noted above, to some degree that may happen.  But I don't think it will happen very much.  Thus, while Volume in September is not likely to reach 0.000 in September, I think it is likely to get so close that it might as well be zero in terms of the effects of ASI on albedo and weather, ecosystem function, ship passage etc.

      So if the Volume line does not bend up to meet the Area line, then the Area line must bend down to meet the Volume line as they both near zero.

False dichotomy: They can both bend up (or down)

And the models tend to show they do both bend up as zero is approached.

Yes, the models do show that.  I think it is less of a false dichotomy than one has to (mathematically) bend [more] to meet the other.  I feel that it is more likely that volume will bend further to meet area.

[Glen Koehler]

     The ice will decide so our words are like a pea shooter aiming at a battleship.  But FWIW, this graphic gives a nice single view of how all the seas except CAB are already reaching or nearing zero Area for extended periods in latter portion of melt season, and the CAB is also moving in that direction.
   Image from Oren post at https://forum.arctic-sea-ice.net/index.php/topic,3447.msg303886.html#msg303886

[interstitial]

I value the information provided the regional ice graphs but my experience is its to hard to find information or insights when staring at the whole thing at once. Opening each graph at full screen size and comparing it to images and other graphs is the only way I learn from them.

[gerontocrat]

I value the information provided the regional ice graphs but my experience is its to hard to find information or insights when staring at the whole thing at once. Opening each graph at full screen size and comparing it to images and other graphs is the only way I learn from them.

Because I downloaded the Wipneus detailed file on regional area and extent (to allow thickness calculations with PIOMAS data) I can give you the CAB graphs according to Wipneus (Area 4.457 million km2)  against the graphs of the NSIDC defined area of  3.224 million km2.

The NSIDC data shows a shallower and somewhat wider profile as the area is more or less confined to North of 80, while the Wipneus CAB extends a couple of degrees farther south and thus includes 1.2 million km2 additional area that now usually all melts out..

[Stephan]

It is time for the monthly update of my extrapolation when the extent [Ausdehnung], volume [Volumen], thickness [Dicke] and area [Fläche] 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 and area in the winter months). The March value now includes 2021.

Position of March 2021 towards the long-term trend lines
Thickness, area, volume and extent are above the linear trend line.

Trend of the trends
The "BOE numbers" increased by 1 year (thickness), 2 years (volume), 17 years (area) and by 8 years (extent) compared to March 2020.
So there is once again a divergence between the "late values" (area, extent) and the "early values" (volume, thickness) which seems to be usual as I already posted for the "high winter months".
The order (earlier → later BOE) generally is volume < thickness < area < extent.

Please note that this is not a forecast but a trend!
See attached table, now widened to see the linear function value (y-AA) at t = 0. Stg = slope.

Click to enlarge it.

[gerontocrat]

When will the Arctic Go Ice Free IN WINTER?

Not very soon.

With April gone the 3 months of maximum sea ice area are past, and therefore also the minimum open water. So I can give some of my open water graphs an airing.

I attache 3 graphs. open water of the 14 seas, open water of the 7 seas of the High Arctic, and open water of the 7 Peripheral Seas.

Total Arctic Seas open water has increased from around 20% in 1980 to almost 30% in 2021.

In the High Arctic winter open water has barely changed.

For the Peripheral Seas that percentage has increased from a bit under 40% to a bit over 50%, though this varies greatly from sea to sea. (see next post)

click images to enlarge

[gerontocrat]

NSIDC data - variation of winter open water in the peipheral seas

Hudson Bay freezes solid every year - being low salinity and enclosed by high latitude Canada.
A significant amount of open water in Hudson Bay in the dead of winter would be a real game changer.

The Being and Barents front onto the open waters of the Pacific NE and NW Atlantic oceans, and have a huge variation of winter open water between years, though the trend is UP.

click to enlarge images

[kassy]

When will the Arctic Go Ice Free IN WINTER?

Not very soon.

The historical data misses inputs like:
When we look at older historical data it does not reflect changes such as atlantification or the heat build up from pacific waters. Old data also does not account for the more fractured state of the ice overall or the fact the ice pack is not clinging to land like it used to or it is not compacted into coast like it used to.

Then again my first sign of BOE would be a certain Gerontocrat pointing out that the Central Arctic Seas were really slow to get to their usual 100% in the refreeze stage.

Before that there would be a summer with lots of open water and mixing up heat from below, maybe some wind fetch could open an area which persists. This should probably happen along the atlantic/siberian side of the CAS.

I expect that in 10-15 years. That is only the first signs but it should accelerate quite quickly since it would interfere with the long term ice building cycle and also local changes compound so what once an ice desert might turn into something more temperate with a slight drizzle over the ice in early season.

[Glen Koehler]

     At risk of revealing myself for being a meathead in case anybody had any doubts, looking at the "Where Will the Last ASI be located" Poll thread, I can't understand how smart trained scientists can publish Extent projections that show >1M km2 out to 2080 and beyond. 

     When I look at a PIOMAS Volume trend graph and see a linear extrapolation hitting zero by 2030, it seems pretty clear that we will hit ASI < 1M km2 LONG before 2080 or even 2050.  This may be a case of knowing just enough to egregiously wrong, but it seems pretty clear to me that an Extent model that shows >1M km2 out to end of the century is counting the trees and missing the fact that the forest is on fire.  RCP4.5 would be great, but I think there is already too much momentum in the Earth energy balance to prevent near-term ASI collapse. 

     Sorry to be a bummer doomer dude, but facts is facts.  Maybe some good ole shrill alarmism can get us in gear to turn things around to recover the prior Earth energy balance and climate regime.  But we have already strapped into the roller coaster and are going for a ride.  Things will get crazier for the ASI long before they get better.

The urgency of Arctic change
James Overlanda, et al.
Polar Science
Volume 21, September 2019, Pages 6-13
https://doi.org/10.1016/j.polar.2018.11.008

Caption to Figure 4:

Fig. 4. September sea ice extent based on 82 ensemble members from 36 CMIP5 models under the RCP 4.5 scenario. Each thin colored line represents one member from a model. Up to five members per model are shown. The thick yellow line is the simple arithmetic mean of all ensemble members, and the blue line illustrates the median value. The thick black line represents observations based on the adjusted HadISST ice/sea ice analysis (https://www.metoffice.gov.uk/hadobs/hadisst/data/download.html). The horizontal black dashed line marks the 1.0 M km2 value, which indicates a nearly sea ice-free summer Arctic (Wang and Overland, 2009). The median suggests a sea-ice-free Arctic Ocean in late summer near the end of the century under RCP 4.5. However, observations and some models suggest that the Arctic Ocean could be seasonally ice-free significantly sooner (10–30 years; Overland and Wang, 2013). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

[Paul]

I think part of the reason why a BOE may not happen until around 2080 is because as we get less and less ice, its going to hit areas which has never melted out(e.g the North Pole) so the rate of decline may well slow down. Of course we could always get one freek year where we reach the 1 million mark but it probably won't be until around 2080 where its happening year in, year out.

Despite last year's record Siberian ice melt, warm weather and very warm SST's, we were still around 2.5 million in extent before reaching BOE status. Admittedly August last year was quite quiet weather wise and that may of saved extent reaching record lows but despite all the heat, it does remind you how much more energy is still needed to reach a BOE.

[Glen Koehler]

     Fair enough Paul.  But it took a lot of energy to reduce September ASI volume by 75% over the past 40 years, and as far as I can tell we are on the same train going down the same track for the next 25%.
     As for the North Pole holding out, you know what happened last year.  And as of May 11, 2021 prospects for the ice at the NP seem worse not better than they did in early May 2020.

[oren]

While I've no doubt BOE will arrive far earlier than 2080, and quite possibly by 2030, I must remind that a part of the PIOMAS trend was achieved by decimation of MYI and its replacement with FYI, and that FYI can be replaced nearly 1:1 every year thanks to winter freezing. So it is certainly possible volume decline should slow down, and the linear assumption prove false in real life. In essence, the "Slow Transition" theory originally by Chris Reynolds.

[Paul]

     Fair enough Paul.  But it took a lot of energy to reduce September ASI volume by 75% over the past 40 years, and as far as I can tell we are on the same train going down the same track for the next 25%.
     As for the North Pole holding out, you know what happened last year.  And as of May 11, 2021 prospects for the ice at the NP seem worse not better than they did in early May 2020.

I still slightly dispute the location of that photo as none of the satalites picked up ice looking like that at the pole. I suspect that was taken in the area between Greenland and the pole where the ice did become diffused which is extraordinary itself.

[oren]

Your suspicions are wrong - unfortunately. I recommend reading the whole article.

https://thebarentsobserver.com/en/2020/08/mosaic-climate-expedition-shares-scary-photos-north-pole

Expedition shares scary photos from the North Pole
Loose and weak ice with lots of melt ponds, partly open water, and no signs of multiyear ice. The powerful photos from the MOSAiC expedition reaching the North Pole on August 19 show the dramatic impact of climate changes.

The MOSAiC crew and scientists gathered on the deck as Polarstern reached the North Pole. Photo: Lianna Nixon / MOSAiC

[The Walrus]

Looking at your graph of the models for RCP4.5, most show a BOE later than 2080, and some much later.  Most notable on the graphs is the slow down is ice loss as time elapses.  This may occur as oren states.  Still, using the thick black line of sea ice analyses (Hadley), the trend is not likely to cross 1M before 2070.  Using NSIDC data, the trend crosses in 2057, although using just recent data (the last 14 years), it would take much longer.  Still, using the data from the past 40 years, and extrapolating out 40 more is precarious, especially if we cannot ascertain how the changes will affect the data in the future.

[kassy]

While I've no doubt BOE will arrive far earlier than 2080, and quite possibly by 2030, I must remind that a part of the PIOMAS trend was achieved by decimation of MYI and its replacement with FYI, and that FYI can be replaced nearly 1:1 every year thanks to winter freezing. So it is certainly possible volume decline should slow down, and the linear assumption prove false in real life. In essence, the "Slow Transition" theory originally by Chris Reynolds.

The decimation of MYI was a step change but of course there will be another one in the future. Yes the ice comes back every winter but actual winters shorten. More and more areas will refreeze later thus gathering more energy which takes more night to refreeze it and thus ending up thinner. There is the intruding Atl + Pac waters and also the fact that we have not actually turned down the emission pump yet so global temps will also add more pressure.

A lot of the statistics of the past have behaviour we do not see that much anymore like really thick ice forming by thick ice colliding with more thick ice.

Seeing how broken the ice was last year was interesting...
I think some time this decade.

Also some things might happen that are not in the recent historical record like a day of slight drizzle over a big area of ice or early open water combined with a big storm might bring up enough heat in the proper area to leave a gaping hole in the CAS ice which would break up or interfere with the ice flows moving towards Greenland or both in the same year.

There are many different inputs over the years and the forces change as the ice changes in relation to the oceans while the planet also changes (general heat budget + snow deficit in areas near sea ice etc).
So any projection made from the recent historical data will probably be long because it misses processes that become relevant at the end.

[gerontocrat]

While I've no doubt BOE will arrive far earlier than 2080, and quite possibly by 2030, I must remind that a part of the PIOMAS trend was achieved by decimation of MYI and its replacement with FYI, and that FYI can be replaced nearly 1:1 every year thanks to winter freezing. So it is certainly possible volume decline should slow down, and the linear assumption prove false in real life. In essence, the "Slow Transition" theory originally by Chris Reynolds.

The decimation of MYI was a step change but of course there will be another one in the future. Yes the ice comes back every winter but actual winters shorten. More and more areas will refreeze later thus gathering more energy which takes more night to refreeze it and thus ending up thinner.

Winter is coming later, as is shown by the monthly average graphs for October 2020 High Arctic sea ice extent and area - both at record lows.

November not so. It will be interesting to see when that delay in refreeze extends into November.

[Glen Koehler]

FWIW, global warming affects both the ocean/atmospheric and solar heat inputs directly. And again, I as I conjectured, it may affect the timing (changing the date at which snow, land, and melt ponds take their part in the process). The effect of the latter on SIE overall may be one of those non-linearities we've been looking for WRT CO2 forcing.

I've been thinking about these non-linearities a lot. But over the years I've been surprised how 'linear' the decline has been. Both for SIE as well as volume.
It may be related to the albedo-amplification factor (caused by snow and ice extent decline) during the melting season. If that factor is constant (as kind of could be expected), then the decline in sea ice volume will be linear. And if the ice thickness of the ice that melts out (mostly FYI) does not differ that much, then SIE will also decline linearly.

I still expect some non-linearity once ice volume gets really low, as best shown in this graph by Chris Reynolds :



Since there is no physical reason why volume would not continue to decline linearly, there will come a point that SIE will decline rapidly, and non-linearly.
As prof. Wadhams once stated : "In the end, it will just melt away quite suddenly. ".

     Whether to rely on Volume trend vs Extent trend seems to be one of those internal faith-based choices that defy logical discussion.  I go with Volume, and that necessarily includes the assumption that Extent will continue following its own path until it can't because the continued decline in the other dimension - Thickness - progresses so far that the legs are kicked out from underneath Extent.  In other words, as Volume approaches zero, Extent has to catch up to Volume, not vice versa.

     Rob Dekker who is a real-life climate scientist who occasionally posts on ASIF, and the great Peter Wadhams, and even the semi-legendary Chris Reynolds, author of the ASIF Slow Transition thread, apparently agree.  We will continue with Extent vs. Volume on different tracks until Thickness gets below some threshold at which the melt resistance and other physical characteristics of the ice begin to change rapidly as thinning continues.  My guess for that threshold is at about 0.5M.  Below that Thickness threshold, as Volume continues steadily grinding downward, Sept Extent will have a reckoning with the multiplicative effects of Thickness, vulnerable thin ice, and 3D math and decline rapidly just in time to catch up to Volume when they both hit zero for the first time (in human terms) ca. Sept 2030.

     (If this "Thinning induced collapse of Extent" idea is true, then we should be able to see it in the data for regional seas that did not melt out at the beginning of the modern records of Extent, Thickness, and Volume, but which do reach zero Volume now.  So it seems like a testable hypothesis.)   
 
     A few years after Sept becomes a regular melt-out month, August and then October will join the club.  July and November will take a while longer, but also remember that long before they reach an annual minimum of zero, they will be depleted remnants far below their original Volume/Extent/Area numbers.  And by 2046-2047 they too could be reaching zero.
 
     Stroeve&Notze and Notz&Stroeve, both 2018, estimated the 50% chance of first 1M km2 Extent Sept BOE arriving around 2033-2035, and regular every-year Sept. BOEs by 2035-2038. 

     Stephan's linear regressions (if extended forward, which is statistically dubious, I take responsibility for applying doing this so don't blame him) show the following ZERO volume arrival dates
Sept 2032
Aug  2035
Oct   2035
July  2046
Nov  2047
Dec  2061
June 2067

     We may bend the trends from about 2050 on, but we have likely already committed the system to those trends continuing for the next 30-years or so, i.e. the climate system has about 30 years of inertia (and ocean heat a LOT longer than that) before changes we make today take full effect.  So between now and 2050 the path may be largely set. 

     If all this holds up to reality, then in 25-26 short years from now in 2046-2047, July-Aug-Sept-Oct-Nov (5 months per year) will have ZERO ASI, and another 4-weeks of late-June and early-Dec, will be nearing zero.

     The Arctic functions as a major component of the planetary climate system.  So in 25 years the Earth's climate system will not just be warmer by this or that amount, it will have a functionally different operational foundation.  I'm sure the climate modelers think about all this, but then again I'm not sure it is possible to foresee the implications of radical change on complex interactive multi-component systems. 

     Richard Alley made that point in a recent Jeff Goodell Rolling Stone article about West Antarctic ice sheet dynamics:
"No human has ever witnessed the rapid collapse of a glacier in Antarctica like Thwaites; ergo, it can’t happen. Alley himself thinks about it simply in terms of risk. “Maybe we’ll get lucky and the ice cliffs won’t disintegrate in Antarctica quite as fast as we predict,” he says. “But if you are even a little bit worried that scientists might have made mistakes in their calculations about what is going on in Antarctica, then maybe we should pay attention to this.” He compares Thwaites and other glaciers in West Antarctica with drunk drivers. “They are out there, they are scary, and they don’t behave as you expect them to,” Alley says. “That’s why it’s a good idea to have a seatbelt in your cars.” "
https://www.rollingstone.com/politics/politics-features/antarctica-thwaites-glacier-how-fast-will-it-melt-1168437/

     What used to be the realm of science fiction is now squarely within the confidence intervals of science fact.  Not without uncertainty, but these are not wild hypotheses, just simple extensions of well-established trends.  And those trends do not look good for the future of ASI.  As for human civilization based upon the climate as it was, well human ingenuity and adaptability is indeed impressive.  I wish it could be focused on advancing forward with less effort needed to correct for the stupid malfeasance of the previous (our) generation.  Our parents' generation was not aware of the climate sensitivity, and we can't blame the next generation for the mess we are leaving them.  So it falls to us as the responsible generation.
 
     Maybe that can be our legacy "The Responsible Generation", the ones who largely caused the problem but also the ones who recognized the facts and faced up to them.
 
     Even if we don't respond effectively and in time, I don't think humans will go extinct, though a lot of other species will.  But I do think that a very large number of people are going to directly suffer from the effects of climate transition, and all will be affected one way or another.  And surviving on a radically altered planet is not what I would call total success.  I just hope we are smart enough to leave Earth the way it was when we arrived.   

[Paul]

Your suspicions are wrong - unfortunately. I recommend reading the whole article.

https://thebarentsobserver.com/en/2020/08/mosaic-climate-expedition-shares-scary-photos-north-pole

Expedition shares scary photos from the North Pole
Loose and weak ice with lots of melt ponds, partly open water, and no signs of multiyear ice. The powerful photos from the MOSAiC expedition reaching the North Pole on August 19 show the dramatic impact of climate changes.

The MOSAiC crew and scientists gathered on the deck as Polarstern reached the North Pole. Photo: Lianna Nixon / MOSAiC

I just seemed to recall an area of diffused ice between Greenland and the pole but the pole itself(at least looking at worldview and the bremen charts) was more compact but of course looking quite thin. So was this actual open water between the ice or melt ponds on top of the ice?

[josh-j]

In any event, the pole is not going to be the last place to melt out. I'm sure there has been open water there in summer before, depending on all the intricacies of the melt season, ice movement etc.

The ice is safer near the cold pole of Greenland and north of Canada; at least, someone will correct me if I'm wring about that!

[The Walrus]

In any event, the pole is not going to be the last place to melt out. I'm sure there has been open water there in summer before, depending on all the intricacies of the melt season, ice movement etc.

The ice is safer near the cold pole of Greenland and north of Canada; at least, someone will correct me if I'm wring about that!

No need for correction.  That ice is better protected and thicker than that in the central arctic.

[oren]

I just seemed to recall an area of diffused ice between Greenland and the pole but the pole itself(at least looking at worldview and the bremen charts) was more compact but of course looking quite thin. So was this actual open water between the ice or melt ponds on top of the ice?

As can be seen in the image taken from above, the deep dark blue is open water between the floes, but the lighter blue is extensive melt ponds on top of the ice. So most of it isn't actually open water. But the state of the ice is still shocking for that location.

And indeed there was more ice above 88N during the Polarstern's voyage.

The expedition ship Polarstern sailed from the northern Fram Strait between Greenland and Svalbard to the North Pole this week.

“I’m very surprised to see how soft and easy to traverse the ice up to 88° North is this year, having thawed to the point of being thin and porous,” said Captain Thomas Wunderlich.

“Even after passing 88° North we mostly maintained a speed of 5-7 knots; I’ve never seen that so far north,” the Polarstern captain said.

He added: “The current situation is historic.”

[Stephan]

It is time for the monthly update of my extrapolation when the extent [Ausdehnung], volume [Volumen], thickness [Dicke] and area [Fläche] 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 and area in the winter months). The April value now includes 2021.

Position of April 2021 towards the long-term trend lines
Extent lies slightly above the long term linear trend, volume and thickness are well above the trend line. Area is below the long-term trend line.

Trend of the trends
The "BOE numbers" increased by 2 years (volume), by 3 years (thickness) and by 5 years (extent) compared to April 2020. Due to the lower-than average area its "BOE number" dropped by 18 years, but this values (years 2659/2770) are far away from reliability.
So there is a slight divergence between the "late value" (extent) and the "early values" (volume, thickness) which seems to be usual as I already posted for the "high winter months".
The order (earlier → later BOE) generally is volume < thickness < area < extent.

Please note that this is not a forecast but a trend!
See attached table, now widened to see the linear function value (y-AA) at t = 0. Stg = slope.

Click to enlarge it.

[BeeKnees]

A question, possibly one that should be on the beginners thread.

How relevant is the definition of BOE?
To my ignorant eye and based on the state of the the ice this year, is extent still a meaningful measure?  It appears that what was previously a nice big patch of solid multi year ice is now just rubble, so are we getting to a point where we never get a BOE simply because there is lots of dispersed ice that's barely surviving to to the end of the melt season in the high Arctic?

[The Walrus]

A question, possibly one that should be on the beginners thread.

How relevant is the definition of BOE?
To my ignorant eye and based on the state of the the ice this year, is extent still a meaningful measure?  It appears that what was previously a nice big patch of solid multi year ice is now just rubble, so are we getting to a point where we never get a BOE simply because there is lots of dispersed ice that's barely surviving to to the end of the melt season in the high Arctic?

Good question.  Extent has always been the most consistent sea ice measurement, so many scientists continue to use that definition.  Area tends to fluctuate more than extent, and volume is a calculation.  With regards to a BOE, by the time that the Arctic sea ice drops below 1M sq. km, most of the central Arctic will be completely ice-free.  Most of the remaining ice will be located near Greenland and in and around the CAA.  For all intents and purposes, when the sea ice extents falls below 1M, it will be a BOE.

[kassy]

NASA has these nice long term ice animations where you can see the thick ice in the pack over the years. The remnants of the thick ice looked like a skeleton of sorts in the later years. In the animations you see thick ice going via Beaufort back into the pack until recently.

Depending on where the ice floats 1 to 1,5M of melting can very well occur.

Ice freezes from the edges of land and thus the remaining ice is an imperfect predictor for next season. The whole Atl/Sib section does not look too well.

In general for Volume the question is where is it located in the basin. Area will hang around that and Extent will fill it up later.

The relevant part in the BOE definition is that it opens up so much water that all earlier patterns are broken. This does not mean that the BOE number is the edge, it is just a number we use to make sense of the world.

A lot of people thought they could just extrapolate the past into the future which means that a BOE would only happen towards the end of the season which implies that all future seasons would be the same as the past where the ice was more or less solid. The 2010s had many a year that started quick but stalled mid year but that might be because then there was still a sort of solid ice pack in the CAS.

Will it be the same if the ice is fractured and there is open water all around?

Probably not because this is also the Arctic switching from an arctic desert to a regular sea. Also there is a whole lot of buried heat below which might come into play with enough open oceans.

Short version: the rubble of dispersed ice won´t stay. What was rubble once was ice and will be sea.

[vox_mundi]

feel free to move this if it's in the wrong place...

Arctic Open-Water Periods Are Projected to Lengthen Dramatically by 2100
https://phys.org/news/2021-06-arctic-open-water-periods-lengthen.html

... Research results show that for every 1 degrees Celsius of global warming, the open-water period (on average) increases by about one month in the Arctic (faster change in the Arctic Ocean, slower change in the sub-Arctic seas).

"We've already seen the world warm about one degree since the 19th century—so this means the open-water period has already increased by about a month on average," Crawford said. "If we limit warming to two degrees, that means an additional one degree of warming over the next 80 years and an additional increase of one month (on average) for the open-water period. But the big implication here is that humans have a lot of control over how much the open-water period changes. There's no tipping point. So any reduction in emissions means shorter open-water periods."

Another key result of this paper is the finding that the entire Transpolar Sea Route, which is the most efficient path for a container ship to take from Japan to Europe and involves going through Bering Strait and across the North Pole, is consistently open for at least 90 days (3 months) with 3.5 degrees Celsius of warming and for at least six months with 5 degrees Celsius warming.

And the final key finding is that compared to satellite records of sea ice, nearly all climate models are either a good match for observed trends or underestimate how quickly sea ice has been changing.

"This means our results are more likely to underestimate sensitivity of the open-water period to global warming than they are to overestimate sensitivity," Crawford says.



Alex Crawford et al, Arctic open-water periods are projected to lengthen dramatically by 2100, Communications Earth & Environment (2021).
https://www.nature.com/articles/s43247-021-00183-x

[Stephan]

It is time for the monthly update of my extrapolation when the extent [Ausdehnung], volume [Volumen], thickness [Dicke] and area [Fläche] will reach zero. The extrapolation occured linearly and by a logarithmic function; the latter one almost constantly resulting in earlier times (valid all year round for volume and thickness, not for extent and area in the winter months). The May value now includes 2021.

Position of May 2021 towards the long-term trend lines
Thickness, area, volume and extent are above the linear trend line, volume is well above it.

Trend of the trends
All slopes have decreased.
The "BOE numbers" increased by 1 year (thickness), 2 years (volume), 18 years (area) and by 11 years (extent) compared to May 2020.
So there is once again a divergence between the "late values" (area, extent) and the "early values" (volume, thickness) which seems to be usual as I already posted for January through May.
The order (earlier → later BOE) generally is volume < thickness < area < extent.

Please note that this is not a forecast but a trend!
See attached table, now widened to see the linear function value (y-AA) at t = 0. Stg = slope.

Click to enlarge it.

[kassy]

September Extent will keep heading slowly down, but won't change radically until one year it does change with shocking speed.  The slow decline of Extent and Area are hiding the inexorable weakening from Thickness decline, fracturing, and mobility.  When the ice for a region reaches some unknown threshold, then Thickness loss accelerates and the ice becomes so weak that it becomes susceptible to the next flash melt event that comes along. 

     I look at Gero's Sept Area - Thickness - Volume graph with a new perspective now.
The crumbling of the southeastern CAB into a rubble field is not fully reflected in the Extent and Area measurements, but we can see that part of the CAB getting weaker as it transitions from a solid ice-pack fortress into a flotilla of vulnerable floes.  In a similar manner, I think that Thickness, the measurement we talk least about because we have the least frequent and accurate monitoring, is eroding the ice from the top due to melt pond albedo reduction, and eating away at the bottom of the ice from Atlantification and open water albedo reduction.  That vertical melting acts like unseen termites degrading the timbers of a house.  The damage is hidden until one day the structure gets so weak that it becomes inescapably obvious when the house collapses due to a big wind or heavy snow load on the roof.  The collapse appears to come out of nowhere, but really just reveals the weakness that was eroding the strength of the timbers all along.

     Gero's graph shows that Area (and even more so Extent which is not on the graph) will cruise along their gentle trajectories until around ca. 2030 when Thickness finally gets so weak that the whole structure of what used to be the September Arctic ice pack goes poof, and we have the first BOE.  Within only 2-3 years after that, the same Poof event starts happening irregularly, then every year, for August and October.  Followed by BOE status spreading into July and November in the mid-2040s, if not earlier due to cascade effects, just as Stephan's regressions are plainly telling us right now.  None of this should be surprising, but it will be surprising when it happens anyway.     

The backbone is volume and thickness. If you look at long term animations you see that the thick ice that used to go via the Beaufort into the Siberian side and back to Greenland stopped doing that or is doing that a lot less. This also means the central ice will be more mobile and there is no reason to think it is in better shape then last year.

All we need is an area to open up relatively early  in the season. This will change the depth of warmer bottom water up plus there will be all kind of weather systems mixing water.

This should happen around the Atlantic/Siberian border of the central seas and should at some point show up in the Central Arctic Seas Area data as a hole which did not use to be there. 

It is not going to wait for extrapolations from past worlds. All the ruble ice melts from the sides too which did not happen when all the ice was nicely together. That happened maybe at the end of the season years ago but it is already happening now and it is only early July June.

I don´t know when but probably this decade. And because of the heat already down below the deciding factor will not be where the ice is, but where it is not for long enough.

[I’M IN LOVE WITH A RAGER]

All the ruble ice melts from the sides too which did not happen when all the ice was nicely together. That happened maybe at the end of the season years ago but it is already happening now and it is only early July.

This is one of the most concerning aspects to me, and seems like it could easily get out of control quickly as the overall structure of the ASI as a whole degrades further. Yet another accelerationary positive feedback cycle?

[Sepp]

All the ruble ice melts from the sides too which did not happen when all the ice was nicely together. That happened maybe at the end of the season years ago but it is already happening now and it is only early July.

Is this actually correct? I understand, that open water increases sun absorbtion, but this is already the case for thin ice and melt ponds.

If the same amount of ice is there, the same energy is needed, no matter, if it is a compact cube or a thin flat plate.

So I think the ruble ice is rather an result (and as such maybe a state of reduced volume) than a reason of ongoing melt.

[interstitial]

Sepp
if the ocean were at equilibrium the shape would not matter but the ocean is a very dynamic place. We know that enough heat to melt the ice exists below the surface if it mixed better. In the ocean ice freezing and melting is all about heat gradients and energy traveling through ice. once the ocean has cooled sufficiently ice forms rather quickly but the thicker it gets the slower new ice forms. Unless of course pressure ridges form breaking up the surface.
Allowing energy to flow around the ice dramatically increases energy transfer. I am not aware quantitively how much faster but it most certainly makes a difference.

[Sepp]

Ah, ok. I did not expect that big change, but your explanation makes sense. Thanks

[kassy]

All the ruble ice melts from the sides too which did not happen when all the ice was nicely together. That happened maybe at the end of the season years ago but it is already happening now and it is only early July. June

Is this actually correct? I understand, that open water increases sun absorption, but this is already the case for thin ice and melt ponds.

If the same amount of ice is there, the same energy is needed, no matter, if it is a compact cube or a thin flat plate.

So I think the ruble ice is rather an result (and as such maybe a state of reduced volume) than a reason of ongoing melt.

The most incorrect part was the month (fixed it in the quote).

If you look at the long term then yes it is a result (2007 loss of big ice, 2012 SIE dive then a number of years which do not equal that but we end up with the 2020 rubble).

If the same amount of ice is there, the same energy is needed, no matter, if it is a compact cube or a thin flat plate.

There are energy inputs beyond solar energy.

But if we first look at that you can already see a difference. If we idealize the ice to thin flat plate that will have meltponding and then drainage all over. Now if we put that ice in a compact cube it will have less meltponding in total but that also means that all the water which is not covered by ice is getting directly warmed. Areas without ice also lose the cooling profile ice provides so that would also lead to mixing up heat from below.

Rubble ice is something in between those two. Of course the simple fact that it has exposed sides will lead to a bit more melting. This should be worse at the ice pack edges were warmer waters come in from the Atlantic.

And of course this ice will be more vulnerable to a future GAC (if the same heat as in 2012 gets brought up it will melt more rubble ice) and then there is the timing. If it happens early enough in the season there will enough time to warm the surface waters and also mix up enough heat from below to leave a hole well into the refreezing season which should be noticeable in the CAS data as reported by Gerontocrat.

This hole need not last whole winter season but just a couple of months delay should set up a  following year which will see a repeat. Also large open swaths of water might change the weather dynamic a bit in the early refreeze season where wind fetch might mix up more buried heat and it will also move the ice around which is easier with rubble ice too.

[Reginald]

Irreversible warming tipping point may have been triggered: Arctic mission chief

AFP, June 15, 2021

"The disappearance of summer sea ice in the Arctic is one of the first landmines in this minefield, one of the tipping points that we set off first when we push warming too far," said Markus Rex.

"And one can essentially ask if we haven't already stepped on this mine and already set off the beginning of the explosion."

Rex led the world's biggest mission to the North Pole, an expedition involving 300 scientists from 20 countries.

[...]

Presenting their first findings, Rex said scientists found that the Arctic sea ice had retreated "faster in the spring of 2020 than since the beginning of records" and that "the spread of the sea ice in the summer was only half as large as decades ago".

"Only the evaluation in the next years will allow us to determine if we can still save the year-round Arctic sea ice through forceful climate protection or whether we have already passed this important tipping point in the climate system," he added.

https://today.rtl.lu/news/science-and-environment/a/1739166.html

[The Walrus]

The concept of an Arctic sea ice tipping point has been debated significantly.  (Not so) Recent studies seem to indicate that there is no such tipping point.

https://escholarship.org/uc/item/67q0r9m0  (2015)

https://authors.library.caltech.edu/25273/1/Armour2011p15718Geophys_Res_Lett.pdf   (2011)

<Corrected. O>

[oren]

The head of the Mosaic expedition visited the Arctic in 2020 for several months, these are models from 5 and 10 years ago. Not comparable, not the same level of reliability.

[The Walrus]

The head of the Mosaic expedition visited the Arctic in 2020 for several months, these are models from 5 and 10 years ago. Not comparable, not the same level of reliability.

True.  The peer-reviewed research has much more reliability.

[oren]

We shall agree to strongly disagree, in this case.

[kinbote]

The head of the Mosaic expedition visited the Arctic in 2020 for several months, these are models from 5 and 10 years ago. Not comparable, not the same level of reliability.

True.  The peer-reviewed research has much more reliability.

Perhaps something recent like 2020?

The tipping times in an Arctic sea ice system under influence of extreme events (2020)
https://europepmc.org/article/med/32611094

Full text:
https://arxiv.org/pdf/2003.02407.pdf

[kassy]

Much too theoretical, from our most recent visit to the pole:

These are the 3 most terrifying results of the Mosaic expedition

1 The ice is melting faster than ever before
During the Mosaic expedition, the ice retreated faster in the spring than it has ever been since records began. The extent of the ice was only half as large in summer and the thickness of the ice had also decreased by more than half compared to the data from almost 130 years ago when polar researcher Fridtjof Nansen researched the ice.

"During the winter of the Mosaic expedition, we measured temperatures almost continuously ten degrees higher than Fridtjof Nansen," said Rex. The question arises whether the ice can still be saved at all. If you saw the ailing and completely melted ice at the North Pole in summer 2020, doubts would come over you, says Rex.

2 Unprecedented heat threatens
The researchers also observed an unusual wind circulation pattern with an arctic west wind jet that was stronger than ever since records began in 1950. The wind pattern and thin ice result in rapid ice drift. The researchers drifted through the Arctic faster than they expected. According to Arndt, the researchers only needed 300 days for the entire drift, while Nansen needed three years at the time. "For me, this dynamic is emblematic of the change in the Arctic," says Arndt. In addition, the ice-free time has become longer in summer.

According to Rex, this band of wind around the Arctic is decisive for our weather. Extreme weather conditions like the heat wave last summer are exacerbated by the wind band and the receding ice, according to Rex.

3 The ozone layer is disappearing

Very much shortened excerpt. Those numbers 1 and 2 are going to repeat for the near future.

Last years polar picture was pretty bad if you compare it to only a few years before:
https://forum.arctic-sea-ice.net/index.php/topic,3280.0.html

You can see area and extent in lots of pictures but there is is a qualitative difference in the ice that is pretty frightening.

This is a much easier way to see a tipping point.

If 2010 to 2020 took us to that how do you expect that to survive another 10 years?