The 2020/2021 freezing season

[Paul]

Really would not be surprised if we have seen the ice maximum as the Barants sea is forecast southerly winds and the ice there is not all that thick. Of course the sea of Okhosk could easily refreeze again if the wind direction is just right but could be a close run thing.

Whilst the Atlantic side is going to be on the warm side, the Beaufort sea could run at quite a bit below average with a large PV parking itself over the region, early indications would suggest we should not see the early retreat as we did in 2019 although a Beaufort high could easily change that.

[uniquorn]

CS2SMOS merged and PIOMAS modeled ice thickness, jan31-feb15/16
Broadly in agreement though cs2smos doesn't show any 4m ice. We'll have to wait and see which is right about the Beaufort.
colours were re indexed to prevent dithering

With Beaufort coming into daylight here is a contrasted aqua modis image from feb18 to show structure.  https://go.nasa.gov/3k5Uuuf

[Sublime_Rime]

CS2SMOS merged and PIOMAS modeled ice thickness, jan31-feb15/16
Broadly in agreement though cs2smos doesn't show any 4m ice. We'll have to wait and see which is right about the Beaufort.
colours were re indexed to prevent dithering

Wow, that CS2SMOS thickness plot shows the thickest CAB ice even more predisposed to being flushed through the Fram outlet over the spring. Does anyone else find this particularly concerning? I'm seeing a very unique pattern of the thickest ice this year, very interested to see the volume anomaly maps from the latest PIOMAS release.

[Freegrass]

Latest Five Day Forecast
Wind + Temp @ Surface
Large GiF!

It looks like we could see some counterclockwise motion of the ice along the Siberian coast...

[Glen Koehler]

CS2SMOS merged and PIOMAS modeled ice thickness, jan31-feb15/16
Broadly in agreement though cs2smos doesn't show any 4m ice. We'll have to wait and see which is right about the Beaufort.
colours were re indexed to prevent dithering

      HYCOM indicates a substantially different distribution of ice thickness, much less around the North Pole (especially to the east of the NP), and much thicker ice to the NW from Banks Island. 
      HYCOM and CS2SMOS show a large incursion of open water between Svalbard and Novaya Zemlya/Severny Island that is covered by thin ice in PIOMAS. 
      Which model is the more accurate representation of thickness?



 

[uniquorn]

      HYCOM ice thickness paints a completely different picture.  I know this has been discussed.  But which one is the more accurate representation of thickness?

Hold on. I'll get my augur and nip over and check.
Personally I prefer something that is measured rather than a model. Probably best to take Neven's advice and compare, compare, compare.
Linking to that gif might be a bit heavy on volume for some

CS2SMOS merged ice thickness isn't a model. It's an attempt to get the best result by merging two satellite measurements.
CS2SMOS product description

Rationale
The European Space Agency’s (ESA) Earth Explorer SMOS satellite can detect thin sea ice, whereas  its  companion  CryoSat-2,  designed  to  observe  thicker  perennial  sea  ice,  lacks sensitivity. Using these satellite missions together completes the picture of the changing Arctic sea ice and provides a more accurate and comprehensive view on the actual state of Arctic sea-ice thickness

well, 'completes' is a bit ambitious, 'more accurate' is better, especially as SMOS is daily and CS2 takes a month to cover the whole basin...and..it's only available from october to april..

[Glen Koehler]

     

<snip> Wow, that CS2SMOS thickness plot shows the thickest CAB ice even more predisposed to being flushed through the Fram outlet over the spring. Does anyone else find this particularly concerning?

<snip> Hold on. I'll get my augur and nip over and check.
Personally I prefer something that is measured rather than a model.

CS2SMOS merged ice thickness isn't a model. It's an attempt to get the best result by merging two satellite measurements.
CS2SMOS product description

     Thanks Uniquorn.  Taking CS2SMOS as the most reliable view highlights Sublime_Rime's concern about the core of the thickest ice being located within a high risk zone for Fram export.  If that ice is broken up and thin enough to be mobile and is still in that neighborhood late in the melt season, it looks like a strong and persistent pressure system (low or high depending on center of wind rotation) could move the heart of the CAB ice south through the Fram. 

     I suppose the locational issue isn't new because the center of the thickest ice has typically been in the triangle between Ellesmere - NP - Greenland.  What may be new is the condition of that ice as seen in the MOSAIC observations of thin discontinuous ice on the way to the NP and the photos taken when they got there last August. 

     Of course, if the ice is so weakened as to allow such an event to happen, that also means that the volume exposed to Fram export would also be less.  Even so, if the core the CAB ice gets beat down to the point of being a mobile collection of fractured chunks instead of a solid ice pack, and major Fram export ensued, that would eviscerate what used to be the area of thickest ice.  That's the kind of wild card event that would accelerate progression towards a first BOE.  It would be another example of how our implicit assumption for incremental change is overly simplistic compared to the real world of dynamic variability from interacting and sometimes synergistic components. 

     All of which is conjecture suggesting that the state of the ASI as we near the end of the 2020-21 freezing season may be worse than the Extent and Volume numbers indicate.  (I had to add the last sentence to keep this post suitable for the freezing thread! ).   
   

[interstitial]

Satellite observations of thickness have severe deficiencies work is being done to improve their performance. At this point they are not very reliable at some observations.
Hycom is most interested in providing an accurate picture of current conditions and a short term forecast used for operating in the Arctic. The starting conditions are updated by any and all observations available. This includes the aforementioned satellites. As satellite thickness observations improve so will the daily starting point for Hycom model. The condition of ice at the north pole surprised researchers on the polar star but were consistent with the Hycom model.

[Stephan]

I evaluated the JAXA data of all freezing seasons from Feb 19 to the individual maximum to find out how often an increase of 0.23 M km² has happened after Feb 19. This is the value needed to be higher than the actual maximum from Feb 16 (13.98 M km²). I grouped these evaluations into the different decades. The evaluation has three outcomes:
A - an increase higher than 0.24 M km² has happened in x years (Pass), which means the maximum has not been reached
B - an increase of 0.23 ± 0.01 M km² has happened in y years (Intermediate), which means the maximum may have been reached, but maybe not
C - an increase of less than 0.22 M km² has happened in z years (Fail), which means the maximum has already been reached and we can start the melting season thread.

                     A       B      C
1979-1989:    5       2      4
1990-1999:    1       0      9
2000-2009:    5       1      4
2010-2020:    8       0      3
sum:             19      3     20

Summary: If we assume that the current situation of the ice and its distribution is more like the 2010-2020 years than e.g. the 1990s then it is more likely than not that the maximum of 2021 has not been reached yet.

[oren]

Thank you for the analysis Stephan.
As the extent maximum is determined by the peripheral seas, mainly the "four Bs" Baffin, Barents, Bering and (B)Okhotsk which have high variability, it is not surprising that years with low extent have a higher probability to grow further at this time.

Regarding the ice thickness distribution, it is a great mystery that can never be resolved without detailed local measurements. I find it rather surprising and gratifying that CS2SMOS and PIOMAS are in such agreement as to the location of the main bulk, though PIOMAS adds some significant hotspot on the Siberian coast and in the Beaufort. Hycom is indeed quite different, my current tendency is to take CS2SMOS as the more probably correct representation of reality, but this is just layman intuition. And indeed the increased mobility of the ice plus its purported location in a region quite easily exportable is a cause for grave concern. Hopefully the Arctic will avoid the export regime of last year around this time, or a large region of thick ice will be lost before anyone can find out if it's really there or not.
Maybe the ensuing melting season will reveal where the ice was thicker or thinner than claimed by the various sources, should a certain region disappear in a flash or persist much longer than it should. I vividly recall Big Block persisting in the Beaufort in 2016 for months only to disappear on the last days of the season, proving beyond a doubt its high original thickness and age.

[Jim Hunt]

It is more likely than not that the maximum of 2021 has not been reached yet.

I agree, but there has been no "rebound" just yet:

[VeliAlbertKallio]

Beaufort sea ice is quite fractured and weak state. https://worldview.earthdata.nasa.gov/?v=-2359552.4009058364,430071.64129245165,-1624832.4009058364,762359.6412924516&p=arctic&t=2021-02-17-T12%3A24%3A34Z

Often white ice is second or multiyear ice, here it appears that at least some of it is recent ice that has been heavily loaded by snow, but as the splits go right through the white ice fragments equally and the darker recent ice (not taking advantage of pre-existing darker leads), it is telltale sign that the whiter ice is not much thicker or stronger than darker ice but is snow. I take it that HYCOM overestimates ice and takes recent snowfall on ice as sea ice that simply isn't there.

Also PIOMAS also shows thick ice on the same area whereas CS2SMOS not.

I re-attach the image and would like to highlight by enlargement the areas where leads cut through white "old" ice and grey "new" ice equally revealing an illusion caused by fractured sea ice cover releasing moisture onto surrounding ice floes that remain.

[Jim Hunt]

My current tendency is to take CS2SMOS as the more probably correct representation of reality, but this is just layman intuition.

If you prefer CS2/SMOS to PIOMAS then it reveals a somewhat more pessimistic view of current Arctic sea ice volume:

[oren]

True, but a part of PIOMAS volume is modeled pressure ridges, which may not be detectable by remote sensing once their top has undergone some smoothing. OTOH PIOMAS may be overestimating these ridges or their resilience under increased mobility. All in all, I simply don't know.

[uniquorn]

Satellite observations of thickness have severe deficiencies work is being done to improve their performance. At this point they are not very reliable at some observations.
Hycom is most interested in providing an accurate picture of current conditions and a short term forecast used for operating in the Arctic. The starting conditions are updated by any and all observations available. This includes the aforementioned satellites. As satellite thickness observations improve so will the daily starting point for Hycom model. The condition of ice at the north pole surprised researchers on the polar star but were consistent with the Hycom model.

CS2 and PIOMAS were chosen because their data is available to us for further analysis. There are many models that could be compared on this or perhaps another thread, Hycom, DMI and CMEMs among them. Regarding the current condition of pole to fram ice, mosaic floe2 is in that area with 4 thermistor buoys, an ocean CTD buoy and 5 snow buoys that no one is looking at. Local measurements that might help resolve the discussion.

Some discussion of the thermistor buoy data here

This image from below the ice perhaps highlights the difficulties encountered when measuring and modelling sea ice thickness.

[gerontocrat]

I've been posting the images attached on the "smart and stupid questions" thread. But now there does seem to be an event unfolding way up high over the Arctic.

There is loads of stuff on SSW, but not a lot on when the stratosphere cools. This link** from NASA  basically says "we don't know". The only consequence stated with confidence is that a cold stratosphere encourages the formation of stratospheric clouds.
** https://www.giss.nasa.gov/research/features/200402_tango/

Elsewhere I have seen that a colder stratosphere tends to favour a stronger polar vortex.

But the fact remains that temps at 10hpa are about 20C below average, and 10C below average at 30hpa. This is at a time when rapid temp increases is the norm. Has anybody any thoughts on the effect of a "sudden stratospheric cooling" on lower altitudes?

Images from http://ds.data.jma.go.jp/tcc/tcc/products/clisys/STRAT/

[Freegrass]

Satellite observations of thickness have severe deficiencies work is being done to improve their performance. At this point they are not very reliable at some observations.
Hycom is most interested in providing an accurate picture of current conditions and a short term forecast used for operating in the Arctic. The starting conditions are updated by any and all observations available. This includes the aforementioned satellites. As satellite thickness observations improve so will the daily starting point for Hycom model. The condition of ice at the north pole surprised researchers on the polar star but were consistent with the Hycom model.

I think you are dead on with this. HYCOM uses multiple sources to create their model, including buoy and submarine data if I remember correctly. I would guess that if they have multiple sources of data, and the satellite shows something different than what the buoys and other data sources are telling them, that their model corrects itself. To rely on one measurement from space is risky IMHO.

And HYCOM is just an awesome visualization. Used by the military, right? So if it's good enough for them, it's good enough for me...

the "four Bs" Baffin, Barents, Bering and (B)Okhotsk

Sea-3BO Oren... 

[interstitial]

My interpretation of the mission sitting on the ice for a year in a ship the major take away thus far has been how much worse the Ice was near the pole than they were expecting. While they do not say this Piomass is the ice climate model that I believe led them to this faulty assumption. Shortly after the mission the Wegener institute started talking about creating a better ice volume map. Before that mission I was only aware of satellite measurements being used for Ice less than 50 cm. Now we have someone looking at thicker ice trying to improve the algorithm. A method to determine snow thickness was created as well as well as identifying characteristics of different types of ice. I do not know if those events are unrelated or tell a cohesive story but no disrespect to the climate scientists who created the Piomass model but it has some major shortcomings with current arctic conditions. The model was made before most of the multiyear ice melted and I believe the dynamic has changed.  An extended discussion of Piomas versus Hycom can be found on the Hycom thread. While a full consensus was not reached I think people left better informed on the topic. I know I did. 

[VeliAlbertKallio]

I'd like to know this expedition's start date making ice observations, if Russians were boarded as in early 2020 I learnt of a novel expectation of major changes in atmospheric system by sea ice losses forecasted within passing of 2 Arctic melt seasons, if this expedition somehow contributed to these conclusions implied. Trying to crack on genesis of these reports for end-of-year cc event. 

My interpretation of the mission sitting on the ice for a year in a ship the major take away thus far has been how much worse the Ice was near the pole than they were expecting. .. snip

[Stephan]

I evaluated the JAXA data of all freezing seasons from Feb 19 to the individual maximum to find out how often an increase of 0.23 M km² has happened after Feb 19. This is the value needed to be higher than the actual maximum from Feb 16 (13.98 M km²). I grouped these evaluations into the different decades. The evaluation has three outcomes:
A - an increase higher than 0.24 M km² has happened in x years (Pass), which means the maximum has not been reached
B - an increase of 0.23 ± 0.01 M km² has happened in y years (Intermediate), which means the maximum may have been reached, but maybe not
C - an increase of less than 0.22 M km² has happened in z years (Fail), which means the maximum has already been reached and we can start the melting season thread.

                     A       B      C
1979-1989:    5       2      4
1990-1999:    1       0      9
2000-2009:    5       1      4
2010-2020:    8       0      3
sum:             19      3     20

Summary: If we assume that the current situation of the ice and its distribution is more like the 2010-2020 years than e.g. the 1990s then it is more likely than not that the maximum of 2021 has not been reached yet.

Updated evaluation for Feb 20, 2021:
                     A       B      C
1979-1989:    3       0       8
1990-1999:    1       0       9
2000-2009:    4       0       6
2010-2020:    7       0       4
sum:             15      0     27

[El Cid]

Elsewhere I have seen that a colder stratosphere tends to favour a stronger polar vortex.

But the fact remains that temps at 10hpa are about 20C below average, and 10C below average at 30hpa. This is at a time when rapid temp increases is the norm. Has anybody any thoughts on the effect of a "sudden stratospheric cooling" on lower altitudes?

I know nothing about this but I guess the current cooling is due to the polar vortex finally "restarting" after having collapsed spectacularly in January (and it stayed collapsed for unusually long). So it is a sort of pendulum-effect: it was down and out for long so it returned in strength.
Now a strong vortex favours ice retention if I am correct BUT with the current extreme heat intrusion into the North Pole region that might not be true.
Anyway, the Pole-region seemed very vulnerable even before this event and now looks even more so.
So, as I said I know nothing.

[uniquorn]

amsr2-uhh atlantic side, feb15-10. Too cloudy for brightness temp on worldview. Ice north east of Greenland is certainly mobile.

[gerontocrat]

Elsewhere I have seen that a colder stratosphere tends to favour a stronger polar vortex.

But the fact remains that temps at 10hpa are about 20C below average, and 10C below average at 30hpa. This is at a time when rapid temp increases is the norm. Has anybody any thoughts on the effect of a "sudden stratospheric cooling" on lower altitudes?

I know nothing about this but I guess the current cooling is due to the polar vortex finally "restarting" after having collapsed spectacularly in January (and it stayed collapsed for unusually long). So it is a sort of pendulum-effect: it was down and out for long so it returned in strength.
Now a strong vortex favours ice retention if I am correct BUT with the current extreme heat intrusion into the North Pole region that might not be true.
Anyway, the Pole-region seemed very vulnerable even before this event and now looks even more so.
So, as I said I know nothing.

Meanwhile,
Temps keep dropping in the Stratosphere above the Arctic, and the surface temperatures North of 80 published by DMI have just zoomed up..

What causes what how where and when is to me a complete mystery.

as usual, click images to enlarge

[glennbuck]

Ouch, cold weather in march and april might save the bacon.

[gerontocrat]

Ouch, cold weather in march and april might save the bacon.

I don't believe that graph. Something went wrong?

[glennbuck]

I don't believe that graph. Something went wrong?

Ye looks like it, but strange Nico Sun posted it rather than wait for corrected data!

[be cause]

regarding the dmi80 graph .. it is tracking perfectly the (stratospheric) rise this time in 2018 that coincided  with the first great opening N. of Greenland . Deja vu ?
  Where has the worldview of the winter Arctic gone since I recommended it a couple of days ago ? Too scary for us mere mortals to see ? .. or have we lost a precious resource ?
   https://go.nasa.gov/2ONoaRl   b.c.

[gerontocrat]

I don't believe that graph. Something went wrong?

Ye looks like it, but strange Nico Sun posted it rather than wait for corrected data!

I believe the update is on autopilot as with a lot of his stuff.

[uniquorn]

regarding the dmi80 graph .. it is tracking perfectly the (stratospheric) rise this time in 2018 that coincided  with the first great opening N. of Greenland . Deja vu ?
  Where has the worldview of the winter Arctic gone since I recommended it a couple of days ago ? Too scary for us mere mortals to see ? .. or have we lost a precious resource ?
   https://go.nasa.gov/2ONoaRl   b.c.

NOAA-20 viirs brightness temperature has come back but it's too cloudy today. https://go.nasa.gov/2OZ85Iw

[be cause]

NOAA-20 viirs brightness temperature has come back but it's too cloudy today. ..
 
 'tis strange .. the missing 2 days have suddenly appeared too .. b.c.

[gerontocrat]

NOAA-20 viirs brightness temperature has come back but it's too cloudy today. ..
 
 'tis strange .. the missing 2 days have suddenly appeared too .. b.c.

Perhaps NASA uses for this operation a facility in Texas and couldn't afford the electricity bill? You never know.

[Jim Hunt]

Something went wrong?

It did indeed:

https://twitter.com/lavergnetho/status/1363424933850841090

Update # 4 (21/02@0805Z)! ESPC is still not receiving DMSP SSMIS data.

The outage apparently started "due to a scheduled maintenance at OFFUT" air base. Indeed, DMSP is military.

[gerontocrat]

Something went wrong?

It did indeed:

https://twitter.com/lavergnetho/status/1363424933850841090

Update # 4 (21/02@0805Z)! ESPC is still not receiving DMSP SSMIS data.

The outage apparently started "due to a scheduled maintenance at OFFUT" air base. Indeed, DMSP is military.

Time for the quote "Houston. we have a problem"

[be cause]

Uniquorn ''NOAA-20 viirs brightness temperature has come back but it's too cloudy today. ''https://go.nasa.gov/2OZ85Iw

 .. not if you zoom in ( as often with wv ) ; today there is open water 30km wide @ 83'N 21'W , and the last few days show what a weak state the region's ice is in .. b.c.

[Stephan]

I evaluated the JAXA data of all freezing seasons from Feb 19 to the individual maximum to find out how often an increase of 0.23 M km² has happened after Feb 19. This is the value needed to be higher than the actual maximum from Feb 16 (13.98 M km²). I grouped these evaluations into the different decades. The evaluation has three outcomes:
A - an increase higher than 0.24 M km² has happened in x years (Pass), which means the maximum has not been reached
B - an increase of 0.23 ± 0.01 M km² has happened in y years (Intermediate), which means the maximum may have been reached, but maybe not
C - an increase of less than 0.22 M km² has happened in z years (Fail), which means the maximum has already been reached and we can start the melting season thread.

                     A       B      C
1979-1989:    5       2      4
1990-1999:    1       0      9
2000-2009:    5       1      4
2010-2020:    8       0      3
sum:             19      3     20

Summary: If we assume that the current situation of the ice and its distribution is more like the 2010-2020 years than e.g. the 1990s then it is more likely than not that the maximum of 2021 has not been reached yet.

Update for Feb 21, 2021. New threshhold value is 0.22 M km². In-between evaluation is valid for 0.21-0.23 M km²

                     A       B       C
1979-1989:    3       0       8
1990-1999:    1       0       9
2000-2009:    4       0       6
2010-2020:    7       1       3
sum:            15       1     26

[BornFromTheVoid]

The goings on in the last 10 days to the north of Greenland and Svalbard
(click to play)

[dnem]

Which has contributed more to the overall loses, this region or the retreat in the Sea of Okhotsk?

[glennbuck]

Sea ice thinner across most of arctic in January!

[Paladiea]

But the fact remains that temps at 10hpa are about 20C below average, and 10C below average at 30hpa. This is at a time when rapid temp increases is the norm. Has anybody any thoughts on the effect of a "sudden stratospheric cooling" on lower altitudes?

If I recall correctly, stratospheric cooling is an indication of heat not making it that high, ie. there is more heat being retained by the troposphere.

There could be many explanations for this, from elevated levels of methane in the Arctic, to a clearer sky due to a lack of air travel due to the pandemic.

[The Walrus]

I evaluated the JAXA data of all freezing seasons from Feb 19 to the individual maximum to find out how often an increase of 0.23 M km² has happened after Feb 19. This is the value needed to be higher than the actual maximum from Feb 16 (13.98 M km²). I grouped these evaluations into the different decades. The evaluation has three outcomes:
A - an increase higher than 0.24 M km² has happened in x years (Pass), which means the maximum has not been reached
B - an increase of 0.23 ± 0.01 M km² has happened in y years (Intermediate), which means the maximum may have been reached, but maybe not
C - an increase of less than 0.22 M km² has happened in z years (Fail), which means the maximum has already been reached and we can start the melting season thread.

                     A       B      C
1979-1989:    5       2      4
1990-1999:    1       0      9
2000-2009:    5       1      4
2010-2020:    8       0      3
sum:             19      3     20

Summary: If we assume that the current situation of the ice and its distribution is more like the 2010-2020 years than e.g. the 1990s then it is more likely than not that the maximum of 2021 has not been reached yet.

Update for Feb 21, 2021. New threshhold value is 0.22 M km². In-between evaluation is valid for 0.21-0.23 M km²

                     A       B       C
1979-1989:    3       0       8
1990-1999:    1       0       9
2000-2009:    4       0       6
2010-2020:    7       1       3
sum:            15       1     26

Nice Stephan.  For NSIDC, the recent decrease brought 2021 in line with the average extent over the past decade (within 0.002 M km2).  The average increase to maximum in the NSIDC data is 0.347, so I suspect that the 2021 maximum has not yet occurred in their dataset.  Incidentally, sea ice extent in the NSIDC only dropped 0.11 M.

[Freegrass]

The goings on in the last 10 days to the north of Greenland and Svalbard
(click to play)

This is actually not a bad thing IMHO because the ice is being pushed - and probably stacked - deeper into the basin. The bigger problem would be if the wind was the other way, and exported all that thicker ice.

I also think that the heat from this system is a bigger problem than the wind and open water, because it prevents the ice to thicken.

I wonder if the pack will act like a spring. Will it rebound when the wind goes away?

[Freegrass]

Latest Five Day Forecast
Wind + Temp @ Surface
Large GiF!

The thickest ice is in the deep freeze, while the thinner ice is not...

[Rodius]

Silly question....
I havent seen an update in NSIDC extent numbers for a few days, is that just me or is it global?

[Jim Hunt]

Silly question....

It's not just you. Scroll up to:

https://forum.arctic-sea-ice.net/index.php/topic,3299.msg301361.html#msg301361

[Niall Dollard]

There has been some recent break up off the NE of Greenland.

Looking back at how extreme it has gone in the past - 3 years ago at this time of the year we had this happen along the north coast :

 http://seaice.de/Wacky_20180225.gif

The temperature at Cape Morris was over + 6 C !

In the GIF you can see plenty of movement/break up in the Lincoln Sea. That year the arch in the Nares Strait did not form until early March which likely made the north coast more susceptible to attacks from the south.

This year the Nares has formed early which has probably helped prevent this event from being as bad as Feb 2018.   

[Rodius]

Silly question....

It's not just you. Scroll up to:

https://forum.arctic-sea-ice.net/index.php/topic,3299.msg301361.html#msg301361

I knew I should have scrolled up.
Thanks

[glennbuck]

Lots of data seems to be going wonky this year!

[glennbuck]

There has been some recent break up off the NE of Greenland.

Looking back at how extreme it has gone in the past - 3 years ago at this time of the year we had this happen along the north coast :

 http://seaice.de/Wacky_20180225.gif

The temperature at Cape Morris was over + 6 C !

In the GIF you can see plenty of movement/break up in the Lincoln Sea. That year the arch in the Nares Strait did not form until early March which likely made the north coast more susceptible to attacks from the south.

This year the Nares has formed early which has probably helped prevent this event from being as bad as Feb 2018.

Ye crazy weather in Svalbard, Longyearbyen 23rd Feb, been like that most of February. March and April swings way below average might make up for it if predictions are correct.

High
Low

Forecast
-1°
-3°

Average
-13°
-18°

Last Yr
-19°
-21°

https://www.accuweather.com/en/sj/longyearbyen/310461/daily-weather-forecast/310461?day=1

[Stephan]

I evaluated the JAXA data of all freezing seasons from Feb 19 to the individual maximum to find out how often an increase of 0.23 M km² has happened after Feb 19. This is the value needed to be higher than the actual maximum from Feb 16 (13.98 M km²). I grouped these evaluations into the different decades. The evaluation has three outcomes:
A - an increase higher than 0.24 M km² has happened in x years (Pass), which means the maximum has not been reached
B - an increase of 0.23 ± 0.01 M km² has happened in y years (Intermediate), which means the maximum may have been reached, but maybe not
C - an increase of less than 0.22 M km² has happened in z years (Fail), which means the maximum has already been reached and we can start the melting season thread.

                     A       B      C
1979-1989:    5       2      4
1990-1999:    1       0      9
2000-2009:    5       1      4
2010-2020:    8       0      3
sum:             19      3     20

Summary: If we assume that the current situation of the ice and its distribution is more like the 2010-2020 years than e.g. the 1990s then it is more likely than not that the maximum of 2021 has not been reached yet.

Update for Feb. 22, 2020. Threshold back to 0.23 M km², "in between" 0.22-0.24 M km²

                     A       B       C
1979-1989:    2       1       8
1990-1999:    1       0       9
2000-2009:    3       0       7
2010-2020:    6       0       5
sum:            12       1     29

The probability that 2021's maximum has already passed, increases slightly.

[Jim Hunt]

Lots of data seems to be going wonky this year!

Actually it's all the same original data, from the SSMIS instrument on the DMSP F17 satellite.

Allegedly the problem is now fixed:

https://GreatWhiteCon.info/2021/02/the-2021-maximum-arctic-sea-ice-extent/#Feb-23

Hopefully the NSIDC’s Charctic graph will update later today. Meanwhile JAXA extent is flatlining: