The 2023 melting season

[Niall Dollard]

Just my opinion of course.

Thanks Paul. Always good to hear other people's thoughts/opinions. It's what forums should be all about.

[Niall Dollard]

That's a massive decrease in 4+year ice for one year

It depends though which of Renerpho's charts you are comparing.

Best comparison is comparing Chart A below with Chart B. Maybe not a whole lot of difference in red pixels.

Chart C should not be used to compare with B because it is after the minimum of 2022 had passed and suddenly there is a big increase in red (compared with chart A). All ice has been moved on another year in age ! The previously 3 year old ice in chart A becomes red.

And looking at the greens at the end of melt season 2023, there should be an increase in yellow compared with last year.

[NeilT]

Just to add to that the amount of 4y+ ice that was flushed into the Beaufort and destruction right at the very end.

I know we now show 4y+ ice.  I can't remember if there was once a graph for ice which has an older base.  However we are moving to a time where 3y+ but not 4y+ is going to become the norm for multi year ice.

This year so much 3+ ice was flushed that replacement 4+ ice should be minimal.

I'll be interested in the charts for next year.

[Renerpho]

It depends though which of Renerpho's charts you are comparing.

Best comparison is comparing Chart A below with Chart B. Maybe not a whole lot of difference in red pixels.

Chart C should not be used to compare with B because it is after the minimum of 2022 had passed and suddenly there is a big increase in red (compared with chart A). All ice has been moved on another year in age ! The previously 3 year old ice in chart A becomes red.

That's why I included chart A. 

[Niall Dollard]

That's why I included chart A. 

Yes. Thank you.

[Niall Dollard]

Oh dear.

Something seriously gone wrong with NOAA/ESRL/PSL thickness chart.

https://psl.noaa.gov/forecasts/seaice/

[Renerpho]

I know we now show 4y+ ice.  I can't remember if there was once a graph for ice which has an older base.  However we are moving to a time where 3y+ but not 4y+ is going to become the norm for multi year ice.

Those maps used to show 5y+ ice (see here for an example of the "old" scale). NSIDC switched to the new scale in April 2019, because there was almost no 5y+ ice left. Compare the image below, from near the end of the 2019 melting season.

[Renerpho]

Oh dear.

Something seriously gone wrong with NOAA/ESRL/PSL thickness chart.

https://psl.noaa.gov/forecasts/seaice/

I sometimes think they are overly optimistic with their numbers for that region, but this is overcompensating.

[be cause]

I think it was in 2016 that I suggested we needed a new season between melting and freezing . I suggested autumn . Freezing has struggled to begin for @ 14 days now and weather could lead to significant melting over the next 14 . Mild temps and bottom melt coupled with strong winds is not a recipe for rapid ice growth . At best i see the stall continuing , at worst , a large fall in numbers and a record late minimum .

[oren]

In any case, day to day posts should go in our new season thread, regardless if about actual melting or freezing. Use this thread for seasonal summaries and discussion thereof.

[grixm]

Oh dear.

Something seriously gone wrong with NOAA/ESRL/PSL thickness chart.

https://psl.noaa.gov/forecasts/seaice/

This is just because of a long period of data outage in that missing area.

[Jim Hunt]

Those maps used to show 5y+ ice (see here for an example of the "old" scale). NSIDC switched to the new scale in April 2019, because there was almost no 5y+ ice left.

I think they just changed the legend to be less confusing. "First year ice" used to be labelled "1", and is now "0-1". The red squares always were "ice over 4 years old", so "5+" could easily be misinterpreted!

[NeilT]

Those maps used to show 5y+ ice (see here for an example of the "old" scale). NSIDC switched to the new scale in April 2019, because there was almost no 5y+ ice left.

I think they just changed the legend to be less confusing. "First year ice" used to be labelled "1", and is now "0-1". The red squares always were "ice over 4 years old", so "5+" could easily be misinterpreted!

This is news top me.  I've never seen it before.

Age is therefore assigned on a yearly basis, with the age incremented by one year if the ice survives summer melt and stays within the Arctic domain. In the model, the ice age is binned from 1 to 10 years, though for analysis, we combine ice older than 5 years in a “5+” age category, as the relationship between thickness and age levels off for much older ice. In other words, there was little difference between the thickness of 6-, 7-, 8-, 9-, and 10-year-old ice.

https://www.mdpi.com/2072-4292/8/6/457

Although from what I've read over the more recent years, there is little ice over 4y anyway nowadays.

[nadir]

For me, the main story this year was the unusual mobility of the ice (one consequence of which was the record high Fram export). As a result, we have lost maybe a third of all multi-year ice this year just due to Fram export, and what's left is sitting in places where we haven't seen it in quite a long time (Laptev).

While the entire CAB looked like a crowd of football fans trying to reach the nearest exit, that ice in Laptev has been sitting there since summer 2022 without moving much at all. I don't know what to make of that, but it will be interesting to watch what happens next.

If I should predict what happens next year just from the maps below, I'd say the Atlantic side looks strong, while the Pacific side may be more vulnerable than usual. However, if that unusual mobility continues then that "strong" Atlantic side may just go down the drain. Surely the Fram export becomes more and more damaging the more the ice pack moves in that direction.

Click on GIF below to start animation.

I took the liberty of editing your excellent animation, playing with frame delays to highlight two major events that happened in 2022/2023.
The first was a major MYI export event between mid-Oct and mid-Jan, that sent down the Fram a huge patch of old MYI of the green, orange and red variety.
This export movement actually continued throughout the year, taking with it more green MYI.
The second was a MYI meltout event in the CAB/Beaufort throughout Aug (and may have lasted into Sep) destroying a large patch of red MYI, the oldest type.
Another thing which can be inferred from the animation is that Nares export took with it another chunk of red MYI that got saved from Fram export but then settled in the Lincoln Sea.
To make matters worse, Fram export continued in the last few frames, taking more old MYI with it. This has continued into Sep as can be seen in the recent ice drift pattern.
The final frame shows how heavily skewed the MYI distribution is at the end of the season, with most of the green and cyan MYI found between the Pole and Fram, a region that is normally exported during the winter.

All in all, I think this was a very bad year for the ice, much worse than apparent in the extent and even area numbers. Some of this may be shown in the mid-month PIOMAS numbers (though I suspect the model is neglecting Nares export and overstating the Lincoln Sea for years).
But in any case, should 2024 turn out to be a strong melt year, it has the potential to easily become a top spot contender.

Very nice animations and comments @all. The distribution of the ice pack makes it look very vulnerable, we’ll see how things play during winter and spring next year

[Jim Hunt]

This is news to me.  I've never seen it before.

By way of example:

[Niall Dollard]

Oh dear.

Something seriously gone wrong with NOAA/ESRL/PSL thickness chart.

https://psl.noaa.gov/forecasts/seaice/

This is just because of a long period of data outage in that missing area.

No.  I have never seen anything like that before.

What has caused the data outage ?

Surely it would be better say that page was "not available" than put out an erroneous chart like that.

[grixm]

Oh dear.

Something seriously gone wrong with NOAA/ESRL/PSL thickness chart.

https://psl.noaa.gov/forecasts/seaice/

This is just because of a long period of data outage in that missing area.

No.  I have never seen anything like that before.

What has caused the data outage ?

Surely it would be better say that page was "not available" than put out an erroneous chart like that.

You can clearly see it over the last few weeks. Here is the plot from September 2nd, for example. Notice how the missing area lines up exactly with the area of low thickness in your image, except that the outlines are more straight because the model hasn't had time to simulate ice drift/regrowth yet. You're telling me this is not a data error?

[Renerpho]

You're telling me this is not a data error?

Sure looks like bad data to me.

[Niall Dollard]

Grixm, of course it is some data error.

I am just surprised they have left it like that for so long.

And I have not seen that before in my years looking at those thickness chart.

[gerontocrat]

NSIDC have produced a review of the 2023 melting season

https://nsidc.org/arcticseaicenews/
The Northern Hemisphere’s summer 2023 in review

With the passing of the seasonal maximum in Arctic sea ice extent on March 6, the melt season started slowly, but daily extents remained among the third to sixth lowest in the satellite record. The pattern of slow ice loss persisted through April—average extent for April ranked tied with 2002 as tenth lowest in the satellite record. The rate of ice loss picked up towards the end of May, dropping extent below the interdecile range after spending most of the month just above the lower part of the inter-decile range. Nevertheless, extent for the month as a whole ranked only thirteenth lowest in the satellite record. June sea ice loss proceeded at only an average rate and by month’s end, it was clear that a record low September ice extent was highly unlikely.

While the rate of Ice loss picked up in August, the sea ice minimum reached on September 19 ended up as sixth lowest in the satellite record. The April through August average air temperate map at the 925 hPa level tells the basic story of 2023—temperatures were below average over much of the Arctic Ocean, limiting melt (Figure 4a). The year nevertheless ended up as notable for the combination of ice-fee conditions in the southern (Amundsen’s) Northwest Passage route and very mild ice conditions in the northern (deepwater) route; this is consistent with the above average temperatures at the 925 hPa level over the Canadian Arctic Archipelago.

Multiyear ice extent at the end of summer was quite low, as has been the case for the last several years (Figure 4b). A band of multiyear ice persists on the Atlantic side, extending from the Laptev Sea across the Arctic north of the Kara and Barents Seas. Much of this ice will likely drift out of the Arctic Ocean with the Transpolar Drift Stream over the next several months to a year.

Very little of the oldest (4+ years old) ice remains in the Arctic, with small patches north of Greenland and an area north of the Beaufort Sea. The total extent of the oldest sea ice is 93,000 square kilometers (36,000 square miles), the second lowest in the satellite record since 1985, only higher than 55,000 square kilometers (21,000 square miles) in 2019. This is in stark contrast to the 1980s when old ice covered over 2.5 million square kilometers (965,000 square miles) of the Arctic Ocean.