Poll: 2024 Antarctic JAXA extent minimum

[Steven]

Today it's summer solstice and about 2 months before the seasonal sea ice extent minimum in February.

Recent conditions:  Antarctic sea ice extent has been record low for most of 2023, but meanwhile it has slipped back to 6th lowest place for the date.

February daily minima for the last 10 years:   

2023   1.95 million km2
2022   2.13
2021   2.79
2020   2.76
2019   2.42
2018   2.21
2017   2.15
2016   2.66
2015   3.59
2014   3.54

[oren]

Thanks for starting the poll. I went for 1.8-2.2, despite slow losses and the season getting long in the tooth, I think the ice is probably still quite vulnerable, due to the decimation we saw last year. Weather permitting we can still go low, even record low in some scenarios.

[HapHazard]

It feels like oren pre-quoted me.

(as in, exactly my line of thinking)

[Stephan]

I chose the 2.2 ± 0.2 M bin because I think we will not see a record minimum.

[Steven]

I voted for 2.2 to 2.6 million km2.  This is based on a linear regression using NSIDC area as predictor variable.  Given the slow melt in the most important region for the minimum (Weddell Sea), a record low doesn't seem likely at this moment.

[interstitial]

2 to 2.4
A warm ocean makes for fragile ice overall but excessive melting provides a steady stream of cold water collecting in the Weddell sea.


Higher local ocean temperatures increase sea ice melting.
Higher local air temperatures melt glacial ice sending more cold water to their drainage seas temporarily protecting sea ice.


I understand at this time El Nino predictions lean towards continued to super El Nino. I know this is a measure of higher ocean temperatures in prescribed locations but I am not sure how that manifests in Antarctic waters. If as I suspect that means mostly warmer Antarctic waters that suggests declining sea ice.


I am unaware of predicted Antarctic temperatures as summer approaches in the region.

[gerontocrat]

The peak of the melting season is over, and daily sea ice losses remain well below average.

An El Nino can apparently inhibit Antarctic sea ice loss, though it can also do the opposite depending on the location of maximum sea surface temperatures in the East Pacific tropics.

Surface air temperatures over the Antarctic seas seems to be mostly cooler than average.

Weddell Sea Extent (5 day average)  is currently 18th lowest, area 27th lowest, and sea ice concentration is 4th highest, i.e. very little easy to melt low concentration ice.

Average extent losses from now would give a minimum of 2.37 million km2.

My guess is 2.5 million km2, so I vote for 2.2 to 2.6

[Steven]

About 24 hours left to vote in this poll, or to change your vote.

[Phil.]

Getting interesting.
https://forum.arctic-sea-ice.net/index.php?action=dlattach;topic=1759.0;attach=398134;image

[gerontocrat]

My guess on December 31 of a 2.5 million km2 minimum already looking very sick, with a latest arithmetical projection of 2.11 million km2, even though the surface air temperatures over the Antarctic seas have been and are still largely below average.

Ho hum.

[kassy]

Well it left my bin now. At least i am equally skilled at predictions on both ends. 

[Steven]

My guess on December 31 of a 2.5 million km2 minimum already looking very sick

Looks like my vote was also too high.  Extent losses have been above average in the last few weeks and even a record low minimum seems possible again.  We'll know in a few weeks.

[Stephan]

I chose the 2.2 ± 0.2 M km² bin and expect this range to win in the end.

[oren]

With extent having dropped to 2.366Mkm2 today, we have one bin out of the running, with 2 of the 15 voters "wiped".
7 voters are currently "in the money", and 6 other votes are subject to further drops.

[vox_mundi]

Deep Learning Forecasts Antarctic Sea Ice Trends for 2024—Projected to Remain Close to Historical Lows
https://phys.org/news/2024-02-deep-antarctic-sea-ice-trends.html

The sustained breaking of historical lows in Antarctic sea ice has prompted international concern and scrutiny regarding the upcoming summer SIE in 2024. In November 2023, a research team based at Sun Yat-sen University and Southern Marine Science and Engineering Guangdong Laboratory (Zhuhai) initiated an assessment of the possibility of new record lows in February 2024, following the "least sea ice year."

To address this question, they employed a Convolutional Long Short-Term Memory (ConvLSTM) neural network to construct a seasonal-scale Antarctic sea ice prediction model. Utilizing deep learning methods, the team conducted predictive research on Antarctic summer sea ice for the 2023/24 season.

Their latest study, published in the journal Advances in Atmospheric Sciences, demonstrates the model's success in capturing interannual and interseasonal variability in SIE over an eight-year reforecast experiment.

The model also accurately reproduced three historical summer lows in 2017, 2022, and 2023, with errors smaller than those in the European Centre for Medium-Range Weather Forecasts (ECMWF) predictions.

What lies ahead for Antarctic sea ice in 2024? The deep learning model's predictions from December 2023 to June 2024 indicate that Antarctic sea ice will remain near historical lows. However, it is highly unlikely to set a new record low in February 2024. The predicted SIE for February 2024 is expected to be 2.105±0.453 million square kilometers, slightly higher than the historic low in 2023.

According to the latest satellite observations, the SIE in December 2023 and January 2024 fell within the range of one standard deviation of the predicted value. Especially, the predicted SIE in January 2024 differs from the observed value by only 0.0015 million square kilometers, indicating the reliability of the prediction.

The results have been submitted to the SIPN-South international call for the 2023/24 season, making it one of the 15 participating contributions. The study underscores the significant promise of deep learning methods in Antarctic sea ice prediction.

Xiaoran Dong et al, Deep Learning Shows Promise for Seasonal Prediction of Antarctic Sea Ice in a Rapid Decline Scenario, Advances in Atmospheric Sciences (2024)
https://link.springer.com/article/10.1007/s00376-024-3380-y

[oren]

With extent having dropped to 2.366Mkm2 today, we have one bin out of the running, with 2 of the 15 voters "wiped".
7 voters are currently "in the money", and 6 other votes are subject to further drops.

With a print of 2.173 today, we have another bin out of the running, so a total of 4 voters are now out. OTOH a new bin with 3 voters is now in the money, including my own vote. 

[HapHazard]

 

[gerontocrat]

There is IMO no doubt JAXA sea ice extent has passed the minimum.

The minimum for 2024 was on 17 Feb at 2,090,437 km2, 0.14 million km2 above the 2023 record low minimum of 1.95 million km2 and 2nd lowest in the satellite record.

I attach graph & table of the minima since 1979.

click image to enlarge
click table twice & maximise for fullest size

[oren]

8 out of 15 got the votes right, not too bad for Antarctic extent where year-to-year volatility is quite high.

[kassy]

Lets see if we can do it again next year. Maybe add an early poll to see how those results compare.