NSIDC 2021 Arctic SIE September average: June poll

[Juan C. García]

ATTENTION: There are two ASI extent polls on the ASIF. This one is for NSIDC sea ice extent monthly minimum or September average (which is also used for the SIPN sea ice outlook), the other is for JAXA sea ice extent daily minimum [provided by JAXA (ADS-NIPR-VISHOP)].

Make sure you are aware of the difference before voting.

These are the September averages on 2000-2020 (in million km²):

Year      Extent
2000        6.25
2001        6.73
2002        5.83
2003        6.12
2004        5.98
2005        5.50
2006        5.86
2007        4.27
2008        4.69
2009        5.26
2010        4.87
2011        4.56
2012        3.57
2013        5.21
2014        5.22
2015        4.62
2016        4.53
2017        4.82
2018        4.79
2019        4.36
2020 (*)   4.00

* Corrected by NSIDC. According to ASIF records, it was 3.92 million km2.
https://forum.arctic-sea-ice.net/index.php/topic,3229.0.html

From lowest to highest:

[Juan C. García]

This year I start in the 4.00-4.50 range.

[oren]

Voted 4.00-4.50 with honorable mention of 3.75-4.25 .

[Brigantine]

4.00 - 4.50

40% weight on 4.23 ±0.60, 40% weight on 4.23 ±1.06, 20% weight on 4.00 ±1.60
Assumed bin chances: (>5.25) 2%; 3%; 7%; 14%; 20%; 20%; 15%; 9%; 5%; 2%; 1%; 1%; 1% (<2.50)

[Paddy]

Guessing it'll be similar to the 5 year average for now (4.25 - 4.75).

[The Walrus]

Guessing it'll be similar to the 5 year average for now (4.25 - 4.75).

I did the same.  No reason to think that there will be any great divergence from the trend this year.

[KenB]

3.75 - 4.25

[Stephan]

I voted for bin 3.75-4.25 M km²

[Aluminium]

4.0-4.5

[Juan C. García]

If we analyze the values that we had on the first 8 days of June, the year 2021 seems to be as normal as the average of 2010-2020... If the melt does not happen on 2021, it will happen on the following years.

So, I am changing my vote to 3.75-4.25M km².

One day left to vote or change your vote. 

[gerontocrat]

Some NSIDC extent data as at 8th June for your edification and delight.

The first image is the September Monthly Averages graph. Sept 21 is based on the  average of the previous 10 years melt from now to minimum (currently estimated to be on 16th September).

The September '21 average comes out at 4.58 million km2, some 255k km2 above the long-term linear trend value of 4.33 million km2, and 587k km2 above the 2020 average.

This has a lot to do with a slow start to sea ice extent loss, extent currently 7th lowest in the satellite record. This is in contrast with sea ice area, which is currently 2nd lowest in the satellite record.

So the question is whether extent will catch up. i.e. by future sea ice extent losses being above the average. Future melt of 5% above average would knock about 350k km2 off the current September projection. Of course you could go for the nuclear option (2012) with above average melt from now at 18% above the 10 year average, knocking 1.25 million Km2 off the current September projection, i.e a Sept 21 average of 3.63 million km2.

Given the state of the Atlantic Front, I am inclined to believe in a greater than average future melt to some degree. Being a wimp, I choose the 4.0 to 4.5 million km2 option, ranging from below the trend value to a bit below the current average projection. Indeed I will state firmly that it will be 4.25 million km2 (fool that I am).

[slow wing]

"Between 3.75 and 4.25 million km^2"

Thanks for doing these polls, Juan! 

[Richard Rathbone]

The SIPN report for June is out. https://www.arcus.org/sipn/sea-ice-outlook/2021/june

PIOMAS predicts 3.73 (with 0.4 SD) which is the lowest, the median SIPN prediction was 4.37

In their review of the record of SIPN I was amused they found that the ice is a better predictor of the scientists, than the scientists are of the ice! i.e. people in general are putting far too much weight on what happened the previous year when recalibrating their models for the current year and the models duly predict the current year will be rather like the previous year, while the Arctic actually has almost no memory from year to year. There's a trend, but virtually zero persistence of anomalies.

[Juan C. García]

The SIPN report for June is out. https://www.arcus.org/sipn/sea-ice-outlook/2021/june

PIOMAS predicts 3.73 (with 0.4 SD) which is the lowest, the median SIPN prediction was 4.37

In their review of the record of SIPN I was amused they found that the ice is a better predictor of the scientists, than the scientists are of the ice! i.e. people in general are putting far too much weight on what happened the previous year when recalibrating their models for the current year and the models duly predict the current year will be rather like the previous year, while the Arctic actually has almost no memory from year to year. There's a trend, but virtually zero persistence of anomalies.

I am surprised of how low they are. Eight are below 4 million km2. That is lower than the NSIDC 2020 September average. That is, 8 predictions put 2021 as the second lowest on record.

[Glen Koehler]

The SIPN report for June is out. https://www.arcus.org/sipn/sea-ice-outlook/2021/june

PIOMAS predicts 3.73 (with 0.4 SD) which is the lowest, the median SIPN prediction was 4.37

In their review of the record of SIPN I was amused they found that the ice is a better predictor of the scientists, than the scientists are of the ice! i.e. people in general are putting far too much weight on what happened the previous year when recalibrating their models for the current year and the models duly predict the current year will be rather like the previous year, while the Arctic actually has almost no memory from year to year. There's a trend, but virtually zero persistence of anomalies.

     Here are the details from the SiPN report that inspried Richard Rathbone's comment about scientist predictions:

"Looking back at the 13-year (2008–2020) record of all June Sea Ice Outlooks, we found that there is a lack of correspondence between the median of all outlooks and the actual September observed value: the two time series correlate at r = 0.24. However, the observed and predicted time series correlate significantly (r = 0.86) when introducing a one-year lag difference. That is, the observed September extent of a given year is a very good predictor of the SIO predictions issued the next year. (See: Predicting Predictions: An Emerging Pattern in the Sea Ice Outlooks Discussion below).