« The pitfalls of prediction | Main | Greenman 3610's video on the 2011 minimum »


Feed You can follow this conversation by subscribing to the comment feed for this post.

Chris Biscan



Unfortunately the ice is not forming there yet.

Kevin O'Neill

Chris, 2010 didn't really see ice forming in the NW passage until the 1st and 2nd weeks of October. Noticeable beginning around the 4th and more than half full by the 10th.


Chris Biscan

pretty crazy considering the entire CA water passages were iced over all year just 30 years ago!

Christoffer Ladstein

Great choice of music Neven! And also a truly incredible melting season. Though I must admit spending too much time on this "affair", and as Kevin so wisely wrote some blogposts ago, a less frequent visit to both this blog and the "attached" exciting chartspage, would perhaps lead to both better health and marriage; I might even get around to read some books....
Nevertheless, when momentum is gone, its time for other affairs. Have fun guys, it's been a joy to hang around, next year will come faster than we like to think...
Last I end with a song from the latest musical heroine from Norway: Ane Brun : Do You Remember (2011...?!)

Andrew Xnn

Curious to know what others may think about the potential for compaction this winter. Apparently there hasn't been very much this summer and that is a significant delta from 2007 (the other being volume).

So, might 2012 be a "recovery" year like 2008 was or could it turn into a year of decisive records?

Anyhow, I'll try to steer clear of the Joe Bastardi school of baseless predictions.

The Blanchard-Wriggleswort paper tells us that good predictions can be made starting in the fall and that volume is one of the better indicators.

Andrew Xnn

Interesting how the DMI 80N Temperature graph is showing below average temperatures while the NCEP diagram is showing almost all yellow (between 5 to 10C positive anomaly) for most areas north of 80.

Andrew Xnn

30 years of Arctic Sea Ice Septembers

Kevin O'Neill

Andrew, the DMI uses a 1958 - 2011 mean; the NCEP graph is operational data and uses a 1985-1996 climatology. I think we're comparing apples to oranges. I


Hi Neven, thank you, and your contributors, for your intelligent commentary on the melt. Good luck with the winter, I'll be keeping an eye out for rain down here in Oz.


There seems a lot of green looking water around the ice pack on Modis. (0 to 150W ish directions)


Anyone understand what is causing this?


Great blog and analysis. Hope you find an appropriate balance between hibernation rest and more interesting posts for the winter and the next melt season isn't 'too interesting' ;o)

Peter Ellis

DMI averages the data based on a regular 0.5 degree grid. That means it weights the region from 88.5N to 90N the same as the region from 80N to 80.5N despite the fact that there's a 40-fold difference in area.

Ergo, the DMI value is very strongly weighted to the area immediately around the Pole and neglects the warming areas around the periphery.

Andrew Xnn

Thanks Peter & Kevin;

Didn't realize how distorted the DMI 80N value was. Come to think of it, the NCEP diagram is also distorted. Believe the diagram below gives a better picture...

Some colder than normal areas, but mostly warmth.


Noel Ward

Neven, thanks for all your amazing efforts, fine commentary, and for bringing the whole crew here together this summer. It has been an exciting (and quite educational) one for sure. And I somehow think next melt season will bring more changes. Now to watch what happens this winter... while hoping it brings good snow and skiing.

Aaron Lewis


Not that long ago, snow in the Arctic was rare. A little diamond dust perhaps, but there was not enough water vapor in the air for significant snow. This tells us how much the Arctic has changed in the last 30 years.

And, a blanket of snow over the sea ice will tend to keep the sea ice from freezing thick and hard.

Rich and Mike Island

When arctic sea ice extent expands by more than 100,000 sq km in a day, what is the term for that? Like century breaks in the melt season.


Rich and Mike Islands, October is the month of daily increases over 200K (IJIS SIE). These are called 'double centuries' in my spreadsheet.

2005, 2006 and 2008 had 3. 2007 had 5. 2009 and 2010 didn't have any, but neither did 2003.


Although we have sea ice as our main objective, we’d better keep an eye on the whole changing Arctic scene. Losing a lot of sea ice is one thing. Another is the ever growing retain of heat in the seas.
I did a comparison of NOAA/NESDIS maps on SST anomaly during 2006-2011 for 22-24 september. I focused on the Barentz and Kara Seas. Maybe it’s an open door I’m kicking in, but they’re both very warm. In fact, average anomaly has never been as high these years than now.
It’s reflected in the very northward boundary of the sea ice in that region. Refreeze doesn’t seem to be in a hurry, too.
I still wonder what effect this may have on fall air circulation.
Hudson and Baffin Bay SST’s don’t look extraordinary this season. The polar jet may be wandering again, but different from last year?
Last two years, december snows and cold were directed to western Europe. Could the cold head for Siberia/Greenland, leaving Europe in a long lasting warm fall? Or is it going to be snowmageddon for the euro-zone?

Chris Biscan


I wrote a post on American about what you just said and I deleted it before I went to post it because I already get flamed for talking honestly about the arctic. I might post it anyways now that I see I am not the only one seeing this.

quite amazing changes happening.

Kevin Adams

@crandles: I was wondering about the green water, too. Could it have something to do with the very low angle of incidence of the sun now? The clouds look weird, too. (Same effect shows up in previous years around this time of the year, too.)


Now, everybody sing!
I tell you, my dear friends, oh, how wonderful it is, no matter wherever you are, we can all be happy singing, no matter if you're out there parked in that car along the highway, or whether you're at home, sitting by the radio, or whether you're having a meal, or whether you're by the TV set, let everyone sing about the melting icecaps, how they're coming down into the sea, and let us all have a swimming time,
as we sing...

Those words are from 1968.
Plus ça change, plus c'est le même message.


A big shout out goes to newtownian 1 who commented about this song over at the Grauniad:

Please take a slug of something to steady your stomach as you delve into the mire of irrational comments which always seem to follow any environment news article over there.

Thank you Neven and all contributors to these discussions. My own blogging fell off a bit lately due to illness, but I was happy knowing that my 'Arctic Irregulars' could avoid withdrawal symptoms by coming here.

Kudos and an e-beer for everyone!


Artful Dodger

Adding additional words of praise for Neven's accomplishment: Newer members may not be aware that, prior to starting this blog in June 2010, Neven was not fluent in English.

Gefeliciteerd, mijn vriend. Goed gedaan!

On changing patterns...
Yesterday I stumbled onto a book by Arnold Taylor, ‘ The dance of Air and Sea’. It seems a good introduction into the intertwined maze of our biosphere. Having taken just a brief look, I found no prediction as to what the Barentz/Kara SST’s are going to do to fall weather. But I’m glad to see there are scholars publishing on this sort of thing.
Better still, in a comprehensible fashion. I thought I’d just mention this for all interested followers on Neven’s blog. You can be sure it’s all there, from the thermohaline circulation to ‘very high atmospheric refraction’. And from changing plankton growth to the spread of bird species.


Thanks for all the kind words. My ego is jumping with joy.

And thanks too for keeping an eye on those SSTs. I had the impression yesterday, watching the DMI SST anomaly map, that SSTs were going down on the Pacific side of the Arctic (anomalies never got really high in the East Siberian Sea, BTW, which also explains the survival of the Arm a bit). But below Novaya Zemlya it's as red as red can be.

This is definitely the thing to keep an eye on for now.

Ben Burch

It occurs to me that when you have a sea ice slush, adding snowfall could cause a quick surface freeze by changing the concentration of salt at the surface.

Kevin McKinney

Belated congratulations on a successful second season, Neven! Glad to hear you plan a less-deep hibernation this year.

Ned Ward

Neven, you've pulled together an excellent virtual community here. I've learned a huge amount from reading the posts and comments over the past few months, even when I didn't always agree with everyone. Thank you very much for all your work.

L. Hamilton

If Neven and I can draft such a thing during the freeze season, I'm thinking this blog deserves its own journal article. It's something new under the Arctic sun.

L. Hamilton

I've posted "final" versions of my daily CT area, UB extent, and IJIS extent bar charts, as those minimums won't be going any lower this year.

The PIOMAS volume results from September aren't out yet, that shoe has not dropped. I'll update the green graph when it does.

Also coming next week: a grand UB/NSIDC/IJIS/CT comparison plot of September mean area and extent, 1972-2011.

Finally, it should be entertaining (for me at least) to look at how all those Gompertz predictions for area, extent and volume turned out, and see whether it's worth offering the same for next year.



Thanks again to all of you! This blog will be even better in 2012 and 2013.

I never thought covering a catastrophe would be so much fun.

Wayne Kernochan

@Neven: for some reason, typepad wouldn't let me post this last night.

Hopefully by now you know that I echo the praise and thanks of others in this blog.

However, I do hope that you will post an assessment of the volume and thickness minima, at the least, before you declare that the "season is over." Because the thickness minimum tends to happen several weeks after the minimum area (all that thin new ice), that part of the season isn't necessarily over yet.

@L Hamilton: I for one would love to see how the Gompertz predictions for volume and area turned out, and would be very interested in their new predictions for next year. Because the minimum area was a bit more than I expected, there was little addition to "negative volume" this year, so I think that the Gompertz fit for volume might still be pretty good next year.


However, I do hope that you will post an assessment of the volume and thickness minima, at the least, before you declare that the "season is over."

You bet, Wayne. Of course, the PIOMAS volume record from 2010 was broken already last month.

But I'll definitely put posts up when NSIDC, SEARCH and the PSC have more news in the first week of October.

Pete Dunkelberg

While waiting for October here are a couple notes from Alaska:

Observations of Climate Change from Indigenous Alaskans

The temp dropped below 32 on Saturday in Barrow, AK for the first time since June 29th. 85-day streak > 32 is longest on record! Previous record 68 days. The average temperature over the streak was 41.9°. The September average temp has been 5.2 degrees above normal.

Average first freeze is September 7th in Fairbanks - have not had a freeze yet. This is the 6th year in a row the first freeze has taken place after September 20th.


Thanks for that, Pete. That's great info.


And now for something completely different (but not entirely unexpected):

Romance in the NWP


There are some big lows forecast for the central Arctic Basin (the AO index is forecasted to shoot way up). What will this do for the ice edge of the pack on the Atlantic side, and thus the waters there, and thus the atmosphere? Exciting.

Chris Biscan

it is amazing how much the ice on the Atlantic side.

The Fram, Greenland Sea, Barents and into the Kara is still taking a melting beating.

the Operational GFS and EURO keep showing the arctic above normal.

But also those super anomalies to return by October 7th-10th over NE Canada.

the same thing we have seen now for years.

Chris Biscan

It is clear even on September 25th to me that the refreeze season is not your grandfathers, or even fathers, lol

Or even older brothers.

This is an entirely new arctic.

At this point the sun going down forces the refreeze. We have seen that in 2007, 2008, and likely again in 2011.

However, this year we me see parts, expecially around 70N or so that will be even more stubborn.

Probably going to see another super late Hudson refreeze.

maybe a new Hudson record for completely frozen over.

ice thickness in the CA, Hudson, Barents, and Kara will be weak and likely skewed by snow.

Rich and Mike Island

Thanks Neven!


I’m trying to finish a calculation anomaly/area on the 2.3 million square km’s of the combined Barentz and Kara Seas. I can’t wait to blog something before I finally have a fixed result on the NOAA/NESDIS maps.
From the gut: anomaly is twice as strong compared to 2010. Of course, that is based on area, not sea water volume. It may not go very deep...
But this could stand for a lot of heat to be released, before any refreeze is possible.
What will happen when it’s released? A very high 250MB level? Resulting in a cold vortex like over the CA last February? It could potentially change the polar jet like we never saw before.

Chris Biscan

what forces the heat out?

Lack of sun?

the weather pattern continues to help keep that area pretty warm for the foreseeable future.

Rob Dekker

Guys, you make me believe that you are all going to go into hybernation now that the melt season is over. Let me say that I hope that some people will stick around, because the fall season (and certainly winter) will be very interesting.

Remember that that we just his the second-to-lowest minimum ice extent in recorded history, and that the vast areas of open ocean contain a massive amount of heat down to quite deep below the surface.

It is going to be very interesting to see what happens with that heat, and how it affects the Northern Hemisphere's atmospheric patterns in fall and winter, and how much that contributes to reduced freezing during winter.

And I'm sure there will be reports of Arctic wildlife and how it handles the "new Arctic reality" of hundreds of miles of open ocean between ice and shore.

So, I will stick around, as I hope many of you will, but let me express my gratitude to our host Neven as well. This was my first Arctic melt season with Neven's blog as a 'primary meeting point', and I hope there will be many to come.

By the way, Neven, are you Dutch too ? Or did you just live in the Netherlands for a while ?

Wat is het met die Nederlanders, dat ze zo geintresseerd zijn in zee-ijs en wetenschap en ongerepte natuur ? Is het misschien omdat in Nederland alles al 'gekultiveerd' is ?


Ik denk dat het door de Elfstedentocht komt, Rob! ;-)

I was born in The Netherlands, lived there for over 30 years, but currently live in Austria. It took me a while to realize you were Dutch too, Rob (we mailed in English).

I guess everyone who develops an interest in the Arctic (sea ice) does so because it's one of the main front lines of AGW. Then they find out it's a captivating subject in itself. At least, that's how it worked for me.

Not much hibernation this year, Rob. I want to increase my understanding of the freezing season.

Rob Dekker

By the way, does anyone know when PIOMAS will be updated ? Seems that the last update is a month and a half ago..

Chris Biscan

i was reading Steven Goodards website, he is insane.

He also uses norsex. He gets owned over and over.

Who is this crazy man?


Re "By the way, does anyone know when PIOMAS will be updated ? Seems that the last update is a month and a half ago.."

Last data is 4.275 for day 243 = 31 Aug 2011. It gets updated monthly; it was about 6th or 7th for the August data.


what forces the heat out?

Lack of sun?

Indirectly, I guess. I thought it was the difference in temperature between atmosphere and ocean, that makes the water release its heat. Lack of sun -> lower air temps -> heat release.

Chris Biscan

Looking at todays Bremen prelim. We might see a drop today on Bremen. Jaxa would be harder since it has a 2 day running mean.

but definitely going to see a few days at least of meager gains. There is no where yet for the new ice to form.

The ESB is cold as bleep at the surface, there must be to much warm water underneath stopping ice to spread there.

Hans Verbeek

what forces the heat out?
Don't forget evaporation. The atmosphere is seldom saturated with water vapour.

Chris Biscan


-8,000K prelim on Jaxa.

Rob Dekker

crandles, on PIOMAS Last data is 4.275 for day 243 = 31 Aug 2011

Thanks Chris, I was confused by the PSC graph, which seems to run only until Aug 15.

So, 4.275 Mkm^3, over 3.8 Mkm^2 or so area on Aug 31, that is 1.12 meter average ice thickness, folks !

That seems consistent with ice measurements by US Army (crrel), and with the Healy happily plowing through much of the area at ice crunching cruising speed.

One may wonder how much thinner it can get before it simply breaks up in little pieces and melts away one summer in the near future with a bit of Beaufort gyre and Fram Strait export.

Rob Dekker

Sorry, that is 4275 km^3 over 3.8 Mkm^2.


The Green Blob north of Svalbard was extremely vivid in yesterday's Modis - could algae be responsible for this?

Also interesting to note that at
Ward Hunt is now designated as Ward Hunt East and Ward Hunt West.

Pete Dunkelberg

News flash from the RC cosmic rays & clouds topic:

Jeff Pierce says:
28 Sep 2011 at 6:41 AM

Interesting timing.

It looks like we might be in the middle of a Forbush decrease right now…


(note, this updates real time, so you view this in the future, it won’t show the decrease in Cosmic rays near the poles that I see.)

There was a coronal mass ejection that just hit the Earth (http://spaceweather.com/archive.php?view=1&day=26&month=09&year=2011).

muoncounter, you seem to know your stuff in this area. Can you weigh in on this?


Thanks, Twemoran. I've put a post up.

I've been wanting to do a piece on this for a long time, but Derek Mueller's work has made it much, much easier (some great images on his web page).


RE: Logicman

Two days ago, Logicman posted a link to an article on the Guardian website. The subject matter was ice loss in general, and was written by Damien Carrington. I wasn't particularly impressed by Carrington's understanding of the topic, and, consequently he rather left himself open to attack.

However, I think Logicman's point was to read the comments - these were frankly astonishing. I often go on the Telegraph or Daily Mail blogs to play with the natives, but this was something else.

If it wasn't for the fact that these people actually get to vote it would be bloody hilarious.

The line that seemed particularly germane to this site was..."The only large place on Earth that is currently warmer than it was 30 years ago is the Arctic."

I seriously doubt if some of these people could find their own backside using both hands.

Cheers Bill F

PS well done Neven


Hi all,


There is now just a couple of days until the money contributed to this blog are due to be distributed to a good cause.

I would like to suggest that over the course of this melting season I have become aware that there is on the internet one source of information on Arctic Sea Ice which surpasses all others. And this is it.

Many of the other organistations reporting on Arctic Sea Ice have multi-milion dollar budgets. NASA. NSIDC. Bremen.

Okay, they have better satellites. But this is the best forum, AFAIK, for a comprehensive overview of what these satellites report.

I feel Neven should keep the money and dedicate it to the further development of this blog. This will amount to an insultingly small sum of money for the amount of time he has apparently spent promptly, courteously and generously editing our various musings here.

From my own point of view, I have learnt huge amounts here from the postings of FrankD, Lodger, Randles, Yooper, etc, and followed threads though to Patrick, ChrisR, etc, etc...

But none of thes people contribute to NASA's forum. NASA ain't got one.

It is impossible for me to conceive that moderating this forum has not caused some distraction from our poor little Dutch boy's attempts to keep his fist firmly inserted in the economic dyke...

It would, IMO, be a tragedy if we all had to convene elsewhere due to a change in Neven's circumstances. Repeat " a tragedy " - oh sure, if this shuts down, any number of people could replicate something similar; but Neven is quite good at what he does. For a cloggy.

Artful Dodger

Rob D: I may snooze, but I won't be hibernating... we've got that little thing about the IJIS 1-Day to work thru.

Hint to Neven: do you have a recent history of SIE updates reissued after the Daily Final is released?

Idunno: I concur. I'd like to see Neven buy some green electricity to power his computers, and hence this blog.



Chris Biscan


animation I made of the NW passage ice vanishing in just a few years.

quite amazing for natural variance.


Nice animation, Chris. I'm waiting for 2006 and 2007 to show up again in the UB SIC archive, and then I'll do a post.


I'd like to see Neven buy some green electricity to power his computers, and hence this blog.

I of course buy my Ökostrom green, but the greenest electricity is the one you don't use. That's why my desktop+monitor use 45 W at idle. My laptop uses 11-15 W at idle.

It is my ardent dream to own a 3 KWp solar panel array (which would hopefully cover all our resident energy needs, heating included). But I still have to build a house first, so I have something to put it on. :-)


But maybe I can get TypePad to invest some of those profits to reduce power usage of their data centres.

Wayne Kernochan

I was staring nostalgically at the volume graph in Neven's main graph page, and it suddenly occurred to me that it appeared that the years of a major volume-minimum downward drop were immediately preceded by winters of unusual volume-maximum downward drops. This was especially true in 2006-2007 and 2009-2010.

If this is true, why? (after all, there are also some years of unusually low or high maximum-to-minimum drops) Is there some signal in the refreeze period that tells us a big volume-minimum drop is coming? Can we factor that into our forecasts of area and extent?

What I guess I'm saying is that as we watch extent and area rise over the next few months, we may pick up some hints of a major drop in volume ahead which would imply a significant drop in area and extent compared to 2010 -- making the next few months a bit more interesting.

But I still can't figure out why a smaller refreeze would happen some years instead of others. - w

Wayne Kernochan

oops, I meant to say "compared to 2011" - w

Chris Biscan

I too agree the extra money should be invested into the blog some how.


Hold your horses, guys. The donate box closes at October 1st. I'll then put a post up to tell you which villa I bought. I mean, a post to democratically decide what the money will do.

Chris Biscan


the ice almost didnt make it this eyar


ECMWF and the AO index forecast ensemble are finally showing some highs moving in in about 4-5 days. I wonder what that will do.


On SST… how convenient, consider this, Wayne...
I finished my CAD-count on the Barentz Sea. The difference between 2010 and 2011, taken on 23 september, is striking. Based on area/NOAA-NESDIS map the anomaly is plus 2.2 degrees averaged over 1.4 million square km’s. To my surprise, 2010 came up with 'only' plus 0.5 degrees.
Over the Kara Sea I calculated an even higher average: plus 3.3 degrees over 0,88 million km².
I haven’t yet compared that to 2010, but it is evident the anomaly was around plus one then.
These Seas averaged -2 to +9 (Barentz) and +6 (Kara) degrees between 1970 and 2000. So 3.5 is now 5.7 (Barentz) and 2,0 is now 5,3 (Kara) degrees on average.
That is a lot of energy.
Normally most of it is emitted to the atmosphere during november-december (while a part is consumed in refreeze). If it is, be sure that release is going to take stage. But if clouds and winds interfere, a lot of energy will remain stored. In that case, the sea ice boundary isn’t going to get far to the south, maybe in february/march. Very thin.

When the energy is released, the proportional area of about 5% of the Polar region combined with the sheer amount of heat, will be sufficient for a strong local effect and noticeable overall influence. The outflow aloft could create a large bulge on the upper troposphere. Consequently, the 250 MB niveau should be much higher than usual. The effects on general Arctic circulation may be threefold.

One: it influences/strengthens existing Rossby-wave patterns, carrying blockades

Two: it disrupts normal lower tropospheric flow through a strong cyclonic pattern (snow in Europe!)

Three: it influences the latitudinal circulation in the Polar Cell, strengthening the Polar low level high.
In december the AO should return to negative. Fiercely.
Where does the heat come from? Not just insolation. It seems the Atlantic did some input.

Chris Biscan

-29,000K tonight on jaxa in the prelim


One thing you might want to consider, Neven, is subscription.

Considering how detailed your own and everyone's interest is becoming, you might want to look at getting yourself a subscription to a journal or other information service. Or for running the blog, some new or updated nifty software.

I deeply envy your skill with animations - and it's one of the things that makes this site so attractive to those who stumble across it. Spending some of the cash on gadgets that make these tasks quicker or easier for you would be a great use of the money.


Morning Neven, others,
Could you inform me how to make a cropped version of a jpg that will fit the TypePad-column? (The one I'd like to add is about 100 Kb; the httm link shows only 1/3d in the column)

Artful Dodger

Hi Werther. Crop your image as you wish it to appear, and then resample it so that the width is 400 pixels. Let the height vary to maintain the aspect ratio.

I assume you have suitable software to do this, but I can recommend Irfanview for Windows Users. It is simple, free, and powerful.


The illustration to my post above:Photobucket
It is made based on NOAA/NESDIS,which is distorted through its projection. But I adjusted it to the known areas of both seas. You can see the 'shadow' of Novaya Zemlya in the middle. Deep red is >+4 degrees, deep grey is <-4 degrees, both of course in anomaly.
Remember there is a large swath of about 0,25 mkm² north of these seas, above 80 N, also ice free!


That's better, thanks Lodger. In my haste I restrained to 448x336. Next time I'll cut at 400.


Very interesting, Werther, thanks!

Adelady, those are some great tips. I'm already using some online software that monitors webpages for me. I'm going to take a subscription that has added features. I've also donated some of the donated money to Peter Sinclair for making the latest Arctic sea ice video.

I'll write a special post on this in the coming few days, with all the numbers, and then we'll have a look.

Chris Biscan

looks like the sun setting has something to do with the modis images having that big empty spot.

Wayne Kernochan

@Werther: Sorry, it took me a while to get to this (a minor matter of an estate :)). That's great stuff.

It seems as if you are predicting a significant bump down in maximum volume in March, because of the remaining insolation around the Barents and Kara Seas, especially if over the November-December run-up to minimum temperature clouds decrease heat release and winds add energy and thus heat. We should see signs of this not only in PIOMAS volume but also in area anomalies in those areas.

As I remember, I had a discussion with crandles about this last year, relevant to how fast we would proceed to no ice year-round once we reached no ice at minimum. Iirc, he said that insolation heat should be released rapidly once the sun went down, and I wondered if increasing air temp anomalies in the Arctic during the winter, even if below freezing, would slow this release. It sounds like what you are describing may be a test case for that future.

Finally, I am curious about your statement that "the Atlantic contributed". My image of the Arctic is that warmer ocean water primarily flows into the Arctic from the Pacific, via the Bering Strait. According to "The Fate of Greenland", Gulf Stream water dives down next to Greenland, and the resulting deep-ocean current goes down to Antarctica, then halfway around it, then up the Pacific to surface in the North Pacific near the Bering Strait. So I'd welcome your explanation of how the Atlantic, rather than the Pacific, affects the heat being stored up right now.


The Barentz-Kara SST showdown...
The basis for my guess on an Atlantic warmth input comes from several strains of information:
1. Lodger’s earlier remarks on this (related to salinity)
2. A comprehensive presentation on the Arctic Oscillation see: http://jisao.washington.edu/ao/
3. Different maps on the flow of the water masses like here: http://www.eoearth.org/article/Barents_Sea?topic=49523
The sink east of Greenland is alive and kicking. It shows on the SST maps right west of Svalbard, the pool of warm, saline surface waters. That’s cooled after spreading, then sinks to be one of the drivers of the thermohaline circulation. But the stream’s branching through the region is complicated (and relevant).
I remember having seen great images of the Bering in(and out-)flow too. It seems to have been an important factor during 2007. I think now we’re watching the Atlantic Decadal Oscillation in full swing.
Since 23 september, the anomaly seems to have grown. Look at not only the NESDIS maps, but NCEP for 1 october. A couple of days ago I checked on four weather stations in the region (Longyearbyen, Malye Karmakuly, Ostrov Dickson and Polargmo). All of the stations reported plus 2 to 4 degrees anomaly compared to the same week, most years, since 2006.
So it's not only the SST, but the lower troposphere too.



Is the 'sink hole' the same as the green area that showed last week in modis?


Search Outlook Forecast verifications

The September average extent per NSIDC was 4.61 m km^2

The forecasts of this quantity were:

July, Adjusted gompertz method: 4.4 m km^2 +/- 1m km^2
August, Adjusted gompertz method: 4.3 m km^2 +/- 0.8 m km^2
September new predict change method: 4.45 m km^2 +/- 0.32 m km^2
September, superceded adjusted gompertz method 4.35 m km^2 +/- 0.4 m km^2

Thus the errors are 0.21m, 0.31m 0.16m and 0.26m km^2.

Therefore, the 4 predictions above were out by 0.4, 0.8, 1 and 1.3 standard deviations with all predictions being too low. The results were therefore not disasterously bad being out by about the amount that could be expected. There were several SEARCH predictions that were nearer: For July, 7 were close and 7 further away. For August, 12 were closer and 8 further away. For September, 3 were nearer and one was the same.

However looking in more detail the results may be a little better than indicated by results above. Neither of my methods attempted to predict weather and what effects this would have. My opinion of the weather in early September is that it could have been a lot worse for melting and advecting ice. There is a risk that such a judgement is clouded by wanting to maintain the methods I used should be expected to perform reasonably well and did not do so this year because of the weather. I hope and think my judgement of the weather is reasonably well backed up by posts and comments at Neven's Arctic sea ice blog.

Had the weather been such that more ice had been lost, the predictions may have outperformed other SEARCH contributions rather than underperforming.

As 3 of the 4 predictions were attempting to update a gompertz prediction as carried out by Hamilton, it make sense to also compare to that prediction. All 3 of the gompertz adjusted method predictions predicted lower than the gompertz fit when the outcome was higher. Again this looks poor. However the variability explained by the methods is less than the remaining variability. A large part of the remaining variability can reasonably be described as weather. Therefore we should expect that in some years the weather could act to cause the direction of the change to be wrong in some years. Given the comments on the implications of the weather for ice loss in September I have seen, this could be one such year and the adjustments to the gompertz fit method could still be regarded as being expected to be useful even though the adjustments were in the wrong direction this year.


While the predictions underperformed relative to other SEARCH contributions, this may in part be down to the particular weather conditions in September and perhaps late August. The methods did not underperform significantly and given the weather that occured this year, the method seem worth continuing or improving rather than dropping them.


NSIDC september average extent of 4.61 as posted above is second lowest extent.

2.89 average September area is also second lowest on record behind 2007.

L. Hamilton

The thing about statistical methods is that we hope they work well over the "long run," but can't expect each prediction will be exactly right. Close, we hope, and you got that.

I offered 4 predictions about September ice back in April based on simple Gompertz curves. Here's the tally so far:

NSIDC extent -- predicted 4.4, observed 4.6
UB extent -- predicted 4.6, observed 4.6
NSIDC area -- predicted 3.1, observed 3.2
PIOMAS volume -- predicted 5.2, observed ___
I think my PIOMAS estimate is going to end up rather high, but we'll know soon.


Re "Close, we hope, and you got that."

Thanks, yes, I am content with the outcome. I hope I am not pushing weather too much as the explantion for the differences.

You did really well, closer and with longer lead time.


Arctic sea ice news is out:

"Overall, sea ice in the wider and deeper northern route through Parry Channel reached a record low, according to Stephen Howell of Environment Canada, based on Canadian Ice Service analysis. Parry Channel had a narrow strip of ice that blocked a short section of the channel, but it did appear to open briefly in early September."



Is it surprising how little the strip of 5 years + ice running from East Siberian Sea to Central arctic has moved from March to September?


AMSR-E Ends 9+ Years of Global Observations


The dialogue on SST...
Two late reactions on posts by Wayne and Twemoran. Twemoran, about the green colouring you mentioned in relation to the sink where Atlantic Deep Water is born. I’m not sure where you’re going. Do you mean the green colouring present on MODIS over all visible sea area in that region? But I hold that to be a prismatic effect through the MODIS lens. It’s present since 20 september. Or do you hint at prolific algae bloom? In august we saw Neven’s wallpaper out of algae in the Barentz Sea. I found just two clear days on the r02c04 pictures (the sink is on the upper left side of those pics); 09/17 and 08/18. When you look carefully, you can see algae patterns there, but nothing like on the ‘wallpaper’. I wonder what relation you suppose.
Wayne, you took my SST post in a direction that hadn’t occurred to me then, yet. At present, I’m very occupied by the possible weather feedback on the Northern Hemisphere. You’re right, that there may be consequences for the maximum extent/area and especially, volume of the sea ice this winter. So far, 2011 has been low through each month of the year. It may for the first time be low on the extreme side during the winter months.
One other reason for the high SST’s in the Barentz region could be a temporary strengthening of the North Atlantic Drift. A lot of ice (0,3 MKM² x 1 m) has melted north of Svalbard, Frantza Yosefa and even Severnaya Zemlya. It could have dumped a lot of cold, salty brine down, thus creating room for a temporary warm inflow?

The comments to this entry are closed.