« Oh, and BTW, the Passages are open | Main | IJIS SIE: 5 million km2 mark passed »


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


So this negative ADA might slow things down, or whatever, but not end the melting season. It will if it persists after the 10th of September.

I hope you will excuse my ignorance but how does an Arctic Dipole Anomaly works? It probably has been explained somewhere on this blog if so my apologies for asking the question again.
I know that the DA means there is High over the CA and a low on the Russian side and that it will create anticlockwise winds that will compact the ice. So does that mean that a negative DA means a High above Siberia and a Low above the CA? Right... But how does this affect the ice, does it mean that it will spread out the ice or does it mean cold weather that will freeze up the slush puppie?

I note that the Arctic Oscillation has just gone positive which is not good for melting

Lord Soth

At the start of the melt season, a negative arctic oscillaton (AO) is a benefit.

During the summer a negative AO is benificial because it is a signal for sunny weather and allows for a high level of solar load on the ice.

Now we are in a period of net heat loss in the arctic, so a postive AO is better as it will

A) Brings storms that will break up the thin ice late in the season.

B) The cloud blanket will retain heat, during the period of solar lost, extending the melt season.

The longer the melt season, the greater the chance of getting a key arrangement of storms that will lead to compression, especially in slush puppy years.


I noticed on yesterday's LANCE-MODIS satellite image that we have some holes showing up in the ice pack in the Beaufort region. And the ice is pulling away mightily from Greenland and parts of the CA. Things to keep an eye on.

Seke Rob

Funny numbers, IIRC, 3rd day in a row where the JAXA prelim number, I only ever see 1, ends in 938.

2007 had 60,000 KmSq exact down.


The UK bookies might consider 29.8.2011 to see a nice round number too, in change from the previous day :D


Lucia (The Blackboard)

Paul K,

Lucia, funny how your number keeps coming down... by the time we hit the minimum of 4.20, you should be right on target : )

My estimate for the mean has gone up and sometimes down depending on recent melt and addition of new variables as I find data. The minimum has never been outside the uncertainty intervals.

Lucia (The Blackboard)


To find the MASIE site
Thanks. I already knew where the MASIE data was. By "These regional data are in the CSV file." Do you mean the CSV file here: ftp://sidads.colorado.edu/DATASETS/NOAA/G02186/masie_extent_sqkm.csv

So, was the answer to my question:
By NSDIC data, I mean the online MASIE data which goes back goes back 4 weeks?

Or do you mean some other CSV file with data that goes back longer? If you mean some other csv file, do you have a link?

(I get that you may wish to be making deep and subtle points and may think my question is trying to imply something deep and subtle. But, as you know, my policy is to avoid rhetorical questions or trying to make points with questions. I am trying to find out where you are grabbing data, what that data is etc. My questions don't contain some hidden subtext.)


Rob bear in mind that this is digital data in units of pixels and is not a continuous variable in units of km^2. A number ending in 000 corresponds to multiples of 128 pixels iirc.


Seke Rob

The MASIE csv needs capturing at a minimum on bi-weekly basis or you're bound to miss a day here and there at the top because there are occasional days missing. Got it set up on a daily query at 17:05 CET-DST, just in case. The time is usually somewhere after 16:00 CET-DST for it to refresh, same as CT. JAXA follows that same time, most of the times, to publish the final.


I would assume NSIDC data refers to the txt files available at


eg ftp://sidads.colorado.edu/DATASETS/NOAA/G02135/Jul/N_07_area.txt

They only go back to 1979 and watch out for the step change in the area data as noted.

If the reference was to NSIDC CSV file then maybe it is the MASIE csv file.

Seke Rob

Think they heard me, or 'read' me. The JAXA final

28th 4964063
29th 4896563

Day change: -67,500 KmSq.

Seke Rob

Phil, one time I got into pixel counting business and the hues, digitizing to get unpublished numbers... got me up the wall, so appreciate your observation. It's the light part, for at our races.... top hat and bowtie not required, so we seemlessly can also attend the splatter at the occasional cock/hen fights ;>)

MASIE Spaghetti Chart updated **, with all the new numbers and the changes for the day included. Arctic CB dropped 28KKmsq... that's the one I'm most curious about this part of the track.

** all other charts I do too.:D


As espen noted above fast ice is rapidly becoming a thing of the past.

I don't think that the ice shelf in front of Independence Fjord has disappeared previously, and it's certainly gone today.

Spring Tides seemed to play a role in that the first racks appeared 2 weeks ago after a SP, and the latest one finished it off.


North East Greenland / Joekelbugt:

The ice pack of Joekelbugt is really getting beaten up, more or less all of it is loose now (1000s km2), when looking at the prelim Modis images.
Regards Espen


No problem Rob, thinking over the IJIS situation ending in 0000 should occur 1/64 of the time so if you can find someone who'll give you better odds, I'd take the bet if I was a betting man.


One of the things I treasure on this blog is the capacity of posters to disagree agreeably.
I've made some egregious mistakes and have always been shown the error of my ways in a gentle, helpful manner. This leaves the door open to explore unconventional approaches without fear of ridicule.

Seke Rob

Re: Twemoran | August 30, 2011 at 16:51

Don't remember if it was posted here, but the Tohoku March 11 earthquake related tsunami reached the Antarctic 18 hour later. Scientists raced over to watch as it happened and how 30cm (1 foot) water rise caused distinct and increased calving. See http://www.prnewswire.com/news-releases/tohoku-tsunami-created-icebergs-in-antarctica-127231473.html On those large slabs of ice they anticipated upward pressure, et voilá.


I just published a post on this year's End Zone (not as thorough as last year). I compare sea ice concentration and SST maps.

Chris Biscan, in the ECMWF maps I'm seeing a nice, little cyclone forming between Svalbard and Franz Josef Land tomorrow and then moving towards Siberia. What will that do to the ice there, I wonder. Probably disperse it and push it towards Fram.

Towards the end of the forecast (Sept. 8th and 9th) ECMWF now has a high forming in the Beaufort region. If that becomes a stable feature, the melting season will last a bit longer.

Christoffer Ladstein

The wind is picking up in speed where Healy is located just 4 degrees away from the North Pole, and by the weathermap you are able to tell that the icesheet in the Beaufort/ CA side is compacted in 1 direction; will this continue todays Big extent decrease, in a day or two!?

OT: If you hate rain then go to Svalbard during the summer; they've had only 30 mm there the WHOLE summer! Compared to that Oslo have been practically drowning, close to 430 mm June-Aug., which is 80-100% above normal. A weather related to the Arctic situation?!

BTW: The MYI seem to be "extinct" after this season! Maybe we ought to put it on a Red Alert List just as the Indian Tigers, just to make it look more serious what's taking place SO FAR away from most peoples living rooms!?

Kevin McKinney

Just to point out the obvious, IJIS SIE had a downward revision from last night's prelim, and we are now below 4.9. To be exact:

4,896,563 km2

A bit of a "bang" for the end of August. We'll see what the last 2 days of the month bring.

William Crump


The MASIE plot for this morning shows 2011 as having greater ice extent than 2007. Does that mean there is no significant difference between these two years? Of course not, as this is an insufficient data set upon which to base a conclusion. It looks like 2007 took a big one-day dive in 2007, I do not know if this was due to wind conditions that caused a steep dip in ice extent in Arctic Basin ice in 2007, but we will see it on future MASIE charts. I do not attach any meaningful significance to the one day dive.

The Cryosphere today chart shows 2011 as nearly identical to 2010 even though the difference on the MASIE chart is greater for today - (this may be an area v. extent measurement difference), but who knows what tomorrow will bring. I think the point is that the analysis comparing 2011 to other years should not be done with just data for one day and it appears you agree which IMO undercuts your statement that my statement that there is no significant difference is incorrect.


If I understand the climate v. weather discussions, the importance of a difference in a single year is not what the discussion is about. The differences being discussed are on a time scale of decades and longer. One day doesn't mean much in this discussion. The average amount of ice (regardles of which measurement system is used) for the Central Arctic Basin in the last 7 years is significantly below the average for the period 1979 through 2004. This should be sufficient to generate some action concerning AGW, but it has not.

You stated:

I looked at the data as it stood when I posted and noted a 7.5% difference. I did not reflect on the season as a whole, or whether there was a difference at maximum, or whether I think it will be 7.5% at minimum. Only that at the time you said that, the data did not support you. So at the time you said that, there was a 7.5% difference in area. Clearly you don't think that is significant. I think it is sufficiently noteworthy that your handwaving was worth pointing out.

FrankD, I think what you did is you found a single data point in the extent data that agreed with what you wanted to say and you did not look beyond this one day's worth of data. You did not do an analysis of the area data at Cryosphere today to see if it was showing a similar percentage drop for that day. You did not even do an analysis to see if the average for August for the two years was significantly different even though the data can easily be seen on the MASIE graph which shows that until August 9th there was virtually no difference. If you did an analysis of the whole season that showed a significant difference in the average for both the area and extent, then you would have a valid argument that 2011 in the Arctic Basin region is significantly different than 2010, but that is not what you did and we both know that there will not be a 7.5% drop in the average ice extent/area for the Arctic Basin for 2011 compared to 2010.

You are quite correct that I do not view a single day difference of 7.5% in extent as meaningful in determining whether 2011 is different from 2010. Daily weather and ice fluctuations are quite common and are easy to see on the MASIE chart. I agree that a 7.5% drop in the average for the year is significant, if it occurs. However, the real test is in the multi-year averages. 2007 took a dive in area and extent in the Central Arctic Basin, but the four following years have not repeated these low levels. Using this data base I could say that ice loss "bottomed out" in 2007, but that would be wrong. There will definitely be a future year in which the extent/area is lower than 2007 in the Central Arctic Basin, but there is no data base for this region alone that suggests it will be "ice free" by 2016 or that any month for the Central Arctic Basin will follow the the rate of decline implied by the volume trend line graph.

If the MASIE chart was drawn with a 0 km2 point as the bottom of the y-axis rather than 2.9 million km2 and the chart included 365 days like the ice area chart on Cryosphere today you would not think there was much divergence in the 2011 line and the 2010 line for the Arctic Basin.

I stand by my statement that the rate of decline in area/extent in the Arctic Basin is much slower than the Arctic as a whole and that area/extent for the year 2011 shows very little, if any, change from the year 2010 (please note that I am using an average for the whole year in this statement and not just a single day's worth of data).

I suspect there is a greater difference in the thickness of first year ice between 2011 and 2010, but I have not found any data set that specifically shows the thickness for first year ice in the Central Arctic Basin other than the ICESAT data, which showed a much slower rate of decline than the Arctic wide volume data.

I was intrigued with the data referred to by Neven, but this data appears to be for a very localized area at the North Pole and not a measurement for first year ice in the Arctic Basin as a whole. Also, it is unclear if the prior year data of 2 meters mentioned by Neven included multi-year ice. Assuming the ice is .9 meters thick over the entire 2.5 million km2 Arctic Basin (if area data is used) or 3.0 million if extent data is used (and yes there is the issue of 15% v. 100% ice detection by sensors), then the volume is 2.25 km3 to 2.7 km3. If this volume is not declining as fast as the volume in the Arctic as a whole has declined in the past, then the trend line analysis using Arctic wide data will give an incorrect trend line for making guesses.

As you are aware, I do not trust the Arctic wide data that have been used to draw trend lines as these appear to overstate the melt rate of first year ice in the Central Arctic Basin. The Central Arctic Basin will not be "ice free" unless the first year ice melts out. This is my very simplistic premise.

Any data base for trend line drawing that includes multi-year ice will give a false reading as it is clear multi-year ice is disappearing faster than first year ice. IMO the volume trend lines are at most showing the demise of multi-year ice by 2016 and not the ice as a whole.

I would prefer a data base for the thickness of first year ice rather than a volume measurement for first year ice as the volume measurement may give a too slow a melt reading as first year ice volume has some room for expansion when it moves into areas previously occupied by multi-year ice. If you come across such a data set please send it.

In the absence of thickness data, the trend line analysis(IMO) should be done using a combination of area and extent for the Central Arctic Basin alone (since there is a data base for this, as this ice constitutes more than 80% of the ice that currently survives at the minimum. The Arctic will not be "ice free" until this region is "ice free"

The bottom line is I am concerned that people using the volume trend lines you have assembled will, despite your warnings to the contrary and your own stated position concerning making predictions of

I don't deal in predictions, any more than I can travel through time. As I have repeatedly stated. I have made one prediction in two seasons here - In late June last year I predicted an IJIS minimum of 4.8 million, 13,600 km^2 out on the final number. That is the only prediction I have made here, and the only prediction I will make. I don't even vote in Neven's polls.

try to use the volume trend lines to support a quick melt out position and that when the quick melt out does not occur by 2016, that people will reject the message that AGW is real.

IMO a better strategy is to forecast a slow demise. If the actual melt is faster, then the message is "AGW is worse than the scientists thought", but if people persist in these fast melt predictions and it does not occur, then the story will be told as "AGW is not as bad as the scientists predicted". The latter message will allow people to ignore the obvious and do nothing.


Will, just one scientist has said 2016, plus or minus three years. The IPCC is saying 2100. If there isn't an ice-free Arctic by 2016 and people eagerly deduce from that that the scientists are wrong and nothing is the matter, then there is nothing that we can say or do to make people wake up. Nothing. You can't change minds that actually beg to be deluded.

So we might as well do what we do now, be transparent about it and not worry too much about how everything might be spun by ideologues.

Seke Rob

Re: Christoffer Ladstein | August 30, 2011 at 18:15
The snipped quote below gave me a chuckle, given the short straw that SG pulled by predicting more > 2 year old MYI... 3 hip hip \o/, just for SG.

BTW: The MYI seem to be "extinct" after this season! Maybe we ought to put it on a Red Alert List just as the Indian Tigers, just to make it look more serious what's taking place SO FAR away from most peoples living rooms!


My Photoshop Ice status tells me the we are below 2007 extend/area and well below if ice concentration also is considered, this info is collected by courtesy from the Cryrosphere today map and the Bremen Map and my own studies of the Modis images, I know it is not a scientific way of doing these things but I can not help mentioning it.
Regards Espen


[not really d johnm33 will do]
Great blog directed here by patrick lockerby. I see that Irene is set to start stirring things up should be an interesting week. Thanks to everyone for the insights.

Janne Tuukkanen

Twemoran: I found this study about sediment cores from the area: http://www.polarresearch.net/index.php/polar/article/download/6177/6856

I only read that cursory, but the now collapsed area is mentioned as "fast ice". In fact they speak about "more or less permanent fast ice" during whole Holocene.

So by quoting this publication, I can, having full scientific rigor backing, confirm you are more or less right :)

Chris Biscan

Just did a 30 min run through of the 12z Euro and GFS.

after day 2-3 they are a mess. All over the place.

the arctic is turning colder though. But also chaotic.

The models are always to cold in the medium and long term because of climo.

However without a DPA we won't get to 2007 min on Jaxa this year.

The ice has dodged the bullet.

The last 5 days 20-30 kt winds barely touched the Pack from ESB-Barrents. If those winds were another 200 miles over the pack like the models showed a week ago. We would have seen much more compaction.

This has been the story all summer.

If I had to give grades in terms of winds for ice loss, I'd give it a D+. 2007 had an A.

But yet we are right there. We won't see refreeze even with widespread -6 to -12C 850s with so many SLPs moving 20-30kt winds all over the place.

we have to take this day by day now.

Only use 3 day forecasts and 1-2 day ice outlooks with the wild variables like Hurricanes, surface feature placement, and temps.

Janne Tuukkanen

D: I was thinking about Irene as well. The remnants of it are quite a cyclone marching over North Atlantic to Europe.

More (water) precipitation and warm air to Greenland Sea? High winds definitely.


Well, Espen, it seems to me we’re on the same track. You Photoshop, my CAD shows 794K less high concentration ice than in 2010 (25 august). Using the AWI and Polarstern reports (I’m no politician, Will Crump) that difference is good for 800 km³ less in volume. I don’t compare to 2007 anymore.
There is less than 30% volume left comparing to the 1979-2000 mean. This year, excess warmth in the open Arctic waters looks more than ever before.
The ice is like the canary in the coalmine. But it is severe weather within changing climate that worry me most for the near future.

Seke Rob

And in other news following the announcement of 6 icebreakers under construction, a deal struck to explore, no develop several blocks in the Kara Sea: http://news.yahoo.com/exxon-rosneft-tie-russian-arctic-u-151024298.html

Lucia (The Blackboard)

Thanks. I knew the monthly existed. Paul Klemencic referred to daily data, but was hoping if daily NSICD data existed back to 1979 (or just more than 4 months), I'd learn of it and find it. If it was available, it would be interesting to look at and compare to other data sets.

Patrice Monroe Pustavrh

Regarding the Janne's comment and Irene. Irene is now extratropical cyclone (cold core) and not tropical anymore (warm core). Are there differences (and how big) regarding that there is still a lot of energy there ? Any good readings on this ? Thx a lot to those who know better than me.

Christoffer Ladstein

As I "predicted" less than 2 days ago, Healy and Polarstern have just met and "winked" to each other. Must have been great fun, exchanging sauerkraut with hamburgers!

What puzzles me though is the fact that these 2 boats report of different seatemp from the SAME sites!


And Healy...


As you are able to verify, the differnce between them are almost 0,5 C in average, so which numbers are most reliable?
Do they measure on different depths?

To me AND the ice, this matters a lot, as Dodger so frequently have stated: ice is created at -1,7 +-0,2 depending on salinity!

Seke Rob

Re: Lucia (The Blackboard) | August 30, 2011 at 22:36

Got extent and area dailies from covering 1978-2007.


Not looked, given the gsfc reference in the file names and the other set floating around of gsfc covering 1972-2002, for similarities / how they differ.


Back on Irene I'm guessing its going to accelerate surface water north into the widening gap between greenland and svalbard which may 'encourage' a shift in the major currents.


Patrice, I just read a post on Davidson’s ‘eh2r.blogspot.com’ site. His sundisk measurements over Montreal may give a clue on the energy redistribution in the atmosphere due to cyclones.

Michael Fliss

Perhaps Christoffer, you are looking at the CCGS Louis S. St. Laurent and the Healy Passing? They both have sea temps of -1.6 at 1500 hours, 8-30-2011 and share approx. coord.

Janne Tuukkanen

There's also Katia, which will probably develop to cat-3 or above, but hopefully miss a landfall. It will transport considerable amounts of energy to North Atlantic.

It sounds like a long shot to connect something born near Cap Verde to Arctic ice, but these systems bring winds with them.

Al Ramsey

Christoffer - the ship you see in the Healy webcam is the Canadian ship - Saint Laurient (red with white stripes) The Polarstern is about 100+ Km away - if you look at the coordinates you will see the different locations - which would explain the different water temps.

Neven - Many thanks for the you web site - have been lurking for a long time and my wife thinks I spend far to much time quoting from it! It has been an amazing summer melt!


Don't remnants of Atlantic hurricanes generally travel rapidly east towards UK or maybe a little north of that. Isn't that too far away for significant effects on arctic ice?

Steve Bloom

Neven: "The IPCC is saying 2100."

More to the point, they said that ~5 years ago and it's already clear that in the AR5 they'll be saying something quite different.

Seke Rob

Re: Al Ramsey | August 31, 2011 at 00:12

The barometric pressure was too different, 6 points, for that to have been right. Mind you, both were at 86.06N, but on different longitudes. Smells like these guys are on a highly coordinated mission.

Lucia (The Blackboard)


Not looked, given the gsfc reference in the file names and the other set floating around of gsfc covering 1972-2002, for similarities / how they differ.

I've been grabbing these:
http://polynya.gsfc.nasa.gov/datasets/nh_daily_observed_or_interpolated_sie_1972_2002.txt They end in 2002.

I've got a note in the file with my R script. I'll line these up and compare. I checked 1 specific day, they are a little different!

The one you found gives me more overlap with JAXA which is nice.

Paul Klemencic

I thought the longer term NSIDC numbers were in the FTP files at the U.Colorado site, but can no longer find them. Sorry.

Seke and others have pointed you to the reports up to 2007. Hope you can make something out of that.

Andrew Xnn

Here is what the IPCC actually reported in Chapter 10 page 750 of AR4, The Physical Science Basis:

The projected reduction is accelerated in the Arctic, where some models project summer sea ice cover to disappear entirely in the high-emission A2 scenario in the latter part of the 21st century.

Looking over the rest of the report, it's obvious that there are a lot of models with varying findings under different scenarios.
Review page 770 for details.


Andrew Xnn

Here is a diagram from the IPCC.
Negative 8.2 would correspond to an ice free arctic summer and as can be seen, it's only some models in some scenarios.

Not very impressive and IMO very conservative.

Pete Dunkelberg

Janne Tuukkanen: "It sounds like a long shot to connect something born near Cap Verde to Arctic ice, but these systems bring winds with them."

Some surprising corelations turn up. The Naval Graduate School thesis of Megan M. Stone, LONG RANGE FORECASTING OF ARCTIC SEA ICE (pdf) is an example.


Our results indicate that viable long-range forecasts of October SIC in the Beaufort Sea are possible via the use of Beaufort Sea SIC and Caribbean SSTs at lead times of one to five months. While our results show a definite correlation between these two variables and October SIC in the Beaufort Sea, we suspect that there are additional factors and dynamics that play an important role in the variability of SIC in the Beaufort Sea. Our study is meant to emphasize how advanced data sets and methods can be used to generate skillful long-range forecasts of sea ice amounts in the Arctic for use in operational planning by the U.S. Navy.

Andrew Xnn

For 2010, JAS Sea Ice Extent averaged 6.3 Mkm2.
That corresponds to an anomaly of -1.9, which is out of bounds for all IPCC models and scenarios.

Paul Klemencic

I tried to deconstruct the JAXA-IJIS reported ice extents to get single day numbers. I picked two sequential days this August, that reported almost the same percent decline in the supposedly two day average , to algebraically calculate the 2nd and 3rd day extent.
If x = the first day extent
then x (1-r1) = the second day extent
and x (1-r1) (1-r2) = the third day extent

Then the average of the second and third days must equal the reported 2d average.

Reported 2d average extent = {x (1-r1) + x (1-r1)(1-r2)]/2

I used initially August 19 with a decline of 37657 and August 20 with a decline of 37187, and calculated the initial extent on
August 18 was 5391518, with
August 19 at 5353994 and
August 20 at 5316944.

Then the prior and following daily extents were calculated using the 2d average formula.

Although the decline in the 2d average for August 19 was 37.5k versus the reported 37.6k and for August 20 the decline was 37.1k versus the reported 37.2k, the daily extents quickly exhibited irrational swings on either side of the test period. Within a few days, the drops in the calculated 1d extent would be either very small or often negative one day, with the next very large, and this pattern repeated on either side of the test period.

Clearly the JAXA series of extent numbers reported cannot be simple averages of the two days of extent measurements.

I repeated the same procedure for July 29 that reported a drop of 23.3k and July 30 that reported a drop of 23.0k, and got the same results.

JAXA-IJIS is using some kind of algorithm to dampen large extent loss days, and 'augment' low extent loss days. Use of either an underlying trend, or longer term average in the algorithm, or some sort accumulated over/under account seems likely. I like the last idea; sort of a "slush fund" for ice extent reporting.

In any case, I was unable to construct a rational set of 1d extent numbers, that would calculate to the set of rolling 2d averages reported by JAXA-IJIS. I really don't know what those reported numbers are, and how to relate them to observations of the ice pack.

Michael Fliss

Seke Rob and Christoffer: The Healy and St-Laurent are working together mapping for the Extended Continental Shelf Project until the end of September.


Daniel Bailey

@ Paul:

I checked with one of the NCAR researchers I know. He's checking with the head of their data support division. However, that individual happens to be away. In the meantime, my contact suggested two options:

1. Try this link:

2. To check with John Walsh, former head of the U of Illinois and now head of U of Alaska. I'll be happy to check with Dr. Walsh for you if you can give me the specific parameters of the datasets you want.

Michael Fliss

...and not to beat a dead horse...or....a picture is worth a thousand words, check out the Healy aloftcon webcam at 1600 and 1900 hours.


Rob Dekker

Note on ice thickness/volume :

* The PolarStern reported 0.9 cm ice all around the Pole.
* The Healy is steaming 3kts all from open ocean to the pole, suggesting ice thinner than 1 meter average pretty much all the way (except for a few patches (2 so far) of thicker MYI, where it needs to cruise around).
* The Beaufort IMBs report 0.6 - 0.8 meter left over.
* The Siberian side of the pack looks like it is in similar shape as the Northern Beaufort.

Going out on a limb here, but this seems to suggests that ice across much of the pack is now on average 0.8 - 0.9 meter thick.

With sea ice area just at 3 million km^2, this suggests 2400-2700 km^3 left over.

With very little MYI that could seriously increase that number, and a month of bottom-melting to go, this starts hinting at yet another devastating blow to ice volume, even w.r.t last year's 4000 km^3 already record low...

I wonder what PIOMAS will report.

Daniel Bailey

Looking at the ice the ships are cutting through, it looks for all the world like a skin of curdled milk at the top of the bowl...

Very thin stuff.


-93,000K loss tonight on Jaxa.

Definitely helped by the big day yesterday. But still another big drop.


Prelim 4803438 a drop of 93125.


And if the correction is small tomorrow, this is already a new third place:
09,18,2010,4813594 was the MINIMUM for 2010

09,09,2008,4707813 was the MINIMUM for 2008
09,24,2007,4254531 was the MINIMUM for 2007

And we have another 2 or 3 weeks to go.


Pending a possible revision, 2011 is now in third position for extent behind 2007 and 2008, about 10 k lower than the 2010 minimum

Minimum extent in 2008 was about 100k lower, which means that bar a very early end of the melting season, there is a strong possibility that 2011 will rank second for minimum extent.



Beat me to it :-)


I'm sure you'll beat me to it when it passes 2008 in a few days :-)


Re "bar a very early end of the melting season"

or a very late and dramatic end to get a record minimum. Loss to minimum would need to be over 50% more than 2007 so such a dramatic and late end looks very unlikely.

Three days averaging 60k drops would reduce that 51% down to just a couple of percent but such a strong run this late is unlikely even/especially? after a 93k drop.

Steve Bloom

Maybe, Chris, but unless my memory of the numbers for the last few days is wrong 93K today means that it's the second day in a row of close to century breaks, plus 2011 has a bunch more vulnerable ice than 2007 did on the same date. Then there are the SSTs. Weather may not be cooperating as much, but as several folks have noted maybe it doesn't need to.

Steve Bloom

If 2011 just continues its average of the last week over the next three days, it'll pull within 50K of 2007. That doesn't sound so unlikely, and after that it seems to me given the various factors I listed that 2011 would if anything be the favorite. But we shall see very soon.


I just had one of those elephant in the room moments.

Michael Fliss said look at the 1900hrs Healy picture, so I did and really looked at it.

been looking at the pictures for several days and observed how thin looking the ice is ( it is a rare picture that shows any ice with large amounts of 3D ice it is all flat and low to the water) and even where the ship is now perhaps 50% being melt ponds. but that 1900hrs picture showing the passage of the other ship through the ice gave me a wake up call.
look at that wake! (using the word wake to describe the channel cut by the ship through the ice. the ship has turned the pack literally into slushpuppy. Biggest piece in sight is less than 6ft square and most of it is just mush. that ice must be pancake thin. not even close to 0.9 metres. or does a ship going through ice make mush like that when travelling through thick hard ice? not seen such pictures before for comparison

Steve Bloom

Just so, Pof. It seems to be pancake ice almost everywhere, with the multi-year having suffered very badly.


Crandles – From what I’ve picked up on the subject, cyclones have a wide ranging outflow high up in the troposphere. Their influence (transfer of heat) reaches much further than the relatively small spinning cloud formation near the Earth’s surface. On the 250mb pressure height that could very well influence the state of the polar front and the polar jet stream. They might play a role in blocking events and, as an example, the exceptional warmth over the Baffin region last autumn. I’d love to learn more on this, the work of Kerry Emanuel seems to stand out on the subject.
Would the effect of cyclones be contributing more to sea ice loss then before? I have no clue on the heat transfer through individual cyclones. But it wouldn’t surprise me that has accelerated under a general pattern of forcing. It means we’re lucky to see them go fishing without striking land, but each one does it’s work, destabilizing ‘normal’ weather patterns.

Mike Constable

The Healy cam does actually show some 3-d ice on the image at http://mgds.ldeo.columbia.edu/healy/reports/aloftcon 20110830-2001.jpeg which contrasts with the no-profile slush in so many other pictures. Looks thick, but I may be grasping at thin ice (to mix my metaphors)


" D: I was thinking about Irene as well. The remnants of it are quite a cyclone marching over North Atlantic to Europe."
maybe this is also relevant here? from 2005/6 when a swell shattered a part of B15:

Alaskan storm cracks giant iceberg to pieces in faraway Antarctica:
Study tracks surprising connection between distant events

Clare, back to lurking & learning lots, thanks Neven, thanks to you all


Thanks to everyone thanking me!

"The IPCC is saying 2100."

More to the point, they said that ~5 years ago and it's already clear that in the AR5 they'll be saying something quite different.

Steve, I know. It just bears repeating, as some people are very eager to turn Maslowski in a strawman that can be used by people to not have to change their lifestyles when some of the ice is still there in 2016.

The IPCC says 2100 and so should be disbanded. ;-)


I asked Anthony Watts to be accountable for his 5,750,000km2 prediction and he deleted my post.

I asked him how he could have possibly predicted that with the data and listed the data we had for him. Good Grief.

Neven, feel free to delete my post if this bothers you. I am just so mad right now.



Water encroachment.

lots of warm water still flooding into the Arctic.

Ned Ward

some people are very eager to turn Maslowski in a strawman that can be used by people to not have to change their lifestyles when some of the ice is still there in 2016.

It's worse than that -- Maslowski said "2016 plus or minus three years" so naturally that got turned into "2013". I've seen many, many claims in the denialshphere that climate scientists have predicted the Arctic ocean would be ice-free in 2013. That's just dishonest.

As ice continues to decline, you're also seeing people either quibbling or being deliberately obtuse about the wording of that prediction. No, a prediction of an ice-free Arctic ocean in the summer does not mean an Arctic ocean with zero grams of ice every day from the summer solstice to the fall equinox. Sheesh.


Frivolousz21 No you are not deleted I just read your comments.
Regards espen


I know, Frivolousz21, it makes me mad too. But we have to try and not get mad, because they feed on that.

Nick Barnes

I'm back from my work trip to NYC (First International Workshop on Climate Informatics, very interesting), where I sat out the hurricane in a friend's apartment. Catching up here, and a bit boggled by the last week of IJIS numbers. I haven't much to say, yet.
However I *will* mention a book which I took on my travels, and which Everyone Should Read, or everyone interested in sea ice, anyway. It's even on-topic with these ice-breaking discussions. "South", by Ernest Shackleton, his own memoir of the Endurance expedition. I knew the story already (who doesn't) but to have it told in his own words, and illustrated by Hurley's famous photographs, is fantastic. My only caveat would be: see if you can find an edition with maps.
I have a feeling that the text may be in the public domain (one reason for this suspicion is that there's no copyright statement on the imprints page), so any of you who like reading books on screens can probably find it for free online.


the guy tried to spin this:

Don’t sweat the ice area statistics. The thickness is much greater today, and we could even say the volume is likely more. Arctic temperatures above 80°N have been colder this summer and September. The ice area will rebound quickly, of course. I projected a 5.75 million sq km min. for 2011 a couple weeks back. I’m sticking to it.

He said that here:


SO he actually said it twice.

Now here:


I guess he forgot his grandiose prediction to feed the heart and souls of his cronies.

I posted this on there, lets see if he publishes it.

He also lied in his most recent post. If you take the NSIDC chart and throw it in photoshop you can see its between 4.8-4.9km2.

Not around 5km2 and that was the 29th on a 5 day mean not including the todays large drop.

and his followers are saying we hit bottom right now at 4.8-4.9.

Does he have any integrity? How can he be like this and sleep at night. He has to be aware of this horrible manipulation.


I hope you guys read his reply to me.

It was priceless and complete garbage..I bet you his followers believe him.


Frivolousz21, try not to link too much to those places and take it to the How thick is your ice? post. :-)

Ned Ward

My comment got eaten, alas. Let's see if I can reconstruct it:

-93,000K loss tonight on Jaxa.

Whee! That's a huge drop, this late in the season. If the final number is revised downward by 4K, it will be a new record drop for this late (there was a 96K drop on 15 Sept 2003).

That 2003 drop was itself something of an outlier -- the next largest one was only 72K.

Here are the other large drops in late-summer IJIS data:

2003 09 15: -96719
2008 09 01: -72188
2008 09 07: -68437
2010 09 07: -66093
2005 08 30: -65781

Stuart Preen

Quite a large negative on the Antarctic sea ice as well at the moment taking the global value well past the -2M level. Perhaps an early end to the winter freeze in the southern ocean?


what was last years Fiasco?

I apologized to him. I have no problems doing that. I don;t have to get over on him for him to know how he really is. He knows.


CT area up to 3.129 higher than each of the last 6 days.

Daniel Bailey

@ Stuart:

"Quite a large negative on the Antarctic sea ice as well at the moment taking the global value well past the -2M level."

Likely the warming oceanic waters surrounding Antarctica taking their toll. This is discussed here:

Lord Soth

Just when I was ready to throw in the towel, we get a 93K melt.

But the next four days will be do or die for a new minimun.

For the next four days 2007 stalled and only recorded a 37K lost. If 2011 can record an average of 55K over the next four days we will be in a tie.

Even if we get within 50-80K, we may still be in the running for a new minimun.

If 2011 stalls also, then its game over.

It will require a complete stall, and an early minimun for 2011 to achive any worse than second place.

Does anybody have the raw NSIDC numbers (non-averaged), as last night may have pushed 2011 in first place for NSIDC SIE?

Seke Rob

No selling the bear skin before shooting it. CT was early today, and it shows a little rebound on the Arctic [see earlier posts] and a bigger on Antarctica. [See Chart]. Just added a dotted lead line to the global CT SIA data based graph, so it's easy to see if above or below, for the day of the year (follow red 2011 line). It's again smack on the 2010 lowest point for the 4th or 5th time since June ;|

Pretty plz, no more Anthony (lobotomo) Watts. The man as the Germans say "Fehlt ein paar Tassen vom Schrank". He met with Nurse Ratched and since flies cookoo nests.


Does anybody believe that area reached a minimum of 2,983,597 sqkm on Day 237 (August 25)? We are now 145k above that value.

If it is the case that we have passed the area minimum, the minimum extent should not be far off : an early end to the season then!


Phil, FWIW: I don't. 2007 also was up 179K around this time from a first nosedive, but went down some more in the end. Besides, SIA usually follows SIE and the latter almost lost a century today.

I expect SIA to hit its minimum between September 7th and 9th (just like the last 4 years).


Phil – A physical explanation I imagine is refreeze of (small) polynia’s inside the high concentration pack. In a lot of grid cells the actual percentage is on the rise again. But on the periphery 93K worth of gridcells have dropped below the 15% norm.
Looking on MODIS, it seems like nilas is forming in some 50-60 km² polynia’s. Also, while the Healy pics illustrate the actual thickness of the pack, they reflect the thin ice on the meltponds too (if you can call them like that; they’re small but a lot seem to go right through).

Paul Klemencic

No surprise anymore to me, that the JAXA-IJIS reported number shows a big decline (93k) on a day when the Bremen map shows the extent falling about half that fast. The day before, the Bremen map showed very large extent losses, the biggest since the "flash melt" day (although nowhere near that day's decline). From looking closely at the reported JAXA-IJIS numbers, compared to observed changes on the Bremen map, it appears that IJIS takes several days to catch up after a big decline, or after a series of observed weak melt days.

If there are delays in getting good confirmed measurements, coupled with a two day averaging procedure, then the behavior of the IJIS reported data may be explainable, particularly after big changes in the ice pack.

My estimate of ice extent is now only 120k less than the IJIS reported extent, whereas a week ago the difference was about 220k. August 26 and August 27 had very little extent losses; even though the ice was weakening, the edges held their ground slowing ice extent declines. With JAXA reporting losses of 52k on August 23, and August 26, two very slow loss days, coupled with the big report yesterday, now the extent reports are beginning to converge.

I still think that its better than 50/50 that the ice extent falls below the 2007 minimum, although the weather is really going to determine this. To beat 2007, we need to lose about 470k from my current estimate of extent.

The conditions the last three days were very conducive to moving ice over warmer water, in the Laptev and Kara regions, and to a lesser extent, the E. Siberian. Couple this with the ice pack receding above Svalbard and Franz Josef Land, and some extent loss in the Chukchi region, led to the big loss on the 29th.

One of the most notable weaknesses that showed up, is the weakening ice area at 170E to 180E at around 81N to 82N. This weak area started several hundred km to the left of where it currently sits on the map, in the Chukchi region. Given the shearing impact of the weather system in that area, it opened up again. It likely shows warm Chukchi region water is being pushed into the pack at this location. There is now a weak low sitting over this area, slowing new inflow, but the heat that already moved into the area may result in continued weakening in this area.

I don't know how the weather system changes showing up today will affect the pack... I guess we just wait and see.


Only 312 square km short of a century break after revision today. Amazing.

Lord Soth

IJIS just updated with 99688 lost on the final update for the 30th

312 sq km short of a century break. They should go back and find a few more questionable pixels for Nevens sake :)


Indeed, LS. I demand a second revision!

Seke Rob

Who wanted a century... will be disappointed. JAXA after final revision -99,688 KmSq for the 29th to the 30th, a change down from prelim by -6,563 KmSq.

Ned Ward

Indeed, LS. I demand a second revision!

Hey, be careful what you ask for. They might revise it back the other direction.

At least this way, we got a new record for the largest single-day decline this late in the season (see my comment up-thread).

There was one single-day decline of 96K back in Sept. 2003. Aside from today's datum, and that 2003 one, there have been no declines larger than 72K in the IJIS record this late in the season.

Seke Rob

P.S. Slammed the 2011 door in the face of 2010 some 19 days earlier, which reached minimum on Sept.18.

The daily numbers to beat for the coming 7 days on 2007, considering 2008 a near formality ;>)

31st -9063
1st + 3907
2nd + 6093
3rd -37031
4th -51875
5th -43594
6th -37500

Pity that post preview does not show any new posts... need to type much faster :)


Hey, be careful what you ask for. They might revise it back the other direction.

I'll take my chances, Ned! ;-)


As recently mentioned on the End Zone thread, although this year trails the 2007 Jaxa SIE by 180,000 sq km, the year-to-date figures (up to and including 30th Aug) are the other way round.

This year's y-t-d is currently 11.029 million sq km, with 2007 at 11.084. Obviously the gap between the 2007 and 2011 varies with the actual daily figures and the number of days these are being averaged over. At present, these two y-t-d figures are converging at a rate of about 1,000 sq km per day.

What some may not have realised yet was the rather odd behaviour of the Arctic SIE during October 2007 when compared to (say) last year. At the end of September, 2010 was about 972,000 sq km behind, but had closed to 72,000 by the end of October.

However, at around the 17th October, the difference had stretched to 1.4 million sq km. The effect this had on the yearly average should be pretty obvious.

Therefore even if 2011 finishes comfortably behind 2007 in terms of the absolute minima, the annual average SIE figures are still very much up in the air.


312 sq km or exactly 2 pixels!


L. Hamilton

On 8/29 the Uni Bremen extent hit 4.6, by a trivial margin lower than the 1-day minimum for 2008.

Christoffer Ladstein

I almost spilled my coffee, when I first encountered todays IJIS numbers!

Who cares for crime-fiction, when we have daily access to Nevens-realtime-"fiction":-).

Ned Ward

312 sq km or exactly 2 pixels!

When the prelim number came in, somebody should have called up the Healy and told them to steam 'round in circles for a few hours, smashing up as much ice as possible... maybe Neven would have gotten his century break.

The comments to this entry are closed.