« A Farewell to Ar... ctic Sea Ice Blog | Main | Open Thread 2 »


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

Artful Dodger

First question of course is "what is the existing record?" The graph above is the cumulative anomaly for both N. and S. hemispheres. Personally, I don't like to make this comparison because it implies there is some connection between Arctic Sea Ice and the Antarctic. There is not. And, it is often seized by disinformers spouting the false 'global sea ice balance' meme.

The more relevant and timely statistic is the record low N. Hemisphere Sea Ice Area Anomaly (SIAA) as reported by Cryosphere Today. It was -2.635 million sq.km on Oct 20, 2007. The current N.H. anomaly is -1.610 M km^2. Going forward, a further drop in the SIAA would indicate slow sea ice growth this fall.

Gas Glo

Beauford Gyre and Transpolar drift look like they may have gone into reverse again.

Does this, during freeze up, have effect such as less compaction and therefore the older ice might be less resistant to melting next year?


I think we are still about 100-125 thousand sq km's above the maximum (negative) anomaly from 2007.

But if the freeze up is slow, Antarctic melt could still push it to a new record low. (Acknowledging Lodgers entirely correct "false equivalence" observation.)

Artful Dodger

Gas Glo: Yes, absolutely. If the main Pack freezes in place, come Spring there will be vast areas of new, saline ice interspersed with sparse fragments of multi-year ice. And that's just the beginning. It looks like a very active ice advection season is shaping up N. of Greenland. There, all multi-year ice lost is replaced with more new, salty sea ice. Not good for the pack...


Personally, I don't like to make this comparison because it implies there is some connection between Arctic Sea Ice and the Antarctic. There is not. And, it is often seized by disinformers spouting the false 'global sea ice balance' meme.

Sure, but a record is a record, and this record would be very effective in killing off that pseudo-skeptic meme. For reasonable people that is.

Does this, during freeze up, have effect such as less compaction and therefore the older ice might be less resistant to melting next year?

That's a very good question, Gas Glo. My eye is currently on the DMI 80N temp graph. I mentioned in one of the SIE updates that a lesson I learned in previous years is that when you see big spikes in the trend line (after minimum extent has been reached) this means that the warm waters are releasing their heat to the atmosphere so that freeze-up can take place. I'm not seeing any spikes yet. 2009 was also late.

So I think there isn't a lot of compaction going on at the moment and the freeze-up is very regular. Maybe the warm waters aren't releasing their heat all at once. If this is the case we might soon see a slowdown in extent increases. But this is all speculation.

michael sweet

There are reports that the Beauford sea is 5C higher than normal. Does anyone here know how deep the heat goes there (and north of Siberia)? Obviously if the heat goes 200 meters it is more heat than if it only goes 20 meters down.

As the freeze up starts ice is filling in near existing ice flows, like in the polar hole Nevin talked about earlier. The water there should be close to freezing since ice is nearby. Once those areas are frozen the freeze up will move into currently open water. Some of the open water is a lot warmer. How long will it take for that warm water to cool? It will be interesting for the next few weeks.

The Northwest passage is still open, but ice is starting to form in it. All the vessels trying to make it through this year must have made it.


Artful Dodger

Thanks for the tip on DA and Wang's paper.

Do you know where I can get the monthly DA data Wang shows in his paper? His plot goes from 1950 through 2008. It would be interesting to updata it for the 2010 melt season and track it in 2011 season.

Artful Dodger

Michael: Read Woodgate, Weingartner, and Lindsay (2010) "The 2007 Bering Strait Oceanic Heat Flux and anomalous Arctic Sea-ice Retreat" (web copy here, pdf download here).

In the Bering Strait, the Surface layer is approx. 100 - 150 M deep (NOAA operates a Buoy Array there, referred to in the paper above). In other areas, the depth of the Surface layer changes dramatically once salinity falls below about 30 psu. In areas such as the Mackenzie Delta and North of Siberia, river outflows freshen the surface water above the 30 psu threshold and then the Surface layer may be only a few meters deep.

Keep in mind, it is not only the current SST in the Bering Strait / Chukchi Sea that matters. The timing of freezeup is strongly influenced by pacific ocean volume and heat transports North through the Bering Strait. Last year, the Southern Chukchi Sea was still open in early January, with air temperatures of +4 C.

Weekly Salinity and Sea Height data is available here, and gives a good indication of volume and heat transport through the Bering Strait.


Extent is shooting though the roof, with a very early 100+K reverse century break. If things were refreezing so fast I would expect the waters giving off a lot more heat (and the waters are still pretty warm according to EORC/JAXA and NOAA), but the DMI 80N temp graph is only showing what could be the start of an uptick.

So is this all due to those big lows, reversing the BG and TDS and spreading out the ice in all directions? As we can see from Jack Taylor's animation the ice is advecting southwards pretty fast, and the Arm is also being pulled in as it were by the ice pack (actually, it's the ice pack pulling towards the Arm).

Things remain interesting. Well, to me at least. :-)


"the Arm is also being pulled in as it were by the ice pack (actually, it's the ice pack pulling towards the Arm)"

NSIDC calls it oscillation.

Did another type of animation of it using the UNI-Bremen "visual" set and posted it between the graphs on my chart page. http://www.polk-nc.com/arctic-tenXten.html

BTW, Neven, I "Laid an Egg" over your North Hole on it to reduce distraction
so the edges are more evident.

Artful Dodger

Nice picture of Akutan Island from the Healy's Aloftcam :


"Extinction its not just the polar bears"


I'll watch the CT "Arctic Tale of the tape" graph showing the anomalies' evolution in recent years.
- no positive anomaly occured since march-april 2003 (or Nov 2004 if you're strict)
- the anomalies amplitude is much larger since 2007, around 2 million km between tops and lows, instead of about 1 million the previous years.

If anomalies stay under 0, I'll read this as "we're still in this big anomaly phenomenon - and it's not recovering in any way".

I analyse the large amplitude as "the system went out of equilibrium, and still is out". And : a big ice sheet, with high volume, had a much larger inertia than it has since 2007.

Kevin McKinney

That last exchange on the video dorlomin linked summed it up well:

Interviewer: Do you think the walrus can adapt?
Inuit elder: I hope so.
Interviewer: What if they can't?
Inuit elder: We're in a lot of trouble.

I'm thinking we're ALL in a lot of trouble, myself. It's just not clear yet how soon it kicks in big time.


NSIDC has updated.


I've also added a small update to this post (scroll up). It looks like the water has started to release its heat to the atmosphere.


About the DMI uptick: Wayne from e2hr.com explained some time ago that arctic water starts freezing at -11 or -12 deg. Celsius. DMI shows 264 Kelvin, which is -9.15 C. Quite right...
(and this proves again that 'some' sites are just cherrypicking, as they just tell about 0 on DMI graph)


Fred, do you remember where Wayne said this? I'd like to read that.


You made it onto a blog posting on Skeptical Science.


Indeed, dorlomin. That's nice! Thanks go out to the Yooper and adelady.

And thanks to you too for linking to Peter Sinclair's latest video (I've subscribed to his feed, but somehow I find out about his videos a few days after they've been released).

Lord Soth

According to NSIDC, we miss second place with 2008 by only 37,000 km^2. I would call that a statistical tie, with regard to extent.

I'm still waiting for the official annoucement on the final volume numbers from PIOMAS, although we all know that this year volume is much less than the previous record, perhaps by as much as 1/3 less than the total previous ice volume.

With the Northern Passage crew making short work of Baffin Bay, they should be home in a fortnight.

The only other thing of major interest (to me), is the tracking of Petermann B, and if it stays ahead of the freeze up as it heads south thru Baffin Bay, or if it will be trapped in the advancing ice.

With this, other than the occasional comment or two, I will sign off for the season.

Thank You Neven for creating a blog, and a home for us that has "can I say healthy obsession", with the state of the Arctic ice.

I will see you all, with the first warm rays of spring.


Neven : Wayne wrote about -11:

It's on eh2r.com :

"Sea ice is an expression of –11 C air temperature

~ Close to the Pole and old notion is confirmed

~ NSIDC forgets to mention: thin ice melts fast!

It is known as the breaking point, when Arctic Ocean ice turns from consolidation to a more fluid state. Years of observation has revealed -10 to –15 C common to spring ocean ice break time. In
Other words, when this happens ice plates flow and collide much more frequently, rather difficult to live on, let alone travel. Now scuba divers from a fantastic expedition near the Pole provide some insight(...)

Its been –9 C outside, and the divers saw the ice melt, literally fall apart from the mere touch, fragile, and full of biological activity. This shows, despite, what appears to be no melting on surface, a loss of balance, between the powerful heat in the Ocean and the colder atmosphere, a balance broken from the higher sun.

Back in frequent fall times, I found that sea water freezes when air equals or is colder than –11 C, especially in no winds. The reverse may be so, by conduction air warms the ice, to the point where it sets the bottom from freezing to melting. Thus sea ice. Is an expression of cold air, when gone, so is the ice."


SIE has grown back by a staggering 0.5 million sqkm in less than ten day! I went through Jaxa records and couldn't find any year that came anywhere near this rate of increase after the minimum was passed.

Anybody has any idea why the ice is growing so fast this year. Is it just the weather? or the fact open water in high latitude freeze up very quickly.
Also it seems that we are having a very strong La Nina this year and I was wondering whether this would have an impact on the arctic sea ice.


Thanks, Fred. I must've read that before, but forgotten about it (happens a lot).

Phil263, I think the weather has a lot to do with it. There is some very strong spreading out going on, as can be seen on the PIPS ice displacement maps (and the CAPIE ratio is still the lowest in the last 5 years). Of course, now the leads that open up freeze over. I believe the real freezing starts now with the sea releasing its heat to the atmosphere. The water should be warm, so as soon as atmospheric patterns become less wild, the extent increase should slow down. But according to the forecasts this weather will continue for a while longer.


I asked Paddy Power to list odds 2011 - 2020 on 95% ice arctic. If other people ask the same question, we might be able to get it carried. Then contact the dailies and at least one should be able to make a fluff piece out of it. But once it is on the books, it is 'real' in many people's minds in the sense that money is backing the the odds, whatever they are.

Kevin McKinney

Another speculation about phil's question: I wonder if there's something of a reversal of the effect we saw at the end of the melt season, whereby areas would 'disappear' off the concentration map as they hit 15%. Since the hallmark of this year's season was greatly expanded areas of lower concentration within the pack, as opposed to a nice neat discrete pack edge, melt patterns looked different than in the past; extent would "hang in" at times while area dropped, leading to those low CAPIE ratios that Neven mentioned.

So--perhaps?--the process could work in reverse: ice existing @ < 15% concentration seeds freezeup areas which then "appear" over the 15% horizon on extent graphs. Presumably that would happen faster than if the whole area must freeze from scratch, just by the arithmetic; but I also seem to recall that, for some conditions at least, existing ice can seed crystal formation and thus accelerate freezing rates. (Commentary from the actually knowledgeable solicited!)

It's also interesting to see these rapid extent growth rates occurring again, thinking back to that extent spike in February/March. At the time, NSIDC attributed that to a cold snap, primarily over the Barents and Bering Seas.


Petermann B is doing lazy circles.


Neven, Kevin

Thanks for answering my queries.It will be interesting to see whether this high extent growth rate is sustained in the coming weeks!


At the risk of revealing what an idiot I am, I return with my idea to ask Paddy Power and Ladbrokes to list odds on sea ice disappearance. The email address for requesting odds are below:


This was my reasoning on suggesting odds:

What has been occurring in recent years is that the older ice has been melting. This causes two things: 1) the extent of the sea covered with ice increases. As the ice thins, it breaks into pieces that spread out. Thus extent goes up as stretches the ice, at least to a certain point. After it is stretched too far, it counts as water. On the other hand, the wind can also compact the ice. In general, as the ice thins, the extent measurement goes up. 2) The dynamics of the ice change. As the ice thins melting and freezing both occur more rapidly depending on weather conditions. Water with ice already in it is more ilkely to freeze than open water, for a variety of reasons including temperature, salinity, reduced wave actions, etc.. When the ice freezes it is only thin first year ice that is likely to melt in the summer. Also, the new ice is patchy and contains a lot of trapped areas of ice with high salt concentrations that are likely to melt first and create holes that allow the sun to heat the water and thus melt from the inside out.

In short, in recent years, sea ice has been going down. Extent is likely to recover for another couple of years as it is fed by older multi-year thicker ice. Area in the winter will recover as the spread out thin summer ice (extent) allows rapid freezing. Area in the summer will likely go down quicker since the thin ice melts. Volume is the key variable and it has gone down to some 30% of the minimum of 1979.

My off the cuff predictions are:

2011: 3% possible new records but likely the only change is further eroding of multi-ice.
2012: 10%, same reasoning.
2013: 15%, same reasoning.
2014: 20%, same resaoning.
2015-2020: 35% + 10% per year up to about 85%, same basic reasoning but at some point the odds of the right weather conditions kicking in increase. The overall trend in ice decrease is real (I believe) but weather factors are hugely important. There will be cold winters and cold summers, there will also be hot ones. Just as 2007 had conditions that were perfect for melting, 2010 had a bad June (i.e. good for melting) but a good july (bad for melting.) as the ice goes down, it loses the capacity to protect the core ice. As a simple time multiplier, the overall chances of another bad summer increase by the number of summers in consideration.


I realize I am an amateur. I ask to not be snickered at, using the bog standard swap of enthusiasm for knowledge. I would really be interested in hearing what other people's odds are. Subjective, please! Numbers make my head hurt. Also, please contact the two agencies above and ask them to list odds. I really think that if the odds are listed, it will make a difference in the world of public opinion.

Greg Wellman

I'm assuming those are your odds of a new minimum extent record that year? Because the odds of a new minimum volume extent record are higher :-) Your odds for extent are ... reasonable IMO.

I might quibble with your "thinning leads to greater extent" explanation, although you also said "at least to a certain point" which is the important thing. I think you understand the process better than those few sentences would suggest. I would say it like this: Some of the ice is in a thick, relatively monolithic and immobile phase, and some is thin, loose and mobile. Thinning of ice in either phase without actually changing the relative amounts in each phase will lead to less area. Conversion of ice from the thick, immobile phase to the loose, mobile phase will increase area. 2007 saw great thinning, but not so much of that conversion. That conversion has been happening a lot in the 3 years since. We're now close to the point where there's not much left to convert, so barring unexpectedly cold temperatures, extent loss should resume.

Greg Wellman

er, "volume extent" should be just "volume" in the second sentence in the previous post. I'll fire my proofreader. :-)


SIE recorded another century increase for Sep 28 after a near century the day before.
Increase from the minimum extent to the last day of sept have been as follows for the last 4 years:

2010: + 660 k (till Sep 28)
2009: + 390 k
2008: +502k
2007: + 338k

We can safely assume that 2010 will end up with an increase over 700 k in as little as 12 days which seems considerable!
I also note that SIA (CT) has also gone up by 619 k over the same period.
While you expect freeze up to start in earnest at this time of year, the speed at which it is happening is quite amazing. Something really strange is going on up there !


Yep, CT area and IJIS extent are shooting through the roof.

Of course, 2007 and 2008 also show pretty steep climbs, but not so early. The DMI 80N temp graph is showing a nice first peak. The AO index has finally turned positive again, with lows projected to continue dominating for the next 10 days. It's very interesting to watch.

Artful Dodger

The Parable of the Match:

I went to a football match once with one of my mates. He brought his shiny new binoculars, and spent the entire match focused intently on the scoreboard. I asked, "what are you doing", and he replied "I don't want to miss anything!"

Every time the crowd roared he'd say "What happened? I don't see anything". But when the first Score registered on the tally board, he shouted "GOAALLLL!!" and celebrated with the cheering Crowd. After the game, he said "Wow, that was really exciting. Let's buy Season Tickets!"

For years, we sat on the sidelines. The crowd would roar occasionally, and my friend would peer intently at the tally board. Most games would end with a goal or two, some nil-nil. Eventually my friend's interest began to wane, saying "this game is too boring. Nothing much ever happens!"

But my friend was a loyal fan. He occupied the same seat for many years. Eventually, the football pitch was bought out by a large multi-national company. So now the grass is gone, replaced by bitumen for a Mall parking lot.

My friend still likes to reminiscence about the years he watched football. Occasionally he'll park his car there and stare at where the tally board used to be, and whisper under his breath "gooaalll...". Then, Mall security comes to run him off, "Nothing to see here, move along, move along."

So when you take a friend to the Game, do you show them the Score, or the Match?


Scores are imperfect indicators, they only reflect results which are defined by pre-set criteria, which are necessarily reductive. Matches are the real story. All agreed!
However, matches are real life stories that will be perceived differently by different spectators. Spectators can relate their own version of the match, but different spectators will have focussed on different aspects of the story or may have their own different interpretations of the same event. The one thing that cannot be disputed is the score. One may argue about whether the score should be computed the way it is or whether another type of score would be more appropriate, but if the determination process is transparent no one will argue about the actual score.
The same could be said about the eternal debate around the best research methodology in social sciences ( and even natural sciences): quantitative statistical research versus qualtitative research. Only qualitative research is able capture the complexity of human societies and human transactions, but only statistical anlysis can deliver hard evidence which no one will dispute if the analysis is rigorous. The two are complementary.
Going back to observations about sea ice growth, I realise that measures such as SIE and SIA are imperfect and do not tell the whole story about what is really happening in the arctic.. However IMO they are still valid signals inasmuch as they lead to questions which suggest more investigation. They cannot be simply dismissed as irrelevant. The fact that SIE has grown so quickly this year is in itself part of the whole story, Information that should be placed within the context of other observations, sure, but still a valid (and important) piece of information.


Ever watch Barca/Spain on one of their possession style matches? Personally I can't stand it, if that's all football was, it wouldn't interest me. If, however, you shift your focus to finding instances of technique, well Barca and Spain are your teams. No doubt that the arctic processes are an incredible thing and worth the study of a lifetime. However, in the last 5 minutes of the match: Get.the.ball.in.the.net., I don't care if it bounces off the back of a player's head unseen, hits a late returning coconut carrying African swallow and barely crosses the line... (or hits a beach ball for all the Liverpool fans out there.) As long as my team wins.

Same point for the arctic, this is the last 5 minutes of the game, although we are rooting for the defense, ...er I guess.


although we are rooting for the defense, ...er I guess.

I'll be rooting for the defense, as soon as everyone agrees there's a canary on goal. And I don't mean Julio Cesar or Heurelho Gomes. The goal was scored for this year anyway, now we're watching the preparations for next year's match.

That the ice is growing back so fast is interesting to watch, but not so meaningful in itself. The ice will grow back fast in winter when the Arctic is largely ice-free during summer as well.

We can speculate all we want and I think that with a lot of things the group here has been on the right track (regardless of confirmation bias), but we cannot conclude anything until the ref blows his CryoSat-2 whistle. That's when we and the rest of the world will know whether there's a canary on goal or not.

Kevin McKinney

Well said, Neven. The rapid freezeup is not, IMO, all that surprising, for the reasons we already discussed above.

Interesting, yes. Certain to bring out claims of "recovery," yes. But I'm pretty sure I made more than one casual prediction that we'd see rapid extent growth this year. And if you look at the IJIS graph, though the slope is steeper than usual, the line still points pretty much at the "variability constriction zone" of November and December.

So I expect we'll see these rates come down, probably sooner rather than later.

Gas Glo

Rooting for defence, match and canary analogies.

I don't think either analogy is right. The canary in the coal mine is something considered to be of little consequence that is worth sacrificing to get warning time.

There is plenty of warning in IPCC reports. I fear loss of arctic ice could change weather patterns and could therefore be something consequential rather than 'considered to be of little consequence that is worth sacrificing to get warning time'.

Hence I think we should be rooting for defence rather than hoping it is a canary that serves as as a wake up call. But rooting for defence isn't the prime aim, gaining understanding of what is happening is more important.

Should (are?) there be models being run that set the ice thickness in April to very thin to see what extreme event become more common?

If we gain an understanding that the loss of summer arctic ice will be rapid then I suppose we should be rooting for this to act as a wake up call and for the consequences to be mild. However I doubt it will be much of a wake up call unless the consequences are severe and I doubt we should be rooting for that as there will be little that can be done even with a wake up call prompting more action.

Maybe I am being too alarmist and pessimistic about the situation but neither analogy seems right to me.


Certain to bring out claims of "recovery," yes.

In fact, I have a hunch this is a bit of a Pyrrhic victory. If you want recovery, you'd want a lot of compaction and not the spreading we are witnessing right now.

I'm not really sure if there is a difference in rapid freeze-up (lots of spreading out with leads freezing up with thin ice) or slow freeze-up (with lots of compaction leading to thicker ice) for sea surface temperatures. Does water somehow stay 'warmer' during quick freeze-up because the thin ice layer prevents it from fully releasing its heat to the atmosphere? Sounds a bit illogical.

That's complicated stuff I know nothing about, but I'm sure I'll learn something along the road this winter.


My amateur opinion on the Pyrrhic-ness of the regrowth depends on the temperatures of the water: I imagine that the very top levels of the waters could freeze but they might act as insulation to trap deeper heat in the water, sort of like how hot coffee, as it cools, can from a thin crust of ice even though, as a whole, the contents of the mug are above freezing. In this scenario, without precipitation adding from the top, the ice would never develop much thickness. The alternative is to imagine that the arctic has suddenly shifted from strong melting conditions to strong freeezing conditions over wide areas. I just don't believe there has been the change in temperatures required to do that. However, all of my ideas are firmly unbacked up by numbers and my intuition still gets me lost in parking lots, so if anyone is knows better, I'd be really glad to be corrected.

Gas Glo

I doubt what is happening now will have much effect. If the ice is thin at the end of the year, this is less insulation allowing more heat to escape and more ice to form.

Nevertheless, the ice this year appears as if not only it is thinner than ever before but perhaps also broken into smaller pieces. Both of these mean much less mass and therefore less momentum and kinetic energy. I would suggest that thin flakes of ice are less likely to be broken up and turned over onto other flakes. Particularly so if ice is spreading away from archipelago rather than compacting towards archipelago.

Doubt that has much effect on rapid increase in extent though.

Is there a significant temperature range (/heat content/distribution) of water where it will neither freeze nor melt ice? If there are significant areas where this is true then thin layer of ice being blown to such areas while uncovering areas where ice would form then this could account for rapid increase in extent. Perhaps we haven't had a Beauford Gyre and transpolar drift reversal at this time of year before?

Just throwing some thoughts out there for any comments they might provoke.


Jason Box has been blogging again, seems a bit of a dash to get some kit from Petermann.

Greg Wellman

If I may make a digression into ocean surface temps far from the poles... what the heck is this? http://polar.ncep.noaa.gov/sst/ophi/nepac_anomaly_ophi0.png
It looks like a fairly strong central pacific La Nina, combined with a moderately strong eastern pacific El Nino, that would be much stronger if the center of the warm water wasn't displaced to the north by a fraction. I'm guessing it's technically not "El Nino water" but is hot water from near the Mexican coast that has intruded on the La Nina equatorial conditions.


Dancing, Shifting or Writhing Ice.

Another amateur observation is how much the Ice is retreating north from Svalbard, especially after some comments from the dark side about it crashing into Svalbard and Franz Josef Land during the first week of September. Even see some separation of the Matusevich Ice Shelf area near Severnaya Zemlya.

To see the shifting - dancing - writhing


Greg Wellman asked "what the heck is this.."

Great question Greg! I track the Nini34 status weekly


and know that we are in a LaNina.

So what is going on,your link shows lots of warm water in Pacific, yet there's a LaNina, right?

I went back to my source, NOAA/ NESDIS twice weekly global image


If you open the 2 images in separate browser windows you'll see the reason.

The Pacific is very large, the warm water in your image only extends to longitude 120 W, the cold water in the Nino region goes from 120 W to 160 W. So your image actually misses the major LaNina cold area.

Kelly O'Day


IJIS extent increased 'only' 57K today, but CT area did a whopping 136K. As a consequence CAPIE shot up to 69.28% and 2010 now isn't lowest of the last 6 years anymore.


A simple layman's hypothesis:
Could the rapid refreeze of the Arctic be a mirror of the Antarctic conditions where the surface water is less saline because of more precipitation vs meltwater in the Arctic? This fresher water will freeze more easily and does not mix well with the saltier deeper water which is warmer.

I.e. is it possible that it's stratification of the Arctic ocean that causes this behavior?


WRT Pacific temperatures -

It seems La Nina is weakening a bit, but is a long way from over. Greg's link is to anomalies, not absolute temps, and the link above (from the Australian Bureau of Met) shows that the SST is still quite cold there, if locally warmer than climate. Yes, it seems to be an intrusion of warmer water from off the coast of Mexico. But Kelly's link shows that the water welling up off Ecuador is still very chilly.

While the below-average temps still dominate, waters off the NE coast of Australia are also rising. La Nina will continue to weaken we probably won't be out of La Nina conditions till the new year.

Kevin McKinney

57K? Must be for the 29th; the 30th is about 27K, IIRC.

cynicus, the salinity idea sounds interesting to me; Lodger posted a salinity map at one point, I think; so salinity data seem like a good place to start in thinking about your idea.

I was playing around with yearly mean SIE numbers, and came up with a graph that perhaps provides some slight provocation. See what you think (click on image to see whole graph):



Very interesting and thought provoking graph... On yearly mean basis, the trend in decrease is not so obvious. While the "magic" numbers of SIE minimum are headline material, the average extent may be a better indicator of the state of the ice... Although volume I guess would be even better. When are those CryoSat 2 reports due to come out?

Kevin McKinney

Thanks, Phil. My idea was to try to get beyond the "snapshot" nature of gauging everything by maximum/minimum. I actually think the decrease is least obvious in the 1st half-year means (yellow in the graph.) That's also an interesting series in that 2006 is the lowest year, rather exceptionally. Overall, 2008 and 2009 do give more suggestion of a "recovery" than I'd have thought, though I still don't think that's an appropriate term.

The artificial nature of dividing the data up by year led to me think of another possibility, too, which I'll link here. That is to consider the mean value for "summer," defined roughly as the period from vernal to autumnal equinox (approximated as the period from March 21 to September 23.) We also have enough data for 2010 to do that, which is a distinct plus for me.

sea ice


Kevin. I am a bit puzzled by the information coming out of your graphs. I suppose that a mean extent calculated between equinoxes can be interpreted as an average extent during the decreasing phase (or during the increasing phase if calculated betwen Sep 23 and March 21), However, I was wondering what sort of meaningfulness we could attribute to this value. Is it a valid indicator of sea ice fragility, i.e its ultimate capacity to resist the effect of climate change? or rather, is it a valid indicator of the effect that decrease of sea ice cover can have on the global climate, i.e if the mean SIE is less, there is more albedo effect over time which acts as a positive loop?
I would like to know what you and others thinks about this?


OK, phil, you sent me on a search. "Mean albedo effect"?

What about the area under the graph I asked myself. Particularly when you look at http://nsidc.org/data/seaice_index/images/daily_images/N_stddev_timeseries.png for 2007 and 2010. That steep drop in June 2010 is noticeable. What effect does that have on a whole period calculation.
The same goes for http://nsidc.org/images/arcticseaicenews/20100927_Figure2.png when you compare 07 and 08. The 08 extent rising steeply away from the 07 indicates that the minimum extent is close, but the average ice cover for the Sep-Oct period is very very different.

I could not find an "index number" or some similar simple indicator, on NSIDC or CT anyway, for annual average total ice cover for a year or a season. And I've not seen anyone else refer to such a simple indicator. I suppose for albedo we're really only interested for the period when there is some sun. But for total effect we're also interested in insulation and other effects of ice cover even after the sun's gone down.

I suppose it's pretty easy to calculate the area under the graph and work from there, or calculate number of days above and below certain benchmarks, but I'm pretty cack-handed at such things.

Gas Glo


has 4.90 for Sept 10 average extent.

No win or lose for bets with stoat.

Kevin McKinney

Phil, you've guessed my idea for the "summer" mean--that is, more or less, the period of greatest insolation, hence the period for which the albedo effect is important. So I was regarding it from the point of view of effects, not so much as an indicator. Of course, if you wanted to seriously look at the energy coming in each summer, you'd need to address cloud conditions, which I imagine would be no small task.

I went ahead and calculated the complementary "winter" means as well--I can put them up, too, if there's the interest--but I'm pretty much just playing around and seeing what pops out. I will say that I'll be interested to see what 2010's mean ends up being, which will depend on what happens next.

Meantime, the increase for 1/10/10 was just about 9K.


I had forgotten to check DenialDepot recently - this guy ("Inferno") is a genius, his latest post, Arctic News and Global Cooling Update,
"there were 2 million km² of ice on Arctic in September" is a pure gem. Home-made comments are somptuous, too.

Also make sure you don't miss this superb one, my all-times favorite, Arctic Sea Ice: Staggering Growth, around this marvellous graph:

(be careful, it's severe irony !)

L. Hamilton

In a retrospective spirit, I added September NSIDC area and extent numbers to update three graphs drawn earlier this season.

Observed and predicted (for SEARCH Sea Ice Outlook) September ice extent:

Monthly concentration, or NSIDC area (adjusted) as a % of NSIDC extent:

Cycle plot showing 1978-2010 trends in ice extent and area for each month:


Sea surface temps look scary (or perhaps scared would be more accurate):


BTW, thanks for the graphs, Larry. And thanks for the link to DD, fred. I really think it's the best climate blog out there.

The extent and area increase seems to have slowed down a bit.


Wayne Davidson has a new update up on his website, discussing some of the things that were touched upon in this comment thread. Some excerpts:

Normally La-Nina during winter is good for cold clear nights, easily disturbed by occasional strong blizzards by the now extra warm Atlantic. Its simplicity itself stemming from lesser clouds, but is it? Looking at ENSO, I get the impression El-Nino is making a return, especially by the frequent Low pressure incursions over the Arctic Ocean. ENSO cycles, should no longer be a mystery, they are triggered by pressure differences exacerbating normal equatorial winds to be stronger or weaker Eastwards.


There is also big time events happening in the Arctic, namely in the wake of a great melt, fresher water covers the Arctic Ocean. This is clearly seen here: Its not freezing faster from a cold wave, it is warm up here in the Arctic, the sea surface is less saline, still about -11 C surface air marks the onset of the Ocean regaining some ice (very similar to 2007). But some seas like Hudson, Baffin Bays have warmer water which will delay winter in many parts of North America. Not apparent, El-Nino appeared during a La-Nina, by a reversal of Hadley circulation, blunting the cooling cloud freeing La-Nina trend. El-Nino should normally appear during summer peaking in December, Winter will be warmer than I expected, in a few weeks I will project 2011.

Read more and see the pics at www.eh2r.com/


Thanks Kevin, I pursued the "influence of stratification on ice growth" idea a bit and, ofcourse, it's not new. Many highly technical papers have been published on this topic. For instance: http://www.whoi.edu/science/PO/people/pwinsor/pdfs/bjork_etal2002.pdf

The general consensus between the papers is -unsurprisingly- that a more stratified ocean surface with lot's of fresh water from either river runoff, precipitation or ice melt will enhance first-year ice growth. This is a negative feedback mechanism.

But on the other hand, thin ice allows for a higher heat flux from the ocean to the atmosphere causing more convection which deepens the upper mixed layer. So as the ice thins the convection process may overtake the ice growth resulting in net ice melt. This is a positive feedback mechanism which might be stronger then the negative component. If so, the last ice might melt itself so to speak.



I am not as convinced as Wayne Davidson that la Nina is on the wane. See the report by the Australian BOM that predicts a La Nina event lasting into early 2011.


Phil, that BOM ENSO page is my main source of information on El Niño/La Niña (I check it almost daily because I keep forgetting when the next update is due). I know they predict differently, but I always find Wayne Davidson's interpretations interesting, so I always link to them. He is in the Arctic (most of the time) after all.


A pictorial quick comparison of Arctic Ice Minimum Extents for 2002 through 2010 with animation if desired. Because my "addiction" continues is the excuse.


in the above link, change html to swf for full screen viewing.

Christoffer Ladstein

+ 130 K today according to IJIS. It's with ice as with buildings (only reverse!), easier to tear apart than build up...

Feel sorry for not contributing lately, but time is a precious commodity, unfortunatly. But think of me as one of the flies at the wall...


NSIDC has updated.

Christoffer Ladstein

Less than 15% multiyearice left in the Arctic, that's very close to "nothing". Though NSIDC doesn't reflect upon volume, wouldn't that be a natural conclusion?!


SIE has now edged above 2009 and SIA (CT) is getting closer .This autumn looks like "The revenge of the ice". I also notice that the Northern passage is now closing up North of the Tamyr peninsula.The temperatures reported by various stations around the arctic are pretty low:Resolute -9C ( -15 forecast on Friday), Barrow : -3C, Svalbard : -1C; I keep thinking that it's La Nina ::)

Kevin McKinney

A quick scan of the Canadian Archipelago looks like temps there are more or less normal for this time of year; -9 for Resolute is about right. (Cold enough for most of us, to be sure!)

And about another 65K+ for the provisional IJIS extent.

Greg Wellman

PIOMAS has updated.
A slight uptick, but barely noticeable. Basically the anomaly is holding pretty steady.

Christoffer Ladstein

Phil & Kevin, the temperature in the Barents Sea and south of the Fram strait (Jan Mayen), report of temperatures way above normal, ranging from + 5-8 C , actually above +8 C at the Bjørnøya, just south of Svalbard. The way I see it, this must be remnants of the hurricaines, now draining the "battery" above our latitudes! (Many places in Norway is facing above 20 C these days, so skiing is postponed until further notice...



The important part in PIOMAS is the sentence in 1st paragraph, that states the Sep average was 4000 cubic km, which is a new minimum (of course).


Peter Ellis


Also note that the volume anomaly is currently dropping at ~1500-2000 cubic km per year. Ice free around 2013 is looking increasingly likely, if PIOMAS is accurate.

Gas Glo

I combined
to see 2008, 2009 and 2010 next to next.

What I really want to see is a table of km^2 for 5+year ice, 4+year ice, 3+year ice for the past 20 to 30 years.

For the 3year+ ice, pixel counting or read percentage off graph and apply to NSIDC September average extent? area? or what? Maybe both as a crosscheck?

This: "There is virtually none of the oldest (at least five years old) ice remaining in the Arctic (less than 60,000 square kilometers [23,000 square miles] compared to 2 million square kilometers [722,000 square miles] during the 1980s). "

sounds omninous if our best guess of shape of curve for disappearance of summer sea ice is the pattern of dissappearance of 5 year+ ice.


Good juxtaposition, Gas Glo. I'd have used it for a blog post, if it weren't for the fact that I'm not writing any. :-B

Artful Dodger

Christoffer: Although the Atlantic has been unusually warm this Summer, I don't think the SST's around Svalbard are the direct result of Hurricane Igor, remnants of which are now South of Iceland, tracking East.

The research vessel Jan Mayen has been on station measuring water temps in the Fram Strait and around Svalbard all Summer. You can see it's SST history here.

Over the last 240 hrs (10 days), J.Mayen's mean SST was +1.89 C. The max SST was 4.0 C, (measured 10 km off the NW coast of Svalbard). The min SST was -1.7 C (100 km North of NW coast of Svalbard).

Near NW Svalbard, the salinity of the sea water is 34.5 psu, which puts the freezing point at about -1.88 C. Since 2010-Oct-05 00:00 SST was +4.0 C, no new sea ice is forming here.

100 km N of there, salinity is about 33.0 psu, so sea water freezes at about -1.79 C. Since the SST on 2010-Oct-04 18:00 was -1.7 C, no freezing is occurring yet here, either. The ice edge will continue to retreat here as long as wind and wave action mix sea water with melt water so it does not form a separate, fresh top layer which can freeze quickly.

Greg Wellman

Gili, Peter,
Absolutely, the new volume minimum is dramatic. Even if the rate of decline were to regress to the 30 year trend (ie. maintain the current displacement from the 30 year trend and run parallel to it) there will be an ice-free summer minimum by about 2020 or shortly thereafter. In order to last longer than that, the last few years have to be a temporary anomaly and the decline has to slow/reverse just to get back on the long term trend. I'm still in the "ice-free summer by 2020-2025" camp until we get some Cryosat data and/or another year continues or breaks the current steepening departure from that trend. As you say Peter, two or three more years at the current short term trend *is* ice free, so I have to be prepared to update my opinion in a hurry. :-)

Christoffer Ladstein

Artful Dodger, YOU're the TRUE sarepta bottle! Thanks a lot. The trend the last years is for a very late freeze up in the Svalbard and Barents region, and if salinity, changes in seacurrents (GulfStream is only one of the many, but by far the most important!) or other weathermechanism, I'm not able to tell for sure. Nevertheless, those hurricaines will definitely create a lot of turmoil and generate a STRONG westerly (the most usual)regime this fall, and I'm willing to bet it will last at least a couple months!?

In the end an article reflecting the typical human greed put against protection of Nature, the Arctic of course...



"In the end an article reflecting the typical human greed put against protection of Nature, the Arctic of course..."

One of the worst consequences of an ice free arctic ( if it does indeed happen within the short term! ) would be the irresistible temptation for oil companies in the US and in Russia to begin exploration and exploitation of potential oil and gas reserves in the arctic basin. (see here ).
I am not very optimistic about governments listening to environmental groups with peak oil looming on the horizon and oil prices shooting through the roof!


Ron Lindsay of the University of Washington has updated his webpage.

There's a compactness graph there (like our CAPIE index):

The method worked poorly this year. One reason may be that the ice was very loose at the end of September. Figure 2 shows the compactness of the ice, which is the ratio area / extent, for the last 30 years. This September was a record low compactness in the modern era of very low ice extents. In the early 1980s the September compactness was also low, but the extent was high. When the pack is loose, the total extent is very dependent on the prevailing winds, which this year did not herd the ice to one side of the basin.

Artful Dodger

CAPIE is even more interesting than 'Compactness'. Because CAPIE uses daily data rather than monthly, you can predict a 'knee' in the SIE plot whenever a strong wind field meets loose sea ice.

Ron Lindsay gambled that 'compactness' would be typical for recent Septembers. Let's do a little math to ask what would the minimum 2010 IJIS SIE have been if Ron got lucky and winds pushed the sea ice to the same compactness as 2007?

- Min CT sea ice area occurred Sep 8, 2010: 3,072,130 km^2
- CAPIE at the min SIE on Sep 24 2007 was 71.18%.
- Since Area/CAPIE = Extent, equivalent 2010 IJIS Extent = 4,315,937 km^2
This is just 61,406 km^2 more than 2007, about a single day of good DA winds.

And, 4.3 M sq.km for Sep is almost exactly what Ron predicted in July. Bad Luck!

To be thorough, let's compare 2010 Sea Ice Area (above) with 2007:
Annual Min CT sea ice area occurred Sep 07, 2007: 2,919,439 km^2.
This is 153K sq.km less than 2010, or about 3 days more DA winds.

The only other Sep measure of interest is Volume. I hope one of the Research groups will compare NASA/ICESAT and ESA/CryoSat-2 data to answer this question. But, I also wonder if all the Sea ice will be gone before that paper can be peer reviewed and published.


- Min CT sea ice area occurred Sep 8, 2010: 3,072,130 km^2
- CAPIE at the min SIE on Sep 24 2007 was 71.18%.
- Since Area/CAPIE = Extent, equivalent 2010 IJIS Extent = 4,315,937 km^2
This is just 61,406 km^2 more than 2007, about a single day of good DA winds.

I love these calculations. :-)
It shows how 2010's extent could even be labeled as artifically high.

Artful Dodger

Anybody watching the traffic on the "Lincoln Expressway"? (click to go to DMI.dk)

Lincoln Sea ASAR

For scale, the min width of Nares Strait (bottom left) is 28 KM. Notice the darker gray grease/pancake ice between the large ice floes. The black lines between floes are fractures or leads where the mechanical stresses of the moving floes overcomes the weak new sea ice, exposing the dark sea water.

This image is from 3 days ago. If you visit the DMI.dk Lincoln page, look for images labeled [ASAR] or Synthetic Aperature Radar. Because the Sun has set on the Lincoln Sea until Spring, there is no longer MODIS (visible band) coverage.


Question about the Nares Strait photos. Rough scaling on photos from the 4th and 5th shows movement of about 40 km/day through the strait. Is this normal for this time of year? I thought it was closing up.

Also, Ladbrokes and Paddy Power both rejected betting on the arctic ice melting date. William Hill is at least seeming to consider it:

"Good day!

Thank you for the interest on the betting markets you have mentioned in your recent email. We apologise for any inconvenience this matter may have caused and we appreciate the chance to assist you.

To further your request, please send an email to pressoffice@williamhill.co.uk stating that you are interested of these betting markets.

Should you have further questions, please feel free to contact our Customer Service Department on UK freephone 0800 085 6296 or International freephone 00800 3551 3551."

If anyone feels like having a tilt at a windmill, it only costs you an email. Cheers.


The Petermann fragment has broken out of its eddy and is now moving into Baffin Bay:

Also regarding Nares strait last winter the ice bridge didn't really form the flow will likely continue for a while and 30km/day+ isn't unusual.


Artful Dodger

"Is it normal to have 40 km/day transport through Nares Strait?"

Well, 'normal' is a moving target in the Arctic, so let's start with 2009. MODIS coverage ended for the Lincoln Sea on about Sep 27. In that image, we can see that Lincoln Sea ice is compacted and without leads. There is open water in Nares Strait South of Petermann Fjord, but Nares is blocked with sea ice N of there. Again, Nares Strait was blocked on Sep 27, 2009.

MODIS imagery for Late Sep is not available before 2009 (unless you care to sift through the 'rapidfire' raw images... I don't ;^). Also, the DMI.dk archive contains images from between 21-06-2009 and 07-10-2010.

The CT Area graph for Baffin Bay also provides some useful information on Ice Transport through Fram Strait. Over the last 30 days, approx 60K sq.km of Sea ice has moved into Baffin Bay from the Central Basin (still the only active transport at this stage of the season). Also, about 20 K of this has advected in the last 3-4 days.

The Area of broken sea ice N of Nares Strait on Oct 5, 2010 is approx 12,000 km^2. This is some of NSIDC's reported 60,000 km^2 of remaining 5 year old sea ice. So we stand to lose 20% of the oldest sea ice in the Arctic over the next few weeks. The amount will depend on the date of freeze-up, which in the Lincoln Sea means new sea ice that is strong enough to resist being flushed South through, or jam up, the Nares Strait.

Artful Dodger

Does anyone have the provisional IJIS SIE number from Oct 6? I logged in too late and got just the final number for Oct 6, 2010: 6,035,625 km^2.

Gas Glo

Re me "Also, Ladbrokes and Paddy Power both rejected betting on the arctic ice melting date. William Hill is at least seeming to consider it:"

Sorry forgot to reply to earlier post on subject. Are you aware that intrade did have a market on 2010 arctic sea ice greater than 2009 per IJIS/JAXA? There wasn't a huge amount of trade on it but maybe enough to run that again 2011 vs 2010 if enough people ask for it. Again only an email - help@intrade.com

I am betting on intrade on 2010 and 2011 global average temperatures and did bet on the arctic sea ice being below 2009 but took risk reducing action (does that mean panicked?) late in July when the reversal of gyre and drift seemed to be going on for ages so took a very small profit.

The trade on global average temperature 2019 vs 2009 and 2010-14 vs 2005-9 is negligable so it may be a tough ask to get date of September arctic sea ice below say 10% of 1980-1999 average September levels listed. It seems people do not want to tie up their money for long periods (and who can blame them for that).

Rather than one contract I would prefer three to be listed:
2011 record low extent per IJIS/JAXA
2011 more than 2010, and
2011 more than 2009
Also for them to be listed as soon as possible rather than wait for start of melt season.


provisional IJIS SIE number from Oct 6

10,06,2010,6015156 = pre

Also, I missed getting the provisional for OCT 3, anybody have it?

Artful Dodger

Thanks Jack, you're a Pip! Oct 3, 2010 IJIS prelim. SIE: 5,761,719 km^2
Cheers, mate!


If I interpret the ECMWF weather forecast right, the Dipole Anomaly might be setting up again in a few days.

Artful Dodger

If the DA gets wound up again, here is the sea ice queued to move out:
(Nares Strait bottom-left, Kap Morris Jessup, Greenland in centre, Fram Strait lower-right, click image to open Original at DMI.dk)

NOAA Kap Morris Jessup

Neven: I can never quite figure out how to get to the ECMWF forecast. Could you post a link, or simple instructions? Takk!


Artful Dodger

Even if the DA resumes and sea ice advects through Fram Strait, will it still melt at this time of year? It looks like the ice has already moved far south along the East coast of Greenland .

Artful Dodger

Phil263: Once Sea Ice passes through Fram Strait, it's fate is basically sealed, since there is no returning to the Central Basin. The Sea Ice moves South following the Greenland Current, melting as it reaches warmer, saltier water:

Multi-year Sea Ice lost in this way is replaced in the Central Basin by new first year Sea Ice. In turn, most of this melts out the following Summer. This is why average age has declined so dramatically since 2007. So a key factor in the loss of Multiyear Ice is the looseness of the Ice Pack, since broken Sea Ice is more easily advected by wind and current.


Dodger, you just go here, click on 'N-Hem', then click on one of the forecasts for '500 hPa, SLP'. For instance 168h will show a nice High over the Beaufort Sea, but that's 7 days away and so bound to change.


Alternatively, go here, click on the little map, on the next page click on Northern Hemisphere in the sidebar, and then you can choose the forecast you want from the drop down menus. Not so "dramatically" coloured, and I think easier to read.


Thanks Lodger.
Nice maps. Interesting to see all these North Atlantic currents and how they could be affected by a change of conditions in the Arctic. Positive Feedback loops and negative feedback loops? hard to predict what the net effect will be .. There was a very good post discussing these climate issues back in August. Can somebody remember who posted this? Was it Jim Dowling?

Artful Dodger

Hi Phil, I remember Charles Wilson's discussion about the possibility of the Atlantic Overturning Current (a thermohaline circulation between Arctic<=>Atlantic). If this were ever to grind to a halt... Well Howdy!

Thanks Neven and Gareth, you're the best!

Um, missed the prelim IJIS SIE for Oct 8... Anybody got it ;^)


I found the post that discussed the weather effects associated with arctic sea ice changes. It was actually a piece by Wayne Davidson posted by Neven on July 30. Here .Scroll down to find comment.

The comments to this entry are closed.