« Arctic storm part 2: the color purple | Main | Further detachment »


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


One thing I thought about (after someone mailed me about it): transport of ice from the Arctic Basin through the straits and channels of the Queen Elizabeth Islands, in other words the Canadian Archipelago. It's very cloudy there, but comparing satellite images I noticed some rapid ice displacement southwards just north of the Northwest Passage. More towards the Arctic Basin there is still a lot of ice floes, but as far as I could see all the ice had cracked.

Will this storm blow ice into the NWP?


If there has been a lot of churn would we be expecting some enhanced bottom melt for a few days afterwards or is this just 'wait and see'?

Espen Olsen


Just spoke to Leif Toudal-Pedersen DMI an hour ago, he said, those figures are prelimerary and will be up for revision since it is a big move, and he also explained the DMI Sea Ice Extend figures are measured 50 kms from the sea shores.


It's all wait and see, dorlomin.

Thanks for that info, Espen!

Espen Olsen

The Norwegian Star Flower is really massive this season (left bottom corner):



Warm temps in Siberia. Tiksi, Russia (Near Laptev Sea) is currently 70 degrees (F). Wunderground forecast is for Tiksi to remain in low 70s through Sunday.

Espen Olsen

Oden Swedish Ice Breaker:

For those interested, Oden is on a mission named Lomrog 3 and you can follow, in Swedish, their reports here: http://www.sjofartsverket.se/sv/Sjofartssektorn/Sjofartsforskning/Polarexpeditioner-med-isbrytaren-Oden/LOMROG-III/


At 970mb thats .3m increase in sea level, a lot of water, sucked in from? bering? baffin?, and released it'll pump out the base. How big is that 970mb area? Anyone up for the math? or am i wrong?

Artful Dodger

In Arctic storm part 2, Chris Reynolds wrote (August 07, 2012 at 19:33)

"The problem with attributing this storm to the open waters is there have been open waters in recent years but no storm."

It's important to understand that this is event is nearly unprecedented. There had never been a Summer (JJA) arctic cyclone before 2006, and most were Fall/Winter (NDJ) when the temp differential between open seas and ice is largest. Since 2007, these storms are getting larger and more frequent. Nevin described the Beaufort Sea cyclone of 2008. No Arctic cyclone has ever been the size of GAC-2012, at any time of year.

This is like playing a game of Russian Roulette, where the climate loads the revolver. Before 2006, there were zero rounds loaded in the cylinder. In 2008, there was one round in the cylinder. Now in 2012, there are three rounds in the cylinder, with one in the chamber.

Question is, Do we feel lucky?

Paul Klemencic

johnm33: Your math is correct. That is a lot of water, although its only near the eye. It would pull the water in from the storm periphery.

The daily MODIS mosaic linked to by the Daily Graphs tab above, is showing the ice around the eye is really broken, and even somewhat aligned with the prevailing wind field.

Paul Klemencic

Darn, I really shouldn't say "eye", we don't really have eye; I should have written "storm center".

Neven's track of the storm shown above is interesting. The LP system migrated up from Siberia, but it almost "homed in" , or slid into the sweet spot where the storm could pick up feeder bands of warmer air from Siberia, Alaska, and Canada. Again, I am thinking this might be a harbinger of events to come in future years.


Just wanted to give Neven a heads up that I send him more frames of the EC composite, reprocessed to get rid of more of that nasty cloud artifact that was in the first set, courtesy of GAC 2012/Gaagi (although I still like Haruhi).

I suppose you all will be seeing them here in some form or another when Neven gets around to it in his (no doubt copious) free time - at least I hope so!

Espen Olsen

With those temperatures in Siberia (Tiksi) now 18C, that can give birth to many storms?

Al Rodger

Re The energy imported by the storm surge.
A very quick calc (it could be checked), assuming the storm is 1,250km diameter averaging a 0.3m surge, the water would be about 367 cu km & that would contain enough energy per increase in deg C (4.18 j/g/K) to melt 4.6 cu km of ice (@334 j/g). So even if it were a lot warmer (say 10 deg C), it doesn't appear a cataclysm for the ice, although an extra 46 cu km of melt is not trivial.

Artful Dodger

The 08 Aug 12Z surface analysis chart is out from EnvCda.


The depression is now centered at W149, N82. Over the past 24 hrs, central pressure has risen from 968 MB to 974.

The 1012 mb isobar (the outer limit of the low) still extends inland South of Barrow Alaska, a distance of some 1340 km from the center of the depression.

Andrew Borst

How many sq km are above 90N? Just trying to figure out if DMI drop is possible. TIA

Paul Klemencic

Andrew: Rephrase your question. There are zero sq km above 90N.

Daniel Bailey

I believe Andrew meant 80N. Though I've been Poe'd before...

Andrew Borst

Sorry, 80N.

Espen Olsen

Ryder Glacier North Greenland:

Location 4th Ice outlet to the right from Petermann, have seen some major calvings in the past 10 days!

Daniel Bailey

Per Howat and others (2008), Ryder Glacier accelerated by 300% over a 7 week period following drainage of a supraglacial lake in 1995. This indicates the ability of an unusually large sudden discharge of water can increase basal water pressure dramatically and enhance basal sliding.

Ryder Glacier has an order of magnitude less melt than Jakobshavn and would be more susceptible to such a sudden meltwater pulses. Perhaps this melt season's discharge in North Greenland was greater than normal? ;)

Artful Dodger

Andrew: None of the DMI drop came from N80 to the Pole. Look at the SI maps, it's more like 75N to 78N, between W165 and W180.

Daniel Bailey

Apologies for not posting this with the above comment (I blame early senility):

Horizontal velocity field of the Ryder Glacier. Contour interval is 20 m/yr (cyan) for velocity less than 200 m/yr and is 100 m/yr (blue) for values greater than 200 m/yr. Red arrows indicate flow direction and have length proportional to speed.

[Source (Scroll down for even more on Ryder Glacier)]

Artful Dodger

Andrew and folks,

Here are maps of Arctic SIE from DMI: (these'd make a fine animation of GAC-2012 ;^)


Andrew Borst

Yes, but I was only using the measurement for reference of size. I found an online calculator that gives me the distance b/w two sets of lat/long coordinates. It gives me the distance b/w 85N-0E and 90N-0E is 568km. Next, the distance b/w 87.5N-0E and 87.5N-90E is 393km (1/4th, had to use that since the calculator would take me through the NP). So 588 * 393 * 4 = ~875,000 KM^2 above 85N. Okay?

Paul Klemencic

Andrew Borst: Here's my guess on your logic. What you really want to know is how much high concentration ice exists above 80N. This is where the MASIE regional maps come into play. Before the storm, the MASIE report for the Central Arctic Basin region (which isn't bounded exactly at 80N but average boundary seems to be around 81N) had 3.1M sq km of ice extent. Almost all of that would have been high concentration ice.

The DMI report shows 30+% SIE below 4M sq km, and if 3.0M exists in the CAB, that leaves only 1.0M for the lower latitude regions.

However, please note that DMI doesn't measure ice extent within 50 km of land, so none of the Canadian Archipelago ice, plus a big section of ice in the CAB within 50km of the CA or Greenland, aren't included in the count.

There is likely some ice in the Greenland sea that counts, and almost nothing in the Laptev region. So that leaves the pre-storm 2.2M sq km that were in the Beaufort, Chukchi, and E. Siberian. Judging from the decimation in those regions, Its likely the SIE in those regions is less than 1.5M now, with a lot of remaining ice at lower concentration.

So it is possible that the DMI is correct, especially if some holes opened up in the CAB. But its difficult to compare the DMI with the IJIS SIE.

Too bad the MASIE regional maps don't show the big melt yet… but wait a week, and the data will begin to sort out.

Ian Allen

We can see multiple tight swirls of around 30km scale near the low centre and comprehensively ripped up ice underneath. The ice near Wrangell island can also be verified as departing.
strong southerly winds forecast over much of the cut-off ice in a few days.


Don't you want Pi.R^2 where R=568 (or is it 588 written later?)

=1013685 km^2

or maybe if you don't like assumption of area being flat:

area = 2*pi*R^2 * [sin(lat1)-sin(lat2)] * [lon1-lon2] / 360

Paul Klemencic

Artful Dodger: That DMI map helped a lot. I wonder if they count the NP hole in the data?

Alberto Silva

ECMWF is now predicting a Dipole Anomaly after the storm ends...


(no comment)

Paul Klemencic

Alberto: Wow… but thats pretty far out, about a week away. But if that Dipole Anomaly forms, and sticks around for awhile, that would just about put the finishing touches on a disastrous year.

( I also note that it shows another Arctic cyclone for the Bering Sea at the end of the week.)

Paul Klemencic

Sorry, end of the forecast period, which would be August 18.

Espen Olsen

Ryder Glacier calving follow up:

The calving happened sometimes after July 10th
and the size is about 50 - 60 km2 or about ½ the size of the latest Petermann piece.

Seke Rob

From the OP

Cryosphere Today has reported the sea ice area number for August 5th, a drop of 53K.

Did I finally convince someone that the first record of CT, 1979.0027 is Jan 1, and the rest following from that ;?

Anyhow, updated some charts, the next 2 having a second lead line with a 17% "What if" melt... yes all prior years are having grass to eat.


And one showing how much melt still to go from this date to minimum with same added second lead line.


MASIE has trouble keeping up with JAXA / DMI [as Paul noted]... in one combo picture:


The last, but all others I'm doing have also been put at the latest state of data availability, the Arctic main annual curves on top of 1 and 2 std.dev off from the 1979-2008 base line. The 2012 anomaly wise continues to run at over 3 std.dev. for the Area metric


Oh, and 11K more area to go and the 85% or 28 prior years becomes 88% and 29 prior years undercut.

Andrew Borst

Thanks, crandles. Circle, duh!

Using the first equation, the sq km above 80N is ~3.9 M km^2. So 3.9 - 1 = 2.9 b/w 80N-85N. Divide that by 24 and that is how many sq km each 80-85N cell in the Bremen concentration graph represents (0.12 M km^2). That helps me a lot to translate these maps in actual numbers.


Dodge - Made you a GIF from the images after 7-22 (my hand got tired)

You can email me at dabize@gmail if you want it.....or just port the images into ImageJ and use the Stack Sorter plugin to make your own

Artful Dodger

From the OP

Cryosphere Today has reported the sea ice area number for August 5th, a drop of 53K.

That's August 6th, as per Neven's Chart. Notice that the CT Sea ice map is labeled 08/07/2012. That should be your alert that the tabular data is out for August 6th.

Artful Dodger

Thanks, dabize. Maybe just send it to Neven so he can share with the group.

Btw, the Aug1=>Aug7 changes are shocking to see, mainly because they foreshadow the 15% threshold losses to come.


Patrice Monroe Pustavrh

Dabize, I can agree with Lodger, it would be very interesting to us too. And you can get many "Nice work Dabize calls" for that :), let's say this is second one (after Lodger's)


Dodger - I already sent the EC composite animation (plus some extras) to Neven - he said updated he'd get it up soon, but he is kinda swamped.

I'll send him this too, but I'll blame you........


Cryosphere Today has reported the sea ice area number for August 5th, a drop of 53K.

I'm sorry, I copied that date from yesterday's post. Should be August 6th. Fixed now.

You know what I always do to check if I have the date right? I go to the interactive CT SIA map and check by hovering over the 2012 trend line. Luckily it's below all the others, so it's easy to do. That shows you day of year (also keep in mind that it's a leap year).


Andy Revkin has a post up

Seke Rob

Just for the record, a reply I had from Bill Chapman

"Yes, the .0027 should be the first of the year and .0000 is the last day (of the previous year). Yes - it is a legacy decision and it works well for my particular plotting package. Sorry for the confusion."

I used to be good with math/algebra and the rest ;>)

Seke Rob

P.S. too quick hitting enter, Bill volunteered more information what he does, so above was just the relevant snip.


Arctic.io's images give a great overview of the storm - Lots of water visible around the center but the area wasn't too solid before our upheaval.


Paul Klemencic

Looks like the MSM has woken up (Andy Revkin is the right person to start this). Dr. Chapman really soft-pedaled it with his response. I hope Andy gets regular updates from him, and doesn't stop with this first set of comments.

I think we need to put some comments up there at the NY Times Dot Earth blog, but don't know where to start. I am a subscriber, so maybe I get some links and put them in the comments.

This could take a lot of time, so I'm not so eager to jump into this. But leaving this go with no comments...

R. Gates

Left comments on Revkin's blog.

In the meantime, here's a quick link to a perfect view of the Great Arctic Cyclone of 2012:



Very interesting discussion above. I think we are witnessing the very scary and surreal transformation of the Arctic.

Has anyone here written on what happens to the ocean currents as the ice pack disengages from the Canadian archipelago and Greenland?

Will the refreeze and salt release in a circle about the pole cause a circumpolar current similar to that about Antarctica?

Here's to a miracle cold snap and an early freeze

Artful Dodger

Hi Seke Rob,

I've been explaining how CT encodes dates on this blog since 2010...

Here is my July 27, 2011 at 22:11

First, look at the top of the CT SIA table. The very first line begins with a year.frac value: 1979.0027 ...

Note that this applies to the Sea Ice maps as well. So when you see Aug 7 on the Map, they mean midnight and the first moment of the day, Aug 7th. So really, it's the day-of Aug 6th. Ugh. But that's how they do it.

The only thing I would add to that now, 13 months later, is that CT designates Jan 1st as Day zero on their interactive chart, which messes with people that are used to the MODIS convention of Jan 1st = Day 1.

Good luck, and Cheers!


Yesterday's full Arctic Mosaic as megapixel zoom:


Low concentrated areas are popping up closer to the North Pole. A cloud free view would be just heartbreaking. Today's zoom will be released once the satellite did his work. Looks even more devastating.


Ryder Glacier calving follow up:

The calving happened sometimes after July 10th
and the size is about 50 - 60 km2 or about ½ the size of the latest Petermann piece.

Comparing two MODIS images from 08/05/10 and 08/07/12 I can't see a signifcant calving event at Ryder Glacier. There are moving some pieces around. But at the front of the glacier is no really big chance visible.

Maybe I'm wrong...

Chris K

Paul Klemencic

I put four comments on Dot Earth (Revkin's blog), covering:

1. Century breaks and flash melt events, with the 188k and 170k flash melts on August 5th and 6th.

2. Dr. Kerry Emanuel's paper "Polar lows as Arctic hurricanes" and Dr. Jeff Masters comments about the lows being stacked in this storm.

3. The location of the ice cams (about 1200 km from the center, versus Barrow at about 600 km).

4. Linking to Neven's first post in this series "Cyclone Warning"

Wayne Kernochan

@Paul Klemencic: Andy Revkin is not not not the guy to "start this." Joe Romm has had a running argument with him for soft-pedaling climate change in a questionable way. It's possible that he cherry-picked the quote to tone it down.

The right guy to start this at the NY Times is Justin Gillis. He knows what he's talking about, and he usually delivers it straight. - w


Paul K wrote:

I also note that it shows another Arctic cyclone for the Bering Sea at the end of the week.

Not so much the Bering sea but the Sea of Okhotsk.

Tropical storm Kirogi is supposed to arrive at the Sea of Okhotsk on 9 August at 21.00 h local time (that's midday 8 in Europe and very early morning in the States).

If it would be like that, Kirogi will send warm air and moist again over the Siberian continent towards the Siberian Islands.

The very same thing happened in 2007. But still, it's only a prediction of course.

Daniel Bailey

Wayne Kernochan nails it (Gillis over Revkin), every day that ends in "y".

Paul Klemencic

Kris, the forecast shows a cyclone in the Bering Sea on August 18th, a week later (and way too soon to see if this happens, but I found it interesting.

Seke Rob

Re Artful Dodger | August 08, 2012 at 21:28

That's fine, Lodger.

Susan Anderson

Glad to see some people who know what they are talking about have already got going on DotEarth. Unfortunately, he enables a lot of fake skepticism but still has enough of a reputation to carry weight. FWIW, you should know he posts comments in batches, so don't think it's anything personal if it takes a while to appear.,

Glad to be behind here, but here's the link for those interested:


Rutgers snow cover anomaly for NH in july is down again - a new record low - who could have guessed:

July snow cover NH

Charles Longway

Paul, Just a note on cold core cyclones inspired by your hand drawn diagram on Monday. Cold core cyclones have an unstacked slanted core with a southern arm feeding in warm air and a northern arm feeding in cold air. Since GAC-2012 is over the pole the warm arm is spread around a 360 circle feeding a northern arm at the pole. This circular symmetry is different than extra-tropical cyclones further south. Given the polar location our current cyclone has the same dynamics and when it is displaced from the pole we should see the core slant with the top displaced north and the bottom south. If we see this slant it will give some confidence as to clasification.


Ryder Glacier appears to me to be about where it was on 09/03/2011 - even to the notch.


I do think once the fast ice melts out completely we may see some action during the succeeding spring tide. (Perhaps the 21st-22nd.)




As I step through the forecasts at http://www.wetterzentrale.de/topkarten/fsecmeur.html (N-Hem, 500hPa,SLP), it projects GAC-2012 to drift over the Canadian Archipelago and dissipate.

However, it shows it being replace by a couple more strong cyclonic systems in the area, dipole anomalies and similar disturbing features for the health of the ice.

In particular, the current remaining thick/concentrated ice seems to be gathering above the Fram Strait. Evolving pressure gradients look like they'll favor strong export out the Fram in the coming days. Could be a one-two punch for the remaining ice volume.

At the moment, O-Buoy #4 is in the Strait and has stopped moving south. When it resumes, it will probably take a lot of other ice with it. See:

Parenthetically, I find it fascinating that the buoy's motion in the strait seems to be strongly affected by tides. At least, I think the odd periodicity of movement must be tidal-related. I don't know if this has been observed previously.

Sigh. Small consolation, for the betting crowd. Intrade has 50-50 odds on whether a new sea ice extent record will be reached in September. Looks like easy money to me:

-- Steve


The proportions of the floe catastrophe are clearly visible already now at the IARC-JAXA site.

Paul Klemencic

Charles Longway: I saw your comment, but I think you are pushing my knowledge to the limit. Some storm experts have hypothesized about the existence of storms like these, and I was hoping some of them might opine on this storm, and chase away any doubts.
For example, I read Kerry Emanuel's argument for arctic hurricanes in Tellus, which actually dates all the way back to 1988/89.

The fact the storm seems to be dissipating a bit faster than originally forecasted, makes it more plausible that this isn't some new type of monster storm. But right now the storm is moving toward the Beaufort, (which will do even more harm to the ice), and the lows are still stacked at least up to 500 mb. Higher, I don't know.

I was hoping a real storm expert could tell us what is going on here. Some of the literature suggests that an Arctic cyclone could hang around for 20-30 days! A summer storm like that would really do a number on the ice. And remember, we don't get big Arctic cyclones in the summer.

Ghoti Of Lod

Is it possible / has anyone tried to look at tethered profiler data for any indication this storm is mixing surface water with deeper saltier water?

Birth Mark

I've been reading this blog for about two years now. I never post here because I honestly don't have much of anything substantive to add to the discussion.

I want to post now to thank you for the blog and to thank the many fine people that post here. Your knowledge is extremely helpful to me and I'm sure many others.

The blogging on this storm and its effects on the Arctic Ice have just been incredible. Thanks for your hard work and willingness to share!

Aaron Lewis

As I look at models of the storm, I note that the dew point at 2 meters is above freezing. Thus, the storm is putting a lot of latent heat in contact with the ice. One gram of water vapor will melt 7.5 grams of ice. At 3C there is not much water in the air, but with those breezes, there is a lot air in contact with the ice. The resulting film of condensate changes the albedo of the ice and fools the eyes in the sky.

Pity the poor seals and bears.


Thanks to Dr Jeff Masters (who I've known since 2004 and blogged for during the super storm years of 2004 and 2005 - which led to me leaving United Airlines and quickly establishing a full-time biz forecasting tropical cyclones for energy traders ever since) - I've 'discovered' this blog. It's been a true pleasure to finally hear from so many scientists who have also become as fanatically interested in Arctic ice and the Greenland ice sheet.

I officially began my career as a forecaster in Alaska in the 1970's, and this remarkable storm is not some new type of 'bizarre / freak' type of storm. (The global models handled this just fine - a 'unique' type of storm would not have been picked up on). It's also not a polar or arctic hurricane in the traditional sense. Those relatively small cyclones can produce hurricane force winds - usually 'triggered' baroclinically as very cold arctic air moves out over relatively warmer waters (and usually eaat of Greenland) - and get much of their energy through latent heat release from the open waters just off the ice pack edge. They also dissipate once they move over land or the ice pack itself, and have only been observed in the winter. This is 'The Great Arctic Basin Storm' (or whatever). But my point is that it is a synoptic scale event normally seen in the cold season - but for reasons already mentioned - are likely to occur with greater frequency within the arctic basinduring the warm season due in part to GW and large areas of open water within it.

I began 'watching' the arctic ice dince the SATY record began (gfirst thru military contacts - and then whilee doing my research work at UAL in the 1980's. I've become pretty proficient at interpreting the imagery & data; and predicting what kind of melt season we would have.

For instance, despite the intense cold this past winter in northern Canada - which normally would of caused a very thick ice formation in the NW Passages (north and south routes) - we've seen it melt out very quickly - pretty much at the same time it has over the last svereal seasons. I suspect that happened because of the very intense and numerous storms last autumn and early winter that moved from the Bering Sea on across Alaska and then northern Canada and put down an early and relatively heavy snow cover on top of the newly formed ice. This snow cover no doubt reduced the rate of ice thickening - just about offsetting what the much colder winter was doing.

In many ways - the hemispheric circulation pattern has not totally given up on the La Nina pattern of the past winter. I suspect 'normal seasonality' forcing simply shifted the high latitude storm track from the Bering Sea / northern Canada further north - and once the upper level cold core polar vortex began to deepen (as the models correctly foreecast) we suddenly found ourselves with a huge and very deep surface storm that has essentially occluded out and will spin itself away over time.

Once the dust settles (so to speak) - we'll have just 1 example of what a major storm within the arctic basin during the summer does to the ice pack. Drawing too many 'conclusions' from it may not be a good idea. It will take several of these over the coming years before anyone will be able to model the impact they have on the ice pack - and just how much of a role it plays in the rate of meltout.

Aaron Lewis

Serious storms are named, not with letters and numbers, but with real names.

I propose that we name this storm; "Dragon King".

No, this is not "the big one". This is the herald that tells us the big one is on the way.

Steve Bloom

Hi Steve, pleased to see you pop up here. I remember you very well from those WU days.

Just to be clear on one point, most here (regular commenters anyway) are amateurs on this stuff, even if very engaged in the observations and science. That said, there are several Arctic/glacier specialist scientists who do comment here with some regularity, a few others with less frequency, and I'm sure a fair number of others who just lurk.

And of course, thanks very much for the info you provided!

Susan Anderson

I still like Gaagii, per Lisa:

Navajo for "Raven". Raven is a Trickster God; the Yupik have a legend that Raven created the world
even if it is only the herald.
Also, what Birth Mark said.

Tor Bejnar

I duplicated Seke Rob’s CT - Arctic Sea Ice Area graph to make sure my Excel routine was right - “1st Day Below each M”. I then created an extended data set for “1st Day Below each 250K” (4 x the data). Using Excel’s linear regression tool, I projected phantom data for a “1st Day Below 1M” for each year (e.g.1994 = Day 279; 2005 = Day 262) . Just as every set of data connecting “1st Day Below each M” on Seke Rob’s graph appear approximately linear, so does the phantom “1s Day Below 1M” set (with the1996 datum being above and 2007 datum being below the trend line). The 1979-2012 best-fit line crosses Day #244 in the year 2020. Because so much changed with the 2007 melt season, I also calculated a best-fit line for the 2008-2012 subset which crosses Day #244 in 2012 (yup, this year!). Day #244 (September 1) marks the early end of the melt season (e.g., 1992) and other years lose, on average, nearly 200K from that date.

In 2006 (I understand) Wieslaw Maslowski predicted the Arctic would be essentially ice-free in September 2016 +/- 3 years. That compares well with my projections to 2020 and 2012, which just happen to average 2016!

This great big storm, like the continuous dipole in 2007, is setting the stage for sea ice area to be reduced to 1 million square km in the near future.

Frankd 1977

Detachment complete!


Paul Klemencic

Hello Steve Gregory:
Welcome to the blog - it sounds like somebody with your expertise could be really helpful here.

Here's a short list of questions - feel free not cover them all one time, or address them at all.

We have not seen Arctic cyclones in the summer before. why is that? Why do you think we are seeing this one now?

Some of the literature suggests that Arctic cyclones can last for 20 to 30 days. What are your thoughts on that? Do you think that summer cyclones could possibly last that long? How intense do you think the cyclones could become, given access to warmer water and warmer air inflow?

Could you describe how this current cyclone works in laymen language?

What are your thoughts on the fact that the polar region's turning from summer high-pressure dominated to low-pressure dominated region?

If Arctic amplification continues to warm the lands surrounding the Arctic Ocean above 20°C in the summer, what is the possibility of a storm setting up, fueled by an inflow of warmer air at lower latitudes, and rejecting heat to the upper troposphere near the pole? If a large storm like this current one set up that does that, could it persist for some time?

Can you anticipate any other persistent weather patterns with predominant wind direction, resulting from the increased summer surface area of Arctic Ocean water?

And finally, the really big one. What do you think of the "polar amplification fuels NH mid-latitude extreme weather events" theory proposed by Jennifer Francis and Steve Vavrus, among others?

Steve Bloom

For now, anyway, very few of the usual denialist commenters have said anything in that DotEarth thread. Hopefully they're feeling a little gobsmacked by reality, as that's the first step toward recovery.

I'll add my thanks to some of the regulars for the good job they did there.

Bob Wallace

Bremen is showing a lot of ice loss on the south end of the Greenland Sea. I would assume that the data from there is reasonably reliable, not messed over by wind from the storm.

The Navy is predicting little to no ice movement through the Fram for the next several days, no replenishing of the GS stock.

I suppose someone could make something of that....

Steve Bloom

Looks like, Frankd. Even if some of that ice reappears as anticipated, it now seems certain we'll end up with an isolated ice island. The question of whether it survives the season or not can become a whole new topic here, as if Neven doesn't have enough to do.

Paul Klemencic

Frankd: Indeed, the Bremen map tonight showed an incredible amount of damage to the ice pack. Maybe the clouds cleared enough to see the extent of the damage better.

I don't even know where to start, so lets start near the storm center. The four quadrangles above 80N on either side of 180 longitude show large zones of ice concentration below 70%. As the storm center moves toward the Beaufort, the wind should shift and begin pushed the fractured blocks and floes away from the main pack, and drop the concentration further. We could see a big bite out of the Central Arctic Basin pack above 80N on this side.

In the Beaufort itself, the storm should push warmer surface water into the pack, and compact some of the pack. Perhaps the end of the McClure Strait will open up, and open the NW passage.

When the new higher pressure systems move in next week, it will be interesting to see what the ice pack around the NP looks like (when the cloud cover dissipates a bit). On the MODIS Mosaic, there appears to be fractured ice within 400-500 km along ranging from 90E to 150W. I won't speculate on the increased chance for open water at the pole this year until we get better pix, and see if the dipole anomaly sets up as forecast.

The IJIS extent numbers should make grim reading tomorrow.

Charles Longway

Paul Klemencic: Thanks so, so much for you insightful commentary on cyclone structure. I second your request to Steve Gregory for any expert help he can give in understanding arctic cyclones.
By the way, I see that Nevin has a really good message on Joe Romm's website. I am really glad to see Joe give Nevin a front seat.


Otto Lehikoinen

I'd imagine the pressure cannot be used straightforwardly to compare this low to the more southern ones as the Coriolis force should (or shouldn't it) weaken the winds around the center of the low, possibly making high northern lows (or storms) lasting longer than those in NP or NA. Can someone find confirmation of this sort of deduction?

Tony Duncan

Steve Goddard is still taunting "warmists, saying they are all basically cowards because they won't bet him that there will be a record minimum extent. He just yesterday posted that he was offering to bet $1000, while his commenters all are bringing up bizarre explanations and flt out denial about what is happening.

Espen Olsen

Ryder Glacier North Greenland:

I am sorry to be a "bit" premature and not checking the images from last year, I admit I only did compare with images early July and they looked like a calving, sorry about the fuss.


Seems I spoke too early when I was grateful we'd be spared the Ekman pump. Still, I think there'll be 2.7m or 2.6m area left at the end of the season (I used to vote for 2.9m).

@Tony Duncan
He seems not to understand people won't trust a person with a history of moving the goalposts.

Arctic Haruhi at 0:54

Steve Bloom

Re Goddard, as with Watts, there seems to be an audience eager to hear comforting lies.


Since my CAD count north of Greenland june 2011 I have come to describe the Arctic Basin pack as ‘a mesh iron pattern of more or less unified floes within broad leads of rubble’.
On MODIS r04c03 2011 that pattern was still clear over 90% day 222.
Today's same tile is not complete yet. But what is showing up through the clouds on the Bering side has no pattern anymore. It has become part of the ’patternless loose floe-area’.
Last year that was indicative for r05c03/r05c04, Idunno’s ‘slush puppy’.
The mean concentration there would be hovering just over 30% and mean thickness of the remains about 40 cm.
If these attributesare consistent, a cardboard calculation shows that the actual volume of the ice may already be below last years’ record low.
Less than 1,8 MK still has the 'mesh' at 90% (mean thickness 1,2 m), 2,2 MK is 'loose' at 70% (mean t=0,8 m), 1 MK is prone to melt out at 60% (mean t=0,4 m).

Andrew Dodds


Long time reader, first time poster..

Was just playing around with Cryosphere Today and came up with this comparison (Aug 6 2007 vs Aug 6 2012):


It's genuinely disturbing, it illustrates just how much has changed. Even in the 'freak year' of 2007, much of the sea ice that remained was fairly contiguous pack (~100% cover); whereas today it looks like there's barely any solid ice areas left.

Rob Dekker

I've been poundering the issue of how much sea ice melt this storm would cause (if any) and look for evidence of that, but have not gotten very far. This is mostly because I am not a meteorologist, nor have any experience in atmospheric thermal dynamics.

So, I was hoping that you guys could help out to get some ballpark estimates done, and shoot some holes in this story.
For starters, a large storm over the Arctic will sure stir up things. Atmosphere, ocean, and ice itself. But is there an identifiable energy source that would cause melt ? Or is it just all rock-and-roll without beef ?

One source of energy I could think of is atmospheric heat.

A large storm like should pull in a large amount of lower latitude 'warm' air, which will likely cause surface melting in the outer regions of the storm. However, once the surface air moves further into the spiral, the pressure drops, and if we assume adiabatic expansion, the air temperature should drop some 8 C when going from 1000 mb to 970 mb. So if the air pulled in was 8 C, it will leave (in the center, and upward) as 0 C, meaning that there is no energy left over to melt any ice. In extreme cases (which this storm may be), expansion may even drop the temps below freezing in the center, which obviously would prevent any melting.
Also, Looking at the NCEP/NCAR data, surface temp anomalies are negative under the storm, which kind of confirms that we should not expect much ice to be melted from atmospheric heat pulled in from lower latitudes.
What do you guys think. Is this correct ?

Then there is ocean water movement.

For starters, as johnm33 pointed out, a 970 mb low (at center) will pull up the sea surface about 30 cm. Over the size of the storm, this should form sort of a 'cone' of water, with a volume of about 1/3 pi r^2 h. With h=0.3, and r=500k radius of the storm, which is some 78 km^3 water (assume diameter of the storm as 500 km diameter).
Where will this water come from ? Well, water follows the path of least resistance, and since the storm is much larger than the Arctic is deep, the entire Arctic basin will kind of move towards the storm a little bit. For the case above, (30 cm elevation in the center of a disk of water with 500 km radius, and something like 1 km depth), the outer bounds of the storm will all move about 250 m inward. Sort of what you would expect during tidal movement.
Needless to say that if the entire Arctic basin moves 250 m towards this storm. there is little heat movement in that process, so we can't expect this effect to have any significant influence on sea ice.

In my opinion, the most likely source of heat would be Ekman pumping, about which we learned a lot last month.
The concept is simple : a low pressure zone will cause a counter-clockwise air flow (a cyclone) which will exert forces on the ice. Coriolis forces will drive the ice outward, which will cause ice divergence of the upper layer of the ocean. This effect reaches down to 10-20 meters from the surface, called the Ekman layer.
Of course, if the upper 10-20 meters diverges, water will be pumped up from beneith that layer to replace it.
Now, it happens that in the Western Arctic, between 20-75 meters, there is a relatively 'warm' layer of stratified water, which is about 1 C above the melting temp of sea ice, which will cause bottom and lateral melt of the sea ice above. To quantify how much heat we are talking about, consider that the outer bounds of the storm diverge at something like 0.5 m/sec (which is reasonable if we look at ARCc ice velocities). Over 20 meter depth, this will displace (2*pi*R*d*v) or 31 Mm^3/sec.

If this water is replaced by water with 1 C potential melt temperature from the 20-75 meter layer, this causes an average of 168 W/m^2 bottom-melt to occur over the entire area and during the entire duration of the storm. Translated to sea ice area lost, we are talking about some 396k m^3/sec ice loss or (for 0.5 meter thick ice) some 68 k km^2/day sea ice area loss.

Note that this ice loss and the 168 W/m^2 bottom-melt is as high as heat absorption during a clear day in July, so Ekman pumping appears to be subtantial and relevant during this (and many other) late-summer storms.

In areas where ice concentration is already low, this could mean the final punch which knocks it below the threshold (don't we see something like that on the DMI graph?).

Espen Olsen

M'Clure Strait:

I looks like the storm is refilling M'Clure Strait:


"Denial is a river in Greenland" commentator on a Guardian blog post.

Thought it had a certain ring to it. :-)

Rob Dekker

Part two :
So that was my back-of-the-envelope calculation of how Ekman pumping under a storm in the Western Arctic may cause significant bottom melt.
Of course, what goes up, must come down, so at the edge of the storm we would expect surface waters to flow down to replentish the 20-75 meter layer, and we would expect a back-flow to the center in that layer (which, incidentally, may follow a clock-wise spiral current pattern).
That's the theory.

What about real life ? Do Arctic buoy data from the past week confirm this story or not ?

Here it gets a bit frustrating. Arctic buoys seems to be malfunctioning or taken out of commission by the buckets, and very few get replaced. Government budget issues ?

Most of the many IMB (Ice Mass Balance) buoys from CRREL seem to have broken top-sounders or thermistor strings, so they don't record ice thickness and melt rates any more, and most of them are in the Fram Strait area, which is not muh affected by this storm (yet).

The only one left alive over in the Beaufort is 2011J :
This one records only temperature and pressure now, but is on the boundary of the storm, and thus interesting.
It does record anomalously high atmospheric temps (up till 10 C!) over the past couple of days, which suggests that indeed this storm pulls in a lot of lower-latitude heat. See the first paragraph of my previous post.

Then there are the Arctic Ocean heat FLUX buoys. These are the good ones, since they record bottom-melt rates (in W/m^2) which is highly variable information that we have used multiple times here at Neven's to make sense of what is happening in the Arctic. Here is the database :
Most of these buoys broke down a while ago, but thee have been alive this year, and two of these are in the Beaufort and thus relevant for the storm. Unfortunately, all three of the last ones alive (AOFB21, AOFB25 and AOFB 26) have been shut down on July 28, without any explanation from NPS. That leaves not a single FLUX buoy left alive in the Arctic at this time, and no way to record any bottom-melt (arguably the most important information we need for understanding long-term ice thickness development) anywhere in the Arctic. AHHH !

So the last hope is on the ITPs (Ice Tethered Profiler) buoys.
From the dozens of ITPs that once were operational throughout the Arctic, there are now only a handfull still alive.
There are two in the Beaufort and on the boundary of the : ITP 53 and good old ITP 41 (a marvelous device operating for two years strait in the harsh Arctic environment).

Here is ITP53's (76.0108° N, 151.4687 W) profile :
And this is ITP41 (74.1969° N, 135.7053° W) :

Both of these profiles show significant 'stirring' in the upper 75 m that we would expect from the Ekman pumping.
The 20-75 m 'stratification' layer spread out all the way to the surface over the past couple of days, with salty water from that layer mixing with the surface water right under the ice.

However (and that is unexpected, at least by me), we do NOT see a significant increase in surface water temps, and second, we see that the effect of the storm reaches much, much deeper than I expected.
Layers of water down to 200 meter, where the water is again much colder, are definitely affected by this storm, and ITP53 even shows a column of ultra-cold water bubbling up to near the surface, while ITP41 seem to show the opposite : warm water from the 20-75 meter layer sucked down to 200 meters.

Even down to 500 meter, where the really warm, really stratified water hangs out, both ITP41 and 53 show some disturbances in the past day.

Remember that Ekman pumping works best when the low kind of stays in the same place for a while. So, now that our storm is not moving around much any more, it will be VERY interesting to see what happens with this deep water in the next few days.

Many, many thanks to Woods Hole Oceanographic Institute for keeping these ITPs alive and presenting the data for everyone to see. We would be virtually blind without them.

Seke Rob

Re Oslo | August 09, 2012 at 00:25

The new record was to be expected for July with June already having fallen through the floor. The one piece having particular interest was the NA and NA minus Greenland

July-2012 2.17 0.20 1.97

i.e Greenland per Rutgers had 1.97 Million Km^2 snow cover on average in July.

The weekly looks like this,

1 2012 31 2.07 0.12 1.95 0.04 (1.91)
2 2012 30 2.15 0.12 2.03 0.08 (1.95)
3 2012 29 2.31 0.12 2.19 0.24 (1.95)
4 2012 28 2.35 0.12 2.23 0.24 (1.99)
5 2012 27 2.40 0.18 2.23 0.24 (1.99)

3rd last column less 2nd last is Greenland net. Think with what we've seen, the question remains: How much first year snow is still on top? (If that were their definition of snow cover). This is what they write on their site: http://climate.rutgers.edu/snowcover/docs.php?target=ssmi

4. A mask is applied for land areas. Due to the complex microwave signature of the Greenland ice sheet, areas with exposed glacial ice or ice lenses and layers do not show up as snow covered. A mask is applied to show the Greenland ice sheet as snow covered (with a depth of 999cm). Snow depths are set to -90cm over the ocean.

Interpretation: If there's more than 999cm of snow/ice depth, it's considered snow covered on [Green]land.

Espen Olsen

Jøkelbugt North East Greenland:

The last remaining grounded fast ice in Jøkelbugt seems to loosen up now:


Artful Dodger

CT Arctic SIA data point is out for Aug 7, 2012:

2012.6000 -2.1992197 3.3401906 5.5394106

The anomaly is down to -2.199, and the daily decrease in SIA is -85k km^2. CAP2E for 06/07 Aug stands and 58.72%, and Solargain is 108.10% of the June 20, 2012 value.

Artful Dodger

IJIS has updated. The revised SIE for Aug 8 is 5,585,313 km^2. There are no revisions to previous data points. 2012 is -63,750 km^2 below the Aug 8, 2007 value.

IJIS SIE has decreased -852,343 km^2 in the six reports from Aug 3 to Aug 8. With the first half the Aug 8/9 data point already known (Aug 8 daily decrease must be in the range -170k to -175k), we are likely to see the first ever week of mega-melt: 1 Million km^2 in a 7 day period.

Aug 9 needs to decrease by about -126k for the first MegaWeek to enter the record books.


If "Denial is a river in Greenland", what is really happening?


Goddard is a rotten island calved from the Watts glacier that drains the denial cap of lala-land.


Artful Dodger

Yes Rob, the 34.20 psu halocline does seem to jump to the surface for itp41 in the past few days. Ominous for Fall freeze-up. That's bad news for the Bears...

Erik W. Kolstad

Hi guys, great to see so much interest in this storm. I did my PhD in meteorology on extreme weather in the Arctic, mainly focusing on polar lows (or Arctic hurricanes, as they were once called by Kerry Emanuel and others). I just wanted to briefly comment on the probability for Arctic storms when the sea ice retreats.

First of all, polar lows are small-scale storms that feed on the energy being released by the (relatively) warm ocean when very cold air masses emigrate south from the Arctic sea ice cap. In this sense they're similar to tropical cyclones. The storm that we're seeing now is not a polar low. They only occur from autumn to spring (in the polar night), because that's the only part of the year that the air gets cold enough for the enormous heat fluxes from the ocean to the atmosphere that drive polar lows to occur. Also, they move south, away from the ice, embedded in cold air outbreaks. The current storm moved in from Russia and merged with an existing low over the ice cap. Jeff Masters pointed out that the low is stacked with an upper low, and this "barotropic" structure is usually a sign that the low will weaken quite soon.

Generally speaking, it is likely that Arctic storms will become more frequent as the sea ice retreats. This will probably be more noticeable in late autumn and early spring, but maybe also during winter. The reason for this is that the waters in regions that were previously covered by sea ice will become exposed and thereby open up reservoirs of energy to be used as fuel for the storms. I've looked at this effect for polar lows, and in a study that I did with Tom Bracegirdle of the British Antarctic Survey, we concluded that the breeding grounds of polar lows will emigrate further north, following the retreating sea ice. Other kinds of storms will probably also be able to penetrate further north than they currently do, following their energy source, the open water.

Although storms are predominantly a winter phenomenon, retreating sea ice may lead to more frequent Arctic storms in the summer as well. This is because most storms (tropical cyclones are one notable exception) feed on horizontal temperature gradients. The temperature difference between the Arctic and the tropics in the dark season is what generates the jet stream and the large, synoptic lows that can grow so powerful from autumn to spring. And as the sea ice retreats, the local temperature gradients will increase and probably raise the risk of summer storms.

I hope this makes sense.


Erik Kolstad
Bergen, Norway

Artful Dodger

Neven, sorry to inform you that due to known issues with the Typepad web-server, the links to comments on this tread contained in your article on Climate Progress are now broken.

The way Typepad works, it can not find links to comments on any "Previous" page, only the last one.

Since this post is so important for our archives here at the ASI blog, may I suggest you consider closing this post for comments, and then deleting/combining a few comments to ensure that Typepad does not create a second page?

You are welcome to delete any of my comments if it's convenient, especially the phuny ones...


Thanks a lot for a great comment, Erik! To others: Erik's blog, Polar Lows, can be found here.

Lodger, I haven't noticed any issues with web servers (except that the web server of my host has completely crashed, but that has nothing to do with this), but I see what you mean. I'll delete a couple of comments.

In the meantime, everyone can continue in the unofficial 4th Arctic storm update: Further detachment

The comments to this entry are closed.