« ASI 2013 update 1: a slow start | Main | New map on the block »


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


This event reminds us to use the full panoply of satellite sensors -- the cloud-penetrating views of microwave is nicely supplemented by cloud top views in visible and near-infrared and cloud top temperatures in farther infrared.

The first photo is not online -- it shows the cyclone as seen by compositing the three infrared sensors of the AVHRR satellite as a false color (R,G,B) = channels (2,4,3) = (0.92µ,10.8µ,3.74). The second shows how the visible and two near-IR channel composite on Modis Terra is becoming useful.

The second group of microwave images shows more progression of water signature today. The faint bullseye radial streaking pattern in Jaxa color reminds me of my windshield after a piece of gravel cracked it two week's back. Note the cyclone has set the goat's head region in motion again, Fram-ward, after a long stall. The strong temporal-geospatial correlation at the sea ice surface with cyclonic features is only to be expected.

 photo NavyCloudTops_zps24a75626.jpg
 photo navyMicro_zps2e2ef67f.png


Yep. The Lance-Modis shot did it for me. That persistent little cyclone is having an impact. And if those ECMWF forecasts are correct, we could be in for another major event.

Definitely something to keep tabs on.


Just want to add a few observations about the jet. It currently appears to be stuck. The cold air outfluxes remain in the same regions. And the warm air outfluxes appear to be mostly stationary as well. This hugely impacted Europe this winter and is keeping dryness/drought heavily entrenched in the western US even as the central US has switched back to flood/storm.

The ECMWF model forecasts have been much more mobile than the actual weather. So we'll see how much this thing actually moves.

One more point to make -- any major kick of heat and/or moisture will pump this storm up. Not to be a doomer. But something to watch out for.


It is called the polar vortex.

It is a perennial cyclone, usually confined within the polar region. It traps cold air within the polar region.

If and when it loosens up, it releases cold air towards lower latitudes, letting warmer air into the Arctic basin, usually from Siberian Russia.

This year it seems to hold steady (so far), meaning the cold air gets trapped within the polar basin, meaning less melting.

The thinning of ice seen directly below the very eye of the arctic cyclone is caused by the enormous forces of the cyclone moving ice out from the center, towards the edges, in all directions.

If the temperature is cold enough in the center of the cyclone, new ice will constantly be made to replace the ice being moved out.

It is an Ice Maker.


Ostepop1000 - forgive me for my skepticism about your particular assertion - but from my limited education in this, that tends to be a winter, rather than summer phenomenon.

I'd further say that that to make ice, its missing another key element - a heat sink. All the heat currently pouring into the arctic needs somewhere to go. The formation of ice *releases* heat, so that would push the heat engine in the wrong direction.

Perhaps I should stop here, taking various people's advice and DNFTT.


Well, Jdallen it is happening right now, as we speak.

The polar vortex (polar cyclone) traps the heat inside the polar basin.

As we have seen in the past - whenever there is a weakening of the polar vortex, a so-called polar outbreak, cold polar air flows our of the arctic basin, usually down through Scandinavia.

And nature hates a vacuum, so warmer air flows into the polar basin, usually from Siberia.

These are not rules, but examples.

My former point was that as long as the polar vortex (arctic cyclone) stays strong within the polar basin, we will have only a moderate melting seasson.

I already predicted 5.5 in extent in the prediction thread.

And I stand by it.


It's not a perennial cyclone, but a temporary one, and it's too small to be the polar vortex. The polar vortex is no longer with us in any significant way, but it will return in winter, of course. It may well be an ice maker, but not now, as I don't think leads are icing over again in June.


Perrennial or not is a question of definition.

The fact is that we have a strong arctic cyclone at the moment, and as long as it is active, and as long as it keeps the cold air within the arctic basin, it will prevent major melting of the sea ice, even create more ice.

Until the cyclone breaks up and lets cold air into the arctic basin.


Until the cyclone lets up and lets warmer air into the Arctic basin, of course.

Sorry about that.


There are always more than one thing happening in the Arctic. The cyclone might be keeping cold air in the centre of the basin, but at the same time it will cause the ice floes below to break up in ever smaller pieces and create more open water between those floes. And who knows how much Ekman Pumping is bringing up warmer water from below the halocline. At this time of the year that won't create more ice.

Usually the ice pack melts at the edges inwards. If this cyclone causes a zone of less concentration well into the ice pack, with June and July still coming up, we'd be seeing something new. Although we did see something similar with the Laptev Bite in recent years. But this is another mechanism.

And like you say this cyclone won't last forever (if it indeed comes back, as the forecast says it will atm), so the warm air and insolation are bound to do their thing sooner or later.

We'll see what happens, Ostepop.


It's not the polar vortex. And it's not really that strong at the moment.

For it to be the polar vortex, we'd have to have a viable circumpolar jet stream to support it. We don't.

As for the 'generating ice' assertion, we'd need to have a lot more cold and a lot less heat for that. At best, the storm at this time would be ice-neutral. But the upwelling its generating is having a deleterious impact on the ice -- which we see both in the models and in the visible.

For you to be correct, we would have to see ice thickening in the region of the storm and an expansion of thicker, more resilient ice at its periphery. What we have is the opposite.

So I've done my part to FTT. Won't be adding any more, though.


@Robertscribbler - as you're pointing out, the "500 lb gorilla" of heat input here isn't the atmosphere, but rather the ocean, because it encompasses insolation collected over a much larger region, and transports it into the Arctic basin. The troposphere makes only a small energy contribution to the melt, and unit for unit, has 750 times less carrying capacity for heat, regardless of what direction it is moving in.

While the cyclone and its clouds may interfere with the other heavyweight - sunlight - you rightly point out that barriers to the big one - haloclines and thermoclines - are being removed.


We will see what happens. Robert.

personally, I put more faith in temperatures, pressure and ice flow.

No "upwelling" for me.


@ jdallen

Exactly. Water packs a very high energy/heat content. And the one-two punch of this storm involves first stirring heat from the ocean via the mechanisms you describe and then, during summer, drawing in the clear skies and warmer temps in their wake to allow the sunlight to do its work.

This storm is forecast to contain both elements. So if, as Neven notes, the forecast is real, then we have something that could be quite a powerful influence on the Arctic this summer.

Is there a precedent for having the central ice hollowed out so early? Because that's what we're looking at here, should the forecast come to pass.


@ Ostepop

Retreating from your 'polar vortex' hypothesis?

If you are interested in looking at the history of this particular cyclone, you'll note that it tread a path through the Beaufort where it churned and thinned the ice before making its presence known more dramatically near the North Pole in the current event.

As for waiting to see what happens, it has already happened. It's just a question of whether the situation grows worse than it already is.

As for upwelling... You can use a technical term for it, as Neven and Dodger here have: Ekman Pumping. For my part, I tend to lean toward the non-technical so my posts will be more generally readable at first blush. Not to say discussions of Ekman Pumping aren't a good idea. Quite the opposite. One of the strengths of this blog is its expansion of scientific literacy.

That said, you should take a look at the impacts of cyclones in general on the ocean, if you're honestly interested. Hurricanes, for example churn the waters of tropical and temperate seas, leaving a swath of cooler air and water in their wake. They pull water up from the ocean depths through both churning wave action and larger Ekman type processes. This cooler water results in cooling where these cyclones pass.

In the tropics, the ocean surface is hot, the depths cold. In the Arctic, the opposite is true. The surface is the coldest region and warmth, through various processes, lies sequestered beneath. So cyclonic churning in the Arctic has the opposite effect, pulling warmer waters up from the depths.

It's not as if this observation, on my part, is a stunning new revelation, as your statement would seem to imply. Upwelling is a standard event that impacts ice. North Baffin Bay is one example. And wind-produced upwelling events have resulted in open water areas forming in the Arctic even in winter.

So was your placement of upwelling in quotations simply to imply derision for a common property of ocean water movement? I don't think upwelling is offended. But, if you hurt its feelings, perhaps you should apologize?


We'll see what happens, Neven.


To the extent that "upwelling" is an entity, or even a strictly mechanical process, even remotely capable of being offended, perhaps I should have been more sensitive, and so apologize.

And if so, I do.

If not...

I do not.

Lou Grinzo

I happened to notice this coming (possible) event in the sea ice thickness animation Neven posted just before stumbling over this post.

Without getting into all the back and forth over vortices and upwellings and such, I think it's worth considering that whatever happens this year we're probably not far from another paradigm shift in the Arctic. Specifically, I would expect that with the continued thinning of the ice and warming of the ocean that huge melt zones far from the edge will become fairly common events, which will only pave the way to virtually complete melt outs in late summers, i.e. Blue Arctic Events.

Pete Williamson

Maybe this paper is describing what you think might come to pass



Enough of this silly troll -- time to block and delete his posts, all of them.

Here is ice movement from Jaxa animated over the last seven days. In addition to massive ice thinning under the eye, the cyclone is noticably rotating the ice pack in the clockwise direction. The ice is not strong enough to keep up, causing it to separate in the radial direction

Not all of the pack is rotating. In fact, the multi-year ice (what is left of it) has gotten caught up between the gear wheel and the CAA and is getting pushed out towards the Fram. A few more days of this ...

 photo navyRotation2b_zpsa27ae441.gif


This might be useful as a simple illustration:


@Ostepop - the presence of an Ekman pump inside of the cyclone is not a question of fact, it is a question of scale and effect.

I've been watching Bremen and other concentration maps closely for a couple of months, and one thing I have noticed is a trend, running from the Fram across to the central Siberian coast. Across this swath of the Arctic, over this time, I have seen consistent prediction (and observed) of ice fragmentation and the creation of leads. I was not sure how the heat required for this was getting to the region over time. While I do know this approximates the flow at depth of part of the relict North Atlantic drift into the Arctic Basin, I had no sense of how the energy could reach the ice quickly enough to produce the effect being reported.

So, I had been suspicious of what I had been seeing, thinking it possibly some artifact of modelling or misinterpretation. But now understanding (roughly) the Ekman mechanic, I have a source for the heat, which could produce effects commensurate with what I was seeing in the combination of models and observed data.

So, as you said to Neven, we shall see; but I think by dismissing upwelling you introduce a flaw in your analysis.



*Wow!* that radial fracturing is both distinct and dramatic. I wonder if we could somehow quantify the expansion of the spiral?

It would seem to me that similar radial effects would be seen in near-surface water flow, and imply the need for substantial replacement water volume from depth.



I do not discount upwelling, I just think it is a fudge factor.

Let's work with what we have. It is too easy to run to the hills.

Bob Bingham

I hope you don't mind me repeating myself but I worry that the loss of Arctic sea ice could be the first fast acting tipping point that could have serious affects to our weather.If we had a quick (quicker) break up of the Arctic ice, say down to 1 mill sq Km in five years it would seriously hamper food production in the North.

Hans Gunnstaddar

Robert Scribbler wrote: "In the tropics, the ocean surface is hot, the depths cold. In the Arctic, the opposite is true. The surface is the coldest region and warmth, through various processes, lies sequestered beneath. So cyclonic churning in the Arctic has the opposite effect, pulling warmer waters up from the depths."

It's pretty hard to argue against that based on the time lapsed visuals. It would seem in laymen's terms from a standpoint of any increase in ice minimum from 2012; that once the ice has been formed in winter, the less the ice is churned in Summer the better. The opposite, cyclones churning ice and upwelling warmer waters seems consistent with greater and quicker ice melt.

Ostepop's diagnosis will either be vindicated or negated, simple enough, but in any case the discourse provides an opportunity to learn. It's a fascinating, cutting edge situation in the Arctic that is holding people's attention like a suspenseful movie, waiting to see what happens next!

Frankd 1977

Thanks A-Team for the Japanese perspective. If a picture says a thousand words, then that animation says a million.

Having said that, the Neven Acropolis Memorial Polynya is showing at 140 E, 84 N on ASRM2 for May 30th. http://www.iup.uni-bremen.de:8084/amsr2/arctic_AMSR2_nic.png



@Bob Bingham:

"If we had a quick (quicker) break up of the Arctic ice, say down to 1 mill sq Km in five years it would seriously hamper food production in the North.

Not to get off topic but it might be a good idea for Neven to start a post on the consequences of the Arctic meltdown on our crops and its social impacts.


@ A-Team

Stunning visual of the storm in progress. The ice-driving effect is tremendous!

Just one point. Persistence can often make up for strength in a storm. One example of this is fetch -- where wind over long periods of time and wide expanses of water can drive as much water as a much stronger wind over smaller areas and shorter periods of time.

With this storm persisting for so long and predicted to continue, I wonder if we won't see a bit of compounding effect? And the storm re-emerging at 980 mb on June 9 in the same region is a bit troubling. A stronger repeat for an already impacted region.

Motion toward the Fram, as your graphic shows, seems a bit ominous. Wasn't it 2010 that saw a large degree of ice thinning due to summer Fram export?


The water contacts at least as much of the ice as air and it is not static. I think ignoring its ability to transfer energy is ignoring a major part of what can cause melt. Air temps, sunlight/insolation, contact with water -- now there's a more complete picture of your melt/freeze balance equation. And storms ramp up the churning process which can have larger impacts on thinner ice.

For my part, I don't think the melt-freeze balance will result in ice growth as Ostepop has asserted. Further, weakening ice and churning, I believe will likely have a deleterious effect. At the best, central regions will vainly to re-freeze as temps rise through June, while the churned fringe is more vulnerable to typical summer conditions.

It would take a pretty cold June to support Ostepop's conclusion. So I guess he can look forward to that, should it arise.

Shared Humanity


That visual seems to suggest the demise of the goat's head. Fractures are appearing in it.

Jai Mitchell

Jennifer Francis of Rutgers postulates that arctic amplification is reducing the intensity of the polar jet as well as pushing ridges further north.

It looks like a cut off low in Iceland is continuously pushing subtropical moisture up into the basin and feeding the vortex. The number of "cut-off lows" this year have been outrageously high.

The Jetstream seems to have weakened enough to enable significant vapor transport.


"Alternatively, decreased meridional and/or vertical temperature gradients would conspire to reduce rainout of subtropical water vapour by synoptic eddies22, decrease isotopic distillation during vapour transport, and lead to D-enriched precipitation at high latitudes."

caveat: I study but have not formally studied.

Jai Mitchell

sorry, meant to include this 5 minute explanation from Dr. Francis.


for a formal presentation see:


R. Gates

Related to the Arctic vortex, of course Neven is correct and the vortex is a winter phenomenon and a stratospheric phenomenon as well. The current surface cyclonic winds around the surface low are actually blowing in the opposite direction to the upper level summer winds in the stratosphere. Also, this cyclone will be a high net energy advection event to the area north of 80 degrees, meaning it will on net advect more energy from lower latitudes to this region. The thinning ice in the area of strongest surface winds should allow more solar to penetrate at the time of maximum insolation. This cyclonic event, plus other divergencies I've seen in many of the basins leads me to see a mid-June "Sea Ice Extent Cliff" event as very likely, whereby we see some jaw-dropping daily declines in area and extent going on for a period of weeks.


@Robert - agree with you mostly, but disagree as regards the thermal transfer from air to ice as a major factor in the melt. Air's thermal conductivity is extremely low, so in the absence of other sources of heat, ice in air close to 0c, assuming net insolation of 400watts/meter, will receive far less than 1/1000th of the heat from air as it does from sunshine.



I have put the 3.125km png's in

Latest is:


I have noticed that there is a pattern of updating. I fully expect the 20130530 file to be updated today again. Sometimes some much older files get updates. I will try to catch the updates of the recent files.

Google sites has limited space, I will remove the files as needed.


A-Team, I'm adding your animation to the post. Thanks.

Not to get off topic but it might be a good idea for Neven to start a post on the consequences of the Arctic meltdown on our crops and its social impacts.

I wrote about it last year (see first post in the Best of Blog section at the top of the right hand section), and will probably write about it again in August/September. In the meantime there's this section of the ASIF that deals with consequences.

John Christensen


Does ACNFS provide any 'actuals' that would allow us to see how close we are today, 5/31, compared to projected thinning?

Darren Wood

Hi, I have an ex-bootneck friend of mine reporting 23 degrees C in Tromso Norway today

John Christensen

ACNFS seems to have some areas on 5/29 down at 40% ice coverage, while CT for this date has it lowered to about 80%, based on my terrible judgment of the colors. Is ACNFS 'ahead' of CT in some way, in disagreement, or could cloud cover make the CT image less reliable?


Good questions, John. I usually dismiss ACNFS as a reliable source (they overestimate thickness for instance), but in this case I thought there could be more to it, especially in combination with the ECMWF forecast.

But as far as I'm concerned, it will only count as soon as the Uni Bremen sea ice concentration map (CT isn't as sharp, IMO) is consistently showing lower concentration in the region of interest. A bit of yellow (and a tiny bit of green) in the region of interest showed up on the UB SIC map today, but it's too early to tell if it's the sensor fooled by clouds or some other artifact, or whether it is seeing the open water between floes we're seeing on the LANCE-MODIS satellite images.

Give it 2 to 4 more days, and we'll start seeing the effect, if it's there.

Remko Kampen

Totally blocked circulation pattern. What is most remarkable is AGAIN the astounding resilience of the this time Scandinavian blocking high.

Murmansk/Kola region is busting June records in May and already at 28° C now for another bust.
Absolute Finland May record is 31.0°, some places right now past 30 so it might fall.


More than that!

On 6th June three of the four models I checked had the Azores High extending all the way to Tromsø - it may open up for advection of Carribean air to Northeast Greenland and Svalbard...

Remko Kampen

P-maker, Monday the T850 will go up to +10° C over Svalbard. That means +10° at about 1450 meters above sea level. Or +7° C on the highest peaks. This wave of hot air will move on though cooling a little to the NP and some of it will thaw a little on the NE Greenland coast. To be followed by another wave by midweek.


Wipneus - your high res pictures are much better than all the other pictures out there. The radial spiral driven by the cyclone and even the cracks at its outer boundary are clearly visible: It is an excentric wheel both churning out the MYI via Fram and removing ice from its center.

Thank you for that wonderful product. Maybe you want to ask the poeple in Hamburg for some money since you did some of their work allready ;-) Normaly they have some money in projects for "contract for services".

John Christensen

Regarding the blocking high in Europe:
-400K KM2 in Arctic SIA does not seem like a lot to me and as we know, we have a number of years with lower SIA for April and May than in 2013.

Is it the lack of ice in Barents in particular or high SST that would cause this blocking high? Or are we just running a higher risk of having these blocking highs with lower Arctic SIA in general, and weather caused it to happen this year, even if SIA is slightly higher than in preceeding years?

Remko Kampen

"Is it the lack of ice in Barents in particular or high SST that would cause this blocking high?"

Differential analysis read a look at previous blockings in the region will show that this question is almost intractable.
What does seem to surface is a propensity for circulation patterns including blockings to last (much) longer. So, to illustrate, if a block causes a heatwave somewhere, it lasts 4 weeks instead of one or two. Likewise for cool trough situations.
Blockings also seem to become more pronounced.
Both might be related to a general decrease in tropic-arctic temp difference.
Primer, don't neglect the magic word 'quasiresonances': http://www.wunderground.com/blog/JeffMasters/comment.html?entrynum=2366

Wipneus - your high res pictures are much better than all the other pictures out there.

Let me echo that, Wipneus. Great stuff.

I'll crop those and turn them into an animation tomorrow or the day after that, depending on the level of divergence.

Is it the lack of ice in Barents in particular or high SST that would cause this blocking high?

Jennifer Francis says that in Spring and Summer snow cover anomalies play a larger role than sea ice.

Here's the negative anomaly (red) for May 30th from the Rutgers University Snow Lab:

John Christensen

Thanks all - very appreciated!

Account Deleted

So, if I understood, storm winds can literally grind vulnerable ice and so favor the heat transfer from the ocean. I find it really amazing. Even a cold storm may eventually cause a net heat increase of the Arctic ice by its mechanical effect.

At this time of the year the grinding effect may be what really matters, not the cold temperatures that the storm brings about.

@ Pete Williamson thanks for the link to that paper

Veli Kallio

Far East Shipping Company (FESCO) are contracted to carry out sea ice demolition projects along Russian coasts "ice chipping". The polar cyclone in summer time conditions with sunlight and heat is a natural sea ice demolition project. In autumn darkness with little light and heat it is sea ice creation project, much like Antarctic ice whipped by winds at the moment to cause sea ice scattering that seeds ice into open waters to stabilise it an cause ice formation. In the present conditions the polar cyclone has adequate sunlight and heat input to cause sea ice melting, in fact, on the pole the sun light peaks at this time of year. Any major helping hand is just the other way round and makes sea ice losses only greater. It is also true that vertical mixing of sea water also increases in the leads as ice breaks up. So there is no sea ice seeding but creation, unless you look at ice now around dark Antarctica!

Aaron Lewis

A "Polar Vortex" is driven by cold, dry air. This "cyclone" is mostly driven by water vapor.

If I am correct, you will always see a leg of "jet stream" at 300 mb over the Arctic, near the top of the cyclone. That is, the segment of jet stream and the cyclone are two aspects of the same circulation, rather than the jet stream blocking surface weather.

As long as we have large patches of sea ice surrounded by water vapor, such cyclones are likely to be stable and persistent, with any highs that intrude into the space tending to be transitory.

Lack of snow at the perimeter of the circulation helps supply water vapor to drive the circulation. I strongly agree that the negative snow anomaly is part of the system generating the cyclone(s).

This summer, I do not see "blocking" so much as "stranding". The jet stream no longer moves and transports weather systems as it did in the past. If the metaphor is a conveyor belt, then with blocking, the conveyor belt is running, but the blocked weather system is not on the conveyor belt. With stranding, the conveyor belt is broke. The foundation for a working jet stream is cold, dry, and competent Arctic Sea Ice. These days that occurs only in the winter.

In coming summers, I expect regional geographic features drive regional circulation patterns (and weather) that generates segments of jet stream flows rather than a global jet stream that transports a sequence of weather systems over regional features. My view is very different from a global circulation that generates Rossby waves.

Remko Kampen

Aaron, 'stranding'. Thank you very much for this phrase. Imagine the north Atlantic jet beaching on flat flow patterns over Scandinavia and retreating by slow desintegration from the east. That happened to the very strong December jet per January this year.
The blocking anticyclone was remarkably weak, especially at ground level. But it proved very robust, resulting in a lengthy run of sub-zero days in Holland. A gray hair for some meteorologists expecting thaw anytime within two, three days but the cold persisted two weeks.

Remko Kampen

As for your scenario, it is very much like mine.
Developed it this day: http://www.wetterzentrale.de/pics/archive/ra/2006/Rrea00120060720.gif .
By far the hottest month in Holland since at least 1706.

Not saying summer this year will be so. But saying 'patternless' patterns like this wil will occur more often during summer up to December as the Arctic sea ice disappears.


As we watch this season progress, I would encourage caution in interpreting the processed information. The rules developed in earlier times that helped stabilize the data may now mislead us.

For example: The rules about ice coverage versus ocean area may serve to increase both the reported ice area and ice extent temporarily and artificially. This most recent cyclone perhaps serves as an example. As the cyclone tore up the ice and opened leads, it also expanded the ice cover. The total ice area likely did not increase and may have decreased. But by spreading the open ocean area across cells, where each does not reach the threshold to assess the cell as ice free, more cells are assessed as ice. IARC-JAXA currently shows a slight bump up in area and extent that may be due to this effect.

The reality is that there is now more edge area and open water in the central arctic. This feeds and drives several potential positive feedbacks (edge melting, depth mixing, wave movement driving mixing, wave breakage of ice, wave action over edges, cyclone formation and associated pressure and wind disruption of ice, etc...

Likewise, other tools (good as they are) that have lower resolution don't see the open leads at all. (e.g. Cryosphere today imagery).

The work by all of you here is vastly better. The discussion and insights here are astounding. Keep up the great work. Even more than this, the interaction is leading all of us to new insights with the cross fertilization of ideas and the new presentations of data in extraordinary graphics. Great stuff.

I think the ice thickness and volume tools serve us much better now in assessing what is happening on a day to day basis.

With that in mind, do you all have thoughts on how to improve the measurement tools to discriminate what it actually happening better?

This is likely irrelevant in the medium to long term, as the ice is going, going - and all too soon - gone. But in the very near term, it may help in understanding what is happening and telling the story.

It may also help in understanding the physics and mechanisms.

For the balance of this season, I now expect the melt out will be greater than I thought before. It appears that we will see the rapid loss of the ice cover over most of the central arctic in June-July. Should we get another (perhaps stronger) cyclone (which seems likely), we may see mass advection or transport of ice through the Fram straight including the bulk of the MYI.

Along these lines, I tried looking at the ice thickness and volume (thanks Chris) broken into thicknesses across time. As brain dead as Excel is for most graphics and science purposes, its 3D graphs work well for this. Plotting each thickness as a separate depth row against the time series makes it much clearer when we are seeing step changes in behavior. What that most clearly shows is the loss of the thickest ices, and a relative minor increase in thin ice vulnerable to sudden melting.

Chris Reynolds

Looking at Wipneus's images, it's like someones pinched a white table cloth and twisted.

Chris Reynolds

Pete Williamson,

Thanks for bringing up the Asplin paper.

What it is talking about is long fetch of open water late in the melt season allowing larger amplitude waves which then can penetrate deep into the pack, as in 2009 when waves broke up the central core of perennial ice.

So it may seem like Asplin et al isn't relevant to the situation now, because the mass of ice will act as an inertial damper on wave formation.


Asplin et al observe that fractured ice has more area at the edges of smaller floes from which melt can act. In churning the pack, this storm in the run up to peak insolation may be creating perfect conditions with smaller floes surrounded by dark water, to allow the ice albedo effect to play a more active role.

Whether or not this storm has such an impact, we'll see. But it is worth noting the reduced thickness in a swath from Laptev into the pack.

Neven has expressed doubts about HYCOM. I'm not sure what these are specifically. My main reservation about this system is that it seems in the past to have modelled thicker ice off the CAA and North Greenland than PIOMAS does. This year has shown that thermodynamic growth is well constrained by physics, with ice thickness being up to around 2m in what has been a fairly typical recent freeze season for ice growth.

Personally I have no doubt that what HYCOM shows is really going on. I am less sure about the role over this melt season. But if I had to put money on it, I'd bet we'll see a rapid growth of open water from Laptev by late July following the form of this impact.

Chris Reynolds

You can see what I mean by a larger periphery and smaller floes from MODIS.

Day 145, then five days later Day 150.



Depends both on air temp and how heavily moisture laden the air becomes. I'd take a hot fog over direct sunlight for its ice melting capacity any day of the week.

The issue, for me, is not to focus on just one condition that may impact ice, but to try to be aware of as many of them as possible.

R. Gates

Not doubt most of you know about these near real-time polar wind charts, but for those who don't you can check out the wind speeds and directions at various pressure levels almost real-time over the north pole here:




The best explanation for the persistence of these blocking patterns has been provided by Jen Francis. In short, sea ice erosion slows down and disrupts the jet. We end up with large meridonal wave patterns (blocking patterns) that keep regions locked in the same weather conditions for long periods. I'd say the current pattern is what's causing both the rather high observed water and air temps in Scandinavia. That and a feeding flow of heat and moisture moving up into our cyclone from this region.


Not doubt most of you know about these near real-time polar wind charts

I didn't. I've added it to the ASIG links at the top. Thanks.

Neven has expressed doubts about HYCOM. I'm not sure what these are specifically.

I find it very useful for comparing between years, but I'm not too sure about the forecasts.

First of all because of the overestimation of thickness, which I believe NRL's researcher Pamela Posey was the first to point to back in 2011.

And maybe it was because I had gotten so used to PIPS, but I've never managed to make use of ACNFS' drift forecast. I primarily use UB for SIC, although a forecast could come in handy. But to be honest, I never check out the ACNFS forecast.

I do find their maps visually appealing, and of course I'm grateful for all the work they're putting into their product. I'm just never sure how serious the Navy itself takes it.


Plus... Neven's addition RE snow cover...

These 10c forecasts for Svalbard are interesting. Looks like 5c temps plunge pretty deep into the central Arctic in some of these model runs.

Core cold air and external heat are major drivers for these Arctic cyclones. So the more surrounding warmer air, the more their potential to develop/strengthen. They also tend to pull that warmer air in behind them. So if you get a pretty strong one in the central Arctic...


question @ Wipneus or A-Team

I just tried some calculation using gimp - but I am not familiar to that (only Fortran...), so I am not sure what I got. From the picture Arc_20130530 I got an SIA of 10.97 Mio sq km (all ice) - that is a bit more than from cryosphere. OK - I got that hole in the middle and one extra lake in canada and some white pixels from the shore, that could explain the error.

So my question is just to be sure: Is the pixel value linear to area in the pixel and are really all pixel 3.125^2 sq km or is there something to be corrected?

Sorry for asking that stupid things here - but maybe it could help someone else to use these new pictures for something more useful, too.



We will be getting there eventually. Without a jet stream, we will end up in a situation, which I have previously (http://neven1.typepad.com/blog/2012/10/naive-predictions-of-2013-sea-ice/comments/page/1/#comments at 11:50 on 11 Oct 2012 ) alluded to as the “Absence of Weather”.

When we end up in a situation with hardly any weather systems being moved around, we may have to revert to, what I would call “Basic Climate”.

Your fine example from July 2006 shows just that. A European weather map with hardly any isobars left.

“Basic Climate” will be defined by classical features such as, 1) a strong diurnal cycle with convection during the day and calm nights, 2) Land-sea breezes will dominate the coastal zone, 3) Katabatic winds will dominate the hilly areas during the autumn and winter, 4) Monsoon systems will be dominating our summers – even in Europe during spring and summer.

The true essence of climate: - gradients from cold to warm surfaces – will play havoc with our ability to survive as a species.

Examples of this new pattern are plentiful over the past few years. Even if we look ten days ahead (see ECMWF forecast for Jun 10 here: http://www.wetterzentrale.de/topkarten/fsecmeur.html ), you will only see one (1) single isobar over the entire European continent.

So, it is time to study “Basic Climate” processes in earnest, if we want to grasp the future of our planet.


and why are there significant more pixel with value 128? (I removed the brown...)


After correcting the satellite-hole (by using values surrounding it) and ignoring lakes and coastal white points the sea ice are from integrating Wipneaus high res picture Arc_20130530 is:
10.92 Mio sqkm (assuming pixel values linear proportional to area and pixel area constant).

If someone would confirm that we could use the pictures easily e.g. to investigate the SIA-change produced by the cyclone in the basin in the next days.


I remember from discussion of GAC-2012 that what gives storms energy is temperature differences. In the case of hurricanes you have a hot interior on a cold exterior. What kills it is the upwelling of deep cold water into the interior that cools it off and therefore kills the engine.
In the case of the Arctic you have the opposite. A cold interior and a hot exterior.If the cyclone were to stay over solid ice the inflow into the core would be the same air as around it and therefore kills the engine. As long as it can say over water (either open water or creating open water by opening up leads) and a warm air mass that it can pull in from, you can have a cyclone that can last for a very long time.
In the case of right now, you have a cyclone that is able to break open the ice to get at the cold temps from the water (have no clue as to have much open water it needs, but the more it is the longer it will last), and as those from Scandinavia are reporting record breaking temps there.
There you have you engine. Cold temps coming from open water (the rise in temps will not be enough to offset air temps) and heat coming from Scandinavia and I would also say possibly from NA as we are in a somewhat of an early heat wave.
Although no expert, I would think that the more the jet stream gyrates the more heat can get into the Arctic the greater possibility a long lasting cyclone can occur.



Temp difference between cold and hot + moisture levels drives it. It'll tend to be coldest where ice is thickest (the ocean water is warmer than the ice). With the meridonal flow, you have hot air beside cold air, beside hot. Plus you have lots of cold air and moisture laden clouds all over the Arctic. These conditions would seem favorable for storm development.

@ P-Maker

Well, you'll not have a push to that degree of atmospheric latency. It's really just a transition period. Eventually melt speeds up in Greenland which somewhat reverses polar amplification by dumping freezing or near-freezing water into the Atlantic and Arctic Oceans. At the same time temps pump higher in the tropics. Then you end up with some crazy storms and temp differentials.


MODIS shows ice circumstances on the Sib side of the Pole to be very transient. 50 km2 wide openings deform overnight, only popping up in other forms tens of km’s away. It all shows that there’s much ‘space’ in the pack.

Comparing ’12 and ’13…at a glance, the Beaufort and CAB north of it looked worse then. The CAA and Laptev showed more blues. The Kara and Barentsz were much more ‘cleared’. Baffin Bay was further in it’s decay.

But it’s the FYI in the CAB that’s looking set to perish… worse than ’12.
IJIS and CT are deceptive…no offense, but these indicators are missing the unusual trip. Much like Sam described above at 19:46.

But what does one expect with CO2 trailing above 400 ppm for five consecutive days last week?.


Forgot to mention that there seems still to be some cold left to create a nilas film in these overnight polynia's.
You could call that 'ice formation...'.

dominik lenné

With all this westerly highs we might approach another appalling summer with northerly winds in western/central europe - brrr - don't like climate change if it turns out this way.


The best cutting edge latest discoveries dictates 3 things:

1- Present Cyclonic circulation stems from adiabatic nature of the warmer sea ice right under a colder atmosphere.

2- In no way is this a "ice maker", accretion is finished. Even while using Lebedev the accretion would be so little. In addition mass buoys are indicating stable thicknesses. Which are not consistent in density throughout the ice column, there may be a a significant difference in the nature of the ice particularly from the bottom upwards.

3- Finally low clouds are the mortal enemy of sea ice, as per my latest research.

Even when the sun is relatively high. It takes a sun
at 23 degrees elevation or higher for 24 hours to make a serious continuous sun ray driven melt. This only happens around the North Pole at the solstice, therefore all eyes at the Pole or thereabouts especially when clear with a few clouds from mid to end of June, because there is 1st year ice there. . Even with open water, the flow edges are relatively stable when the sun shines high. Only warmer temperature advection (amongst other heat sources) break the apparent stalemate.



This is what I intended to do in the first place, but I got distracted by A-Teams suggestions.

Take care, the 20130530 file has changed (as I expected).

Further from a private email exchange with the Hamburg scientists a word of caution:
"These data are not officially released and are still work-in-progress."

The data should be proportional with concentration.

Don't know why 128 (50%) is overpopulated. Interesting observation though.

The grid is described here: http://nsidc.org/data/polar_stereo/ps_grids.html
My understanding is that the 3.125*3.125km2 is exact at 70 deg latitude and nearly so at other latitudes. What I don't know is what the error is exactly and whether it is accounted for in the extent/area products.

With grid size of 3.125km I don't expect the same values as with 12.5 km. We have discussed the differences between Uni Bremen's extent (6.25km) and Jaxa/IJIS (12.5) at length in this blog during season 2011.

If you want to seriously continue these calculations, I would very carefully suggest a programming language:

In R I can read the data as an array, called here sic. Dimensions are:

> dim(sic)
[1] 3584 2432

Total sea ice area:
> sum(sic) *3.125^2
[1] 11321135

Total sea ice are of all cells with cover over 15%:

> sum(sic[sic>0.15]) *3.125^2
[1] 11294299

Extent calculation:

> sum(sic>0.15) *3.125^2
[1] 12381094


My question to you: how much where the "errors" you found for ignoring th NP hole, and how much for the coastal ice?

Espen Olsen


Fantastic map, and very useful!


Morning all.

Neven, thanks for the summary above on ‘reality’. To make more sense of the CRWS Jet stream analyses, it is instructive to overlay them on a 300Mb NCEP/NCAR Reanalyses graph.

Did that for 29 May:
 photo JetoverGeo300Mb29052013verysmall_zpsdf52d42b.jpg

Should be obvious, but reassuring to see the Geopotential heights nicely lining the black jet stream fields.

The Polar dip in the heights (wouldn’t call it a P.Vortex anymore since April), the cut-off sinks over SW Europe and the NE Pacific PDO-basin, they’re all lining up pretty linear. Creating a strange pattern of ridges centred over Northern Scandinavia and Alaska.

The jet seems to be only strong where there are large regional pressure/temp differences. It almost fades over the Atlantic, Central Asia.

If this was just a single event, it wouldn’t be interesting. But these sort of weird patterns have been over us for some time now.



You are right, a programming language is much more flexible - but that would be nerd-only. Gimp is n00b-proof (I just installed it yesterday and got the SIA without any look in a manual). And the Nerds may control gipm by scripts and write a plug-in for anything sophisticated. So gimp is a possible solution for A-Teams dream of "sea-ice investigation for everybody" and your picture-processing is the door to that dream.

The NP-hole: I draw a selection box around it and counted the black pixels via information - histogram: 2051 pixels.
I switched the lower boundary of the histogram to 1 to ignore black and got the average value of ice-concentration surrounding it: 249.7/255
So ice-are in the hole is: 249.7/255 x 3.125^2 x 2051=19,613 sq km.

To ignore the white coastal pixels and the lakes I just painted them black. So at coasts with ice the error is still there.

To get SIA I put a selection box arround everything, used the histogram function ignoring black pixel to get average pixel value and multiplicated that with number of non-black pixel and pixel area.

I think after reading a bit in the manual to get a clue about A-Teams "mask-magic" and the scripting commands it should be easy for everybody to do this automatically for any region of interest and maybe also to produce the corresponding movies automatically.


To get SIA I put a selection box arround everything, used the histogram function ignoring black pixel to get average pixel value, _divided by 255_ and multiplicated that with number of non-black pixel and pixel area.

Chris Reynolds


Interesting, I usually use the 5600m height band of 500mb GPH, but 300mb works fine, possibly better?

It's also worth plotting surface temperature (mean not anomaly) now that you have two layers in a file (assume that's how you're working).

Not sure how unusual this is though.

As an aside, using the GPH field allows wavenumbers to be counted. I make it a wavenumber of 5 at present,

Neven, thanks for the summary above on ‘reality’. To make more sense of the CRWS Jet stream analyses, it is instructive to overlay them on a 300Mb NCEP/NCAR Reanalyses graph.

It sure is instructive. Thanks, Werther. I've decided that I really have to make an effort to start understanding all of this.


It looks like the forecast has changed. The cyclone won't return to the CAB, but intensifies over the CAA for a day or two and then de-intensifies. That seems to be it for now.


The effect of the storm on SIA was not very large up to now in the Arctic Basin. In a circle of 2.761 Mio sq km the SIA dropped from 2.725 Mio km2 (05/20) to 2.680 Mio km2(05/31) - that is a loss of 46,000 km2 only. Maybe the melting there is still to come?

BTW - the number of pixels in the dark circle varies in the different pictures from Wipneus. A most likely value is between 2400-2500 pixels at a value of 37.

The surplus number of pixel with value 127-128 is zero in the Basin - those are probably artifacts from clouds or coastal pixel further south.



NSIDC has grid cell area's on http://nsidc.org/data/polar_stereo/tools_geo_pixel.htm
From psn06area_v3.dat a 6.25km grid:

Horizontal is the 6.25^2 nominal area. Real cell area's can differ quite a bit and should be used.


I see, cell area is dependent on pixel number and should be used.

I found a function "layer mode “Multiply”" which could do the trick before numerical integration via histogramm mean. But we would need a picture with the correction value as function of x and y (or in one color of the RGB-picture). I did not find a methode to use a function of x and y in gimp yet - really a strange thing for a programmer ;-)

But there is still a lot to read and they call everything so differently (no math only coloring-words) - so there is still some hope.


Both NSIDC and Uni Hamburg document the area/extent calculation using the real grid cell area. But I just remembered the fact that Jaxa/IJIs values are all multiples of 12.5^2. That means they don't. It should be visible in the annual extent curves.


I think it is OK to work wit the constant grid area - since it is a systematic error < 6% it will not increase by substraction like a statistical error. So the error in the SIA decrease of 46,000 sq km due to the storm is 2-3,000 sq km and quite similar to the dititalisation error of 1,000 sq km. Other errors like may be even larger (weather, melt ponds, ...)

So I think, we can do nice things without correction. And I am tired of searching in the gimp manual for today anyway.


I guess a config like on 29 May ’13 can be found somewhere in the NCEP/NCAR archives. It is not the individual case that’s interesting. It is the slow repetition of peculiar configurations that line up. They do, with more and more amplitude, since (my estimate) the great Russian heatwave/Pakistan monsoon floods ’10.

This is what I’ve used as a ‘ridge-count’since last September:
 photo Ridgeanalyses500Mb29052013verysmall_zps3cbcf935.jpg

For this, I use 500Mb, like you, Chris. The ridges seem to be more manifest under the Jet-level, about 3 km lower in the troposphere (the Jet ploughing on 9 km above the boreal region).

Here’s the surface air temp with the jet as an overlay. There’s much less coordination.:
 photo Jetandtemp1000Mb29052013verysmall_zps8ceaf514.jpg
The Pacific jet-field locks colder air over the Bering region. The troughs over Greenland, Europe and central Siberia feed colder than usual weather in south-central Siberia, France-Spain and mid-Greenland.

Is this weird? No individual feature is. It is the line-up. And note, the CAB is not cold. And it didn’t get much ‘winter-power’ during the freezing season…


Meanwhile...take a look at MODIS r05c05 tile...it shows the Lena river flooding south of it's still iced lower bed.
This marks the warm ridge over that river basin.


No one, yet? Allright, good morning all, I’ll continue my sudden boost in presence…

I did some CAD counting on the MODIS tile r04c04, the one with the persistent Low. I’ve digitized the most important open water poly’s in that ephemeral day 153 scene. Without pretence that I could get them all, I got 152, spread out over a 1000 km stretch, totalling about 2500 km2.
That’s 16 km2 for the mean. The bigger ones.
There’s enough to double that (if one could stick to the job…).

5K on a 1000K km2 isn’t much. I can’t check the CAPIE in the CAB. Pity, because that is where all these polynia’s and leads have formed.

But I could fit the 2012 tile day 153. That was a good white swath of ice with the usual rubble filled leads, but no myriad of 16km2 poly’s.

So what? I think at this stage SIA doesn’t represent serious indication of the state of the pack. For one, it overrated the ’12 blue ice in the Laptev Sea for day 153 for about 210K.
For second, it does a comparable thing for the Bering Sea (fog?), about 60K. To me, it looks like the 628K lag for ’13 is deceptive. It must be more like 250K.

For third, and after comparing the ASCAT for day 153, I get the impression that the whole FYI band in the CAB is bound to melt out. And as this awesome process has been prepared for the last three years, it may well happen whatever the weather conditions may be…

PS hey, ASCAT shows that, indeed, the Goat's Head is now over (or 'on') the Pole. Could that temporarily prohibit open water there? And how long?

Chris Reynolds

Werther, and anyone else.

I've done a slideshow of jpgs about the Jetstream and reading it from NCEP/NCAR.
It should work if you click on image 01 and use the right arrow key to scroll through the series. This seemed to be a better approach then an animated gif (which you can't pause), powerpoint (not everyone has a viewer) or a Youtube video (which has poor resolution).

NCEP/NCAR plots from here:

SFSU Jetstream plots from here:

Chris Reynolds


Another one in the spam filter.

[Released, N.]

Bill Fothergill

@ Chris R

Thanks for the Jet Stream file, and the earlier link to your June Crash article.

I apologize if I am teaching you how to suck eggs, but Powerpoint Viewer is available as a free download from Microsoft.

cheers bill f


SIA in Arctic Basin did not drop further yesterday. In the high res concentration pictures with (3.125 km)^2 grid I draw a circle (selection ellipse) in gimp from upper left position 800 1,450 with size 600 600 and (2.76 Mio sq km) and counted the mean ignoring values <38 (the black hole) and multiplied by area/255.

Since 20th May the loss there is only about 40-50,000 sq km with "biggest" drops on 27-28. May. I thought effect of divergence and/or upstreaming would be maximal just when the storm moved and before it slowly damps. But the effect is either still not there or overlooked by the sensor.

Werther, I like your "ridge count" - is is a good guide to the eye. If I understood wetterzentrale right, the European ridge is bending towards the Basin bringing Laplands heat (hottest place in Europe) to the Basin. Wheater is always a bit surprising - now it is surprisingly surprising and weather forecasts in the newspapers are a joke if you compare with yesterdays paper.


BTW: Wipneus' usable version of the high res concentration pictures from Uni Hamburg are here (including 06. 01.): https://sites.google.com/site/apamsr2/home/pngcby32/

Chris Reynolds


"I apologize if I am teaching you how to suck eggs, but Powerpoint Viewer is available as a free download from Microsoft."

Thanks, I wasn't aware of that, but I think the problem is that (as I've found with Excel) there's a significant number of people using Linux. Mac and Linux are about 1/5 of the hits to my blog, I presume that's fairly representative.

Chris Reynolds

Part 1 of my May Status post is about the May storm.
Linked to as it's relevant to this thread.


Chris, I can not find a significant indication for divergence in the AMSR2-pictures - a loss of 40-50k sq km SIA in ten days in the complete storm affected zone in the CAB is nothing to write home about. The 600-pixel circle spans also the circular shear region surrounding the spiral.

I see two possible explanations: First - it takes some more days until the SIA is reduced there visible for the sensor. Maybe the clouds have to go or the relatively slow movement has to be completed or something like this to get a final measurement.

Second possible explanation could be, that some of the water between the floes detected in visible and IR spectra is in fact only wetted ice by waves and spray? Albedo of water and sprayed wet ice is quite similar - immersion is the effect used in microscopy to get a clear look through a rough scattering surface.

In anyway - water or wet ice will both drive the albedo-feedback and that will kick the CAB in the next weeks as you mentioned allready.

Kevin O'Neill

The onset of ice melt at the north pole typically begins in early June. This is seen first with water-logged snow followed quickly by the formation of melt ponds. Melt pond formation is usually visible in late June or early July.

The effect of the late-May cyclone has been to mechanically weaken and spread out the ice. This occurs at a time (relative to 10 or 20 years ago) when scientists are already having difficulty finding decent-sized floes on which to place their equipment. The early evacuation of the Russian SP-40 research station is just one more incident making this point clear.

There has likely been a reduction in ice volume in the vicinity of the cyclone due more to mechanical and chemical processes than surface temperature or insolation driven melt. The effect of this preconditioning of the ice should become apparent as the 'real' melt season soon begins.

And this is in addition to the other preconditioning effects that have become the norm these past few years. Chief among them (IMHO) is that DMI north of 80 temperatures rarely dip down to the 244K (-29C) baseline during DEC - FEB anymore. This leaves the ice both mechanically weaker and warmer.

I'm unsure whether this will lead to dramatic year over year changes at this specific point in time. It will, eventually, if not this year perhaps next year or the year after - sometime in the not too distant future.

In the end I always come back to Waslowski and Wadhams and their prescient 2007 statements. Maslowski predicting a 2013 summer melt out and Wadhams commenting:

"The implication is that this is not a cycle, not just a fluctuation. The loss this year (2007) will precondition the ice for the same thing to happen again next year, only worse.

"There will be even more opening up, even more absorption and even more melting.

"In the end, it will just melt away quite suddenly. It might not be as early as 2013 but it will be soon, much earlier than 2040."

Chris Reynolds


You won't find as much divergence in AMSR-2 because the HYCOM divergence leading to the low concentration region is due to the modelled ice having a higher tensile strength than in reality. The divergence will occur most evidently outside the region of fracturing, with the region of fracturing having no large single openings, just a mass of tension fractures.

Furthermore AMSR-2 doesn't have the sort of markers that allow spotting of movement that are available in the HYCOM images and (in terms of MYI) the ASCAT images.

But the divergence is there around the opening in the HYCOM model.

And ASCAT shows the same movement of the MYI area as HYCOM shows.
ASCAT Day 146
ASCAT Day 152
ASCAT also shows the bulk anticlockwise movement due to the storm, as seen in those two images.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment