« Looking Forward, Looking Back | Main | Melting of the Arctic sea ice »


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


How will we know when the Nares ice bridge fails? Good question.

A lot of people watch the Petermann Glacier. It seems lasting fame is achieved by whoever can report calving events first, even it if is two minutes ahead of the next person.

Otherwise, that channel is one cloudy part of the world. As you can see below, cloud-penetrating radar lacks enough resolution, while the IR and visible can barely see through the clouds.

There are flyovers and sometimes people on the ground maintaining cameras and seismic sensors. More likely, we'll see it first from a break in the weather (but when?).

 photo naresCloudy_zps1eb67d58.jpg


This time of year, the sun is finally high enough that the visible range of the spectrum reflects enough sunlight to give a decent channel 2 image on the AVHRR satellite. (Actually this is near-infrared, 0.92 µ). Channel 3 (3.74µ) came on earlier but will always be a little fuzzy. Channel 4 (10.8µ) is what we have been using all winter on the Beaufort and Ellesmere regions.

Three channels, hmmm. Enough to make a false color RGB. Gimp can combine the channels using Colors -> Components -> Compose. However it is better to 'equalize' the grayscale scenes before combining, then invert, and color-balance to taste.

The final product combines the information from the three wavelengths, and highlights numerous features of the ice that aren't apparent in the individual originals. The first image shows a small slice from each; the larger color product (965 pixel width) shows the entire lower Beaufort Gyre region on 28 Mar 13.

 photo infraredFalseColor_zpsd57ac45c.jpg
 photo beaufortFalseColorBig_zps0ef2b7f6.jpg


Wow. Amazing work as always A-Team. You really should start posting some of this stuff in the ASIF. Maybe make a thread just for your images and animations so that we can better archive it?

Oh, and sorry about this stupid question, but I'm fairly new here, so I can't quite tell exactly what region that covers. You said it's the lower part of the Beaufort Gyre, but the little bit of land mask in the picture isn't quite enough for me to fully recognize where it is. Could you clarify?

Chris Reynolds

A couple of days ago I posted a link to a recent Env Can NOAA IR satellite image, but for some reason it was reverted to 'private' so nobody could see it.

Here's the public link.


Fair question, HIB. The coastlines are outlined in white in the initial satellite imagery that we download from the Canadian receiving station, here beaufort.130328.2121.4.png

However nothing is labelled. So for the names of obscure islands, gulfs, sounds and straits, I use a map called mcr_0036.jpg:


On the larger image, which is rotated 45º from the download, Prince Patrick and a bit of Banks Island are visible as white outline in the larger image.

I really should have masked out the land at the first step because its coloration, while fascinating, obscures features of the ice near the shore. This is easily done with the Gimp, PS or ImageJ contiguous color picker because the photo reserves pure white for its boundary.

 photo islands_zpsd9ee118e.gif


Here is the entire Beaufort image with a land mask, then a blow-up of the coastal and inter-island waterways around Banks and Patrick.

While the scene is confused in spots by clouds and ice fog from fractures, a lot of complexity can be seen in this non-pack ice. I do not know what all the colors represent in terms of the physical state of the ice.

The land mask is a bit tedious to build because they have quite a few perimeter gaps (inexcusable cartography errors) in their island outlines. The image purveyor should supply the mask to begin with, as Jaxa does.

My image above has its channels labelled backwards, it was built as beaufort.130328.2121.432 for RGB, not 234. The one below is -(43-2) in terms of channels and their inversions. (After getting a final product, Gimp history will help you remember how you got there!)

 photo beaufortTriple_zps3458a424.jpg

Jim Hunt

A-Team and Neven still aren't getting all the credit they deserve:






Several people have asked about compositing 'Beaufort' and 'Ellesmere' 10.8µ imagery, our highest resolution infrareds. Just set canvas size to 2852 x 1829, open the Beaufort to the lower left corner by using an offset of (x,y) =(0,269) and the Ellesmere to upper right corner, offset (1002,0). Alternatively just open the Beaufort in a double canvas, set Ellesmere to half-opacity, double the zoom, and move until Banks and lat/long are perfectly co-registered in the see-through.

Crazy-makingly, the Beaufort frame does not extend up to the 0-180 meridian line, so for a complete half-ocean the upper left corner (Wrangel) has to be stubbed in from 'Arctic Composite' (which we didn't want to use in the first place because, more crazy-making, its resolution was dumbed down 274% so as to barely display ice fractures).

The image below shows a seamless composite for 29 Mar 13, both taken at 02:22 UTC. This give the full fracture picture for the western Arctic. I had to scale down to keep the file size within bounds for the blog.

It is not common for images taken at the same time have full cloud-free scene coverage in full focus but if times are too different, the ice will have shifted a few pixels. Because area above Ellesmere is generally colder than the Beaufort, a different temperature scale is likely to be used, giving rise to a sharp seam on the overlap. This can be removed by resetting contrast end points with the "Curves' tool in Gimp. This temperature scale changes without notice by the season, by the hour and by the satellite -- below is a small sampler of that.

Ellesmere 3:26:07 -38.5 -19.0 noaa19
Ellesmere 3:26:14 -38.2 -19.0 noaa19
Ellesmere 3:26:17 -37.9 -18.7 noaa19
Ellesmere 3:26:16 -38.2 -19.0 noaa16
Ellesmere 3:26:20 -37.8 -18.8 noaa16
Ellesmere 3:25:22 -38.6 -19.1 noaa16

Beaufort 3:29:02 -35.6 -16.1 noaa16
Beaufort 3:29:19 -35.7 -16.2 noaa16
Beaufort 3:29:21 -35.4 -15.9 noaa16
Beaufort 3:29:23 -35.1 -15.6 noaa16
Beaufort 3:30:00 -34.9 -15.4 noaa16
Beaufort 3:29:22 -35.3 -15.8 noaa19

 photo Composebeaufort13032902224_zpsd9d12722.gif


The lower corner of Beaufort, the part above the Alaska/Yukon border, has been obscured in recent days by dark (warm) clouds boiling up from the southeast and out the Bering Strait. It's opening up a bit now to display something that has crossed the divide between 'lots of fractures' to 'lots of little ice floes'. The infrared temperature of this ice has also noticably warmed.

So here we are, still in March, still way ahead of schedule, already looking like mid-May. If so, we're moving right along through Kübler-Ross's five stages of sea ice grief (anger at inconvenience, paid denial, negotiating payoffs to policy makers, starvation and acceptance of a few centuries of bad climate).

Double-wide image: top left unretouched 10.8µ 30 Mar 13, top right inverted; lower left 28 Mar 13 false color infrared (rgb = channels 432), lower right unretouched

 photo BeaufortBreakup_zpsbcc41502.jpg

Jim Hunt

Is there a moderator in the house?

My comment about A-Team and Neven not getting proper credit seems to have ended up in amongst the spam.

For some strange reason it does that most other places too!


Ice Thickness Comparison:


[edited the images, but TypePad doesn't seem to allow sizing, N.]


I looked to see whether melt season onset could be quantitatively described by warming of the upper ice surface (the temperature reported by AVHRR 10.8µ infrared).

Of course, the ice is really melting now more from inside and from below, so observed surface temperature depends on this heat played out against mean surface air temperature, as refereed by conductivity profile and the heat equation.

To do that, I clipped out a central region of the Beaufort Sea from the last couple of months. This region is mid-latitude --300x 300 pixels centered on 80ºN 140ºW -- and generally cloud-free, so hopefully representative, the idea being to make an animation of the histogram window (whose frames capture the parameters for later statistical analysis).

However this imagery is subject to their constantly changing inept scale renormalizations, so a proper key has to be re-done for each frame so the histograms can be scaled to their temperature.

Not saying this is infeasible, but all in all, a project better assigned to my imaginary assistants.

 photo histogramAnimation3B_zpsdea6995f.gif


Jim, I just released all comments.

A-Team and Neven still aren't getting all the credit they deserve:





NASA Earth Observatory links to two blog posts on the ASIB. Kind of flattering. :-)

Jim Hunt

A small step in the right direction at least Neven!

Steve Bloom

Thanks for that comparison, wanderer. Yowzah.

Steve Bloom

Hmm, spam filter just got me for a very short comment without links or graphics. Odd.

But to flesh out that remark (about wanderer's comparison) slightly, the apparent tight internal rotation of the remaining thick ice seems like a harbinger of a very tough season.



Thanks for the compilation, I had noted the NASA EO mentions, but this is better.

Thanks to Neven, A-team and others who post solid science that have given this resource a significant level of credibility.


I'd say raise a glass in toast to everyone who is posting or quietly learning here, especially those who then go out into the world to balance off the media and inform the broader public. And those like Neven who had the sense to step out on the ice while it was still thick.

The baton will likely be passed this fall to the meterologically minded -- but the blog is very well-positioned for that already with some patient and expert mentors.

Narcissus is a big hit (of course he thinks this is all about him). It goes out Monday to 300,000 members of the Center for Biological Diversity. They are the group that got the polar bear endangered listing based on the threat (beyond that unfortunately) of climate change.

They did contact me about permissions, credits and sea ice blog links but I am not recalling how we left it ... probably that they just use it in their fundraising.

Jim Hunt

Cheers A-Team,

Particularly since I now find myself in the position of endeavouring to inform the broader public over at the Economist about the views of Jennifer Francis. People are saying all sorts of unkind things about me, and all entirely without foundation!



People are wondering how ice pack fracturing, more melt ponds and earlier break-up will affect the biology of the arctic. Sea ice made the cover of 22 March issue of Science -- A. Boetius et al on the 212 Alfred Wegener Institute's 2012 Polarstern cruise to the Amundsen Basin -- in the form of the meter-long under-ice diatom, Melosira arctica.

Too much information to really summarize here but in the short term, a lot more diatom habitat with younger translucent ice, a lot more growth from May to August, the diatoms falling past the pelagic uneaten, dead diatoms littering the ocean floor at 3500-4400 m depth, but not fixing any carbon to sea floor sediments because of being devoured by echinoderms -- sea cucumbers and brittle stars (notably Kolga hyalina, Elpidia heckeri and Ophiostriatus striatus).

These have conventional fully aerobic metazoan metabolism, so the CO2 from water that the diatom photosynthetically fixed as reduced carbon get fully re-oxidized back to CO2 (the part not assimilated into echinoderm body parts, except for some fraction utilized by their enteric bacteria (also fully oxidized with sulfate the probable terminal electron acceptor). Whether something bigger swims in and eats the echinoderms, I couldn't say.

The net effect, which they quantitate to 9 grams of carbon per sq meter per season, is transferring a climatically insignificant amount of CO2 from the surface to the bottom waters. Worse, the diatoms cannot grow in the open ocean, they need the sea ice environment. So even this is an ephemeral phenomenon -- when the May-August sea ice is gone, there'll be no more Melosira carbon fixation at all.

It's not clear that other photosynthetic diatoms will pick up the slack. Certainly not the obligate freshwater fern Azolla (with associated cyanobacterium Anabaena) that once grew to great abundance on the Lomonosov Ridge in the old days (when it was above sea level). Everything will be limited, as it is now, by phosphorus (and nitrogen) availability.

So, while there might be a few good seasons left from re-freeze of first year ice, primary production -- and thus arctic biology -- may otherwise be cut off at the knees.

 photo venusHalfShell_zps22c66cfd.png

Martin Gisser

A-Team, thanks for looking into diatoms and all. Arctic phytoplankton gave me a little "positive" perspective on sea ice melt. Perhaps one day The Blog will feature beautiful images of Acrtic phytoplankton blooms?

The nutrients would come from melting permafrost (according to my humble theory). I'm no biologist, but general Gaian/entropic reasoning would suggest something to come and use these nutrients. I guess there would also be free-floating singular diatoms, unlike the current Melosira colonies.

So you think the carbon sequestration would be tiny, if any? The echinoderms produce calcium carbonate in their mesoderm, so is "safe" to some extent against ocean acidification (being covered by skin). The question is, what happens further down the road: How quick will the calcium carbonate be buried in sediment?

(BTW, Diatoms are extremely beautiful stuff: Ernst Haeckel's Kunstformen der Natur has a plate. Another German, J.D. Möller, made stunning microscopic artwork of them.)


Not directly related to cracking ice, but Dr. Jeff Masters has written 3 pieces on his blog that directly ties in what is happening in the Arctic due to AGW and what is happening to the the weather. Granted is US centric, but still applies the rest of the North.


http://www.wunderground.com/blog/JeffMasters/article.html link will change tomorrow.

A major new multi-year fracturing pattern has appeared north of Ellef Ringnes/Ellesmere and is working its way west. It's not so clear what novel array of forces are causing cracking orthogonal to the previous coastal CAA rupture event -- the usual Beaufort Gyre and Transpolar Drift motions appear on Navy Hycom animations.

 photo ellesmerePerpCrack30MarInvertWBEqB2_zps560de748.gif


A-Team and the group - first post from an enthusiastic lurker - I saw that orthogonal pattern as well, and was both fascinated and disturbed. The comment I made elsewhere was that the arctic ice was literally being converted to cubes.

In petrology, long ago and far away in my life, we'd see clevage patterns like that from pressure at depth; generally from perpendicular to the shortest corner-to-corner dimension. Not sure it applies here but seemed a useful thought in analyzing the vectors of force involved.

Trying to do an analysis heat flow to predict early season melting I think is exactly right, but a huge challenge to get the sea ice surface measurements.

Just as an FYI, and response to earlier discussion about the value of what's happening here - I've been pushing A-Teams animation aggressively, both with my on-line community, and contacting media. I finally got some attention from the Boston Globe. Hopefully, they will come this way for some education.

I'm sure there are other folks like me quietly pursing this sort of thing. Let us hope we can overcome "normalcy" bias and get people talking.


One follow on thought... If I recall from some research I did when prompted by some earlier posts here, I figured that ice would build at the rate of about 10cm per degree C average below freezing. There would be a certain amount of "inertia" inherent in the latent heat of crystallization and insulative qualities of the ice, but would seem a good rule of thumb. At that rate, the heat flow would keep up with the rate of transfer. Move the average, and your thickness changes accordingly. We might be able to do that to get a sense of new volume created when the leads freeze, and what the limits are to how thick it will get.

Steve Bloom

Wow. Has this sort of cracking pattern been seen before at any time of year?

Steve Bloom

Ah, Venus on the Half Shell (for any Kilgore Trout fans present). IIRC it didn't start well.


Was there a small earthquake? It looks like a sonic wave front passed from right to left orthogonal to the existing fractured ice. There seemed to be a pause just before the final cracking finalized in the western quadrant. It may have been a reflection at right angles to the land forms locally, partially canceling wave energy. It could also be the ice itself being stretched and compressed along vertical shear zones much like the fracture zones radiate from the undersea central ridge of the Earth.

Artful Dodger

@Llosmith57 | March 31, 2013 at 05:54

"Was there an earthquake?"

No. The full moon was a couple days ago. We saw similar results after the last new moon. Together, these two events are the strongest Spring tides of the season.

Jim Hunt

I'm a Kilgore Trout fan Steve!

So it goes?

Espen Olsen

A-Team and Sigmetnow:

Today April 4:
Pre-Fractures are now clearly seen into the CAA from the north, between Ellesmere and Ellef Ringnes Islands.

I am sure we will soon need some animators cover this story.

The comments to this entry are closed.