« Yamal to the rescue | Main | SEARCH 2013 Sea Ice Outlook: June report »


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


3.6 Mkm2
I tend to believe PIOMAS and it thinks there is more ice now than last year. There has been the suggestion that PIOMAS overestimates the thickness / volume of thinner ice. There seems to be more of that this year which could then prove me wrong.
I do think the PAC has tended to spread the ice and this year will start to catch 2012 soon. I am expecting the the Laptev bite to see to that.

Remko Kampen

Mignonette, I do not use the graphs or single days to arrive at an estimate for the minima.
However I do use them, in combination with the state of the ice and meteorological conditions, as sign & symptom.
I also do not extrapolate from previous years (or if I do, I do so with Wipneus' exponential curve on Piomas (yep) ending in zero by 2015).
The state of the ice and meteo suggest we are in for a good number of even worse drops, possibly far worse.
And I am reckoning even more than earlier this year with the almost total wipe out this year, only excepting the pushed up zone off part of the CAA and Greenland.
I am also daily more surprised at the underestimation of current events even by many profs.


re Stormy June, that would be T-Bone Walker, not the Allmans.


@Remko Kampen: Fair enough. I was not suggesting that you were basing your estimates on graphs or single day's data but rather that this kind of short-term noise should be kept away from this thread.

The exponential trend for PIOMAS fared well in the last 3 years (3 consecutive drop years) but it is uncertain yet that it will keep doing so. Let's see.

Kevin McKinney

OT, but--

"...T-bone Walker, not the Allmans."


Eventually, it was T-Bone Walker and the Allmans and "Albert King, Eva Cassidy, Chris Farlowe, Question Mark and the Mysterians, Jethro Tull, Eric Clapton, Davey Graham, Lee Michaels, Gary Moore, and others."


the Wikipedia article.

Aaron Lewis

Less than 1 million km^2 average ice extent for month of Sept. 2013. That is functionally 0 for critters like polar bears, walrus, & seals.

Large areas will have dispersed ice that will not register as ice extent, e.g., ice will remain a hazard for shipping.

This estimate is based on sustained low pressure in polar region continuing to drive the polar cyclone that pulls latent heat from the south. Thus, ice melt is based on hemispheric heating, rather than on direct local solar heating. The latent heat then condenses on the ice, thereby keeping local atmospheric pressure low, and driving the cyclone. (I do know that wind driven by phase changes in water is not part of the conventional wisdom on weather. See for example; Makarieva, A. M., Gorshkov, V. G., Sheil, D., Nobre, A. D., and Li, B.-L.: Where do winds come from? A new theory on how water vapor condensation influences atmospheric pressure and dynamics, Atmos. Chem. Phys., 13, 1039-1056, doi:10.5194/acp-13-1039-2013, 2013)

I suggest that the slow start of the melt season is the result of warm/weak ice spreading out across the surface of the water. That increase in area of ice detected fools PIOMAS into overstating current volume.

Kevin O'Neill

Aaron - I'd be very skeptical of Makarieva et al: Should Eli Believe His Lying Eyes or Makarieva


Same source for me. I knew the Fillmore East version and assumed they wrote it, which is remarkably stupid since they actually say a T-Bone Walker song. Being aware of that cognitive dissonance, I actually looked it up.


Remko is 'daily more surprised at the underestimation of current events even by many profs.'

Me too. I think where this starts is with never looking at the data. Today, that consists of Modis, 89Ghz, 36VH 18V, Ascat, SLP. Five open tabs, five minutes a day to refresh and review -- that's not asking too much.

Tremendous daily confusion on this blog between real data and its secondary interpretation (by automated, often antiquated pipelines driven by algorithms not plausibly inapplicable to today's ice).

The animation shows the last 21 days of Uni Hamburg high resolution sea ice concentration (dramatized palette); the 15 Jun 13 is the polarized mid-frequency passive microwave.

Let's take another poll: tell us what you think these two are showing, then remind us of your estimate.

 photo SicHburgJun13_zpscb4f084c.gif

 photo 15Junjax_zps1a7d3030.jpg


After seeing some of the rationale for the more extreme high/low predictions, I'm beginning to wonder whether we need to identify predictive methods other than heuristic, statistical or scientific. I think we also need to add "Hubris".

It seems to me that predictions that increase or decrease the 2013 results from 2012 in any category (SIE/SIA/Volume)by many orders of magnitude more than any previous year-to-year change are not well founded when using any statistical method or scientific analysis. That leaves me concluding that the predictor has a good case of hubris.

Highly educated and otherwise well informed people are not immune to catching a good dose of hubris. In the latter years of my career, I was responsible for performing a Win/Loss analysis for a major aerospace firm at the completion of competition for major contracts, typically those whose initial value exceeded $150M. Since the analysis was expensive and time consuming, not all contract award decisions were analyzed. However, in almost every contract competition that we lost, internal hubris was a contributing factor, if not the primary factor leading to the loss.

This is something we all need to be aware of. Individually, we can become so enamored with our own intelligence, analytical skills and thought processes that we lose sight of what realistically can or can NOT happen in the near-term.


4.5 million km^2

Just a generous "dead cat bounce" & a feeling that the slowness of the melt might persisy.

Account Deleted

Talking about dead animals, where is the goat

Nightvid Cole

I think 3.75 M km^2 is reasonable. Not as high as 2011 (since the snow over Siberian sector is going faster than in 2011 and the ice concentration over European Russia sector is lower) but not quite as low as 2012 (snow melt is behind, season started off colder, PIOMAS volume higher).

Nightvid Cole

And to add to my previous comment, range/error bars will be somewhere between 2011 and 2012.



That certainly is a fascinating and alarming animation in your last post.

That fact that you have some amazing image processing skills as well as your references to living in Tucson made me think of something. To the best of my knowledge, there are only two employers in the Tucson area that have employees with in-depth knowledge of image processing. I retired from one of them. Obviously, image processing was not one of my skill sets.

Just wondering.


My first comment after reading with great interest for many months. I find the loss of the arctic ice deeply unnerving and refuse to participate in a guessing game about it's TOD. That being said, I sense that a hastened demise may well benefit us in several ways vis a vis our current woeful ambivalence.

The NH weather is now becoming so extraordinarily chaotic that an ice free zone at the pole may have some sort of stabilizing effect. Perhaps the weather will become more predictable rather than less so. Of course, the disruptions to civilizations will be enormous but they're getting to be already.

As if to declare the Arctic summer official launched, there is a major fire NE of Fairbanks AK. It is visible on Weather Underground satellite (wundermap). They'll be at or near record high temps (95F!) over the next three days there.


Lurking here for a year (= still a newbie & still learning how to evaluate the graphs & graphics). Can you provide links to the 5 tabs that you refer to as most important (Modis, 89Ghz, 36VH 18V, Ascat, SLP). I know these are probably in the 'Arctic Sea Ice Graphs' link above, but frankly I get overwhelmed with the shear volume of data. If these are key, I want to watch them daily as you suggest.
Thank you for all the info you provide. I AM learning a lot (slowly)...


@ulicescervantes. If you check out A-Team's excellent animation near the end of the comments to Neven's Update 2: Shaken and stirred, I think you will see what appears to be the goat's head dancing near the North Pole.

As a naive spectator, I am dumbfounded at the persistence of this landmark. I am starting to think there might be an actual giant goat staring up from the arctic basin. I can't otherwise understand how it survives the weather for so long.


Put me down for 2.0 km2.

I'm no scientist. I'm only guessing. But what I'm basing that guess on is the way that the ice area and anomaly are both moving rapidly downward right now. I'm also looking at how the North Pole might be completely ice free before long, and I figure that can't be a good sign come September.


.95 million square kilometers with at least 2 days being ice free.

Rationale: purely subjective and based on a unique life experience. Many years ago when Dad was still alive he wanted to go fishing to his favourite lake late in the spring. When we got there the lake was still 98% covered with rotten ice with only the shoreline being open. We had to boat a long circuitous route to get to the place he wanted to fish. Well the fish weren't biting but the wind kicked up and started to move the ice mass. We were forced off the small section of the lake that was ice free & then we witnessed three miles of rotten ice get smashed to bits on the far shores in a matter of less than an hour. Our row back to the boat launch was across open water.

Is not the Arctic showing plenty of melt along the shore? Has not the ice been pounded into submission by the storm? Is not the overall quality of the ice poor & rotten? IMO, all that is needed is one good August storm (ala 2012) to complete the analogy. Fractal patterns repeat in nature all the time. It's a matter of scale.

Most people on this fantastic blog are referring to the possibility of a rapid ice loss as "the cliff". IF/WHEN it does happen, I will refer to it as the Belmont Lake Effect or BLE... to go along with all the other acronyms that have kept this lurker on his toes for some time now. Carry on Gentlemen. It will be an interesting summer for sure.


3.25 million sq km. I believe the current rate of melting is around 5% of the originally measured 1980 - 2000 extent and coverage, so an addition 350,000 sq - km should melt this year. the season started late, but there is so much melt ponding from the pole to the euro-russian side, and it appears east siberia is melting fast now from a strong heat wave. we will see open seas closer to the pole than ever before, and it may reach the pole from above svalbard.


@Kevin: Last off topic comment, promise: just noticed you mentioned Eva Cassidy. Funny that, it turns out she was in school with family. I I had no idea who she was until yesterday but from a different source. I was trying to explain Funk to one of those cultural barbarians from Europe, so naturally looked up Chuck Brown, hence Eva Cassidy (yes, they did a duet album but Chuck was a local god and hence everywhere locally). Still what are the odds of two so disparate and unlikely streams producing the same results within hours.

Bad year for ice, i think anyway.



Speaking fro myself, it isn't hubris. It is the nature of the poll that asks us our best guess. In the end, that is a number. I don't think any of us take it so seriously. I know I wouldn't put any money on my guess (2.2).

I freely admit I have no understanding of ice, arctic, etc. On the other hand, i've watched this system bounce around for a few years and believe that my ability to gist the wildly fluctuating data is probably as valid as any expert model. Again, not hubris, just looking at the piomass type graphs and distrust of models in such a complex situation with so limited data.

Hans Gunnstaddar

Ice extent is not dropping like it did in 2012, and I'm wondering if a positive feedback in what I refer to as the 'Slurpee Effect' is taking place. (The Slurpee was or maybe still is an artificially flavored crushed ice 7-Eleven concoction.) In this case, open waters where there were once multi-year ice floes are generating thin ice, which gets crushed or cracks open providing more ice forming opportunity and so on in an unexpected positive feedback that (for the many years until greater warming sets in) keeps the Arctic chock full of 1st year ice in the minimum range of 3-5 million square kilometers. 2012's record melt acting as the initiating event that opened waters sufficiently to usher in the 'Slurpee Effect'.

After all, we need to keep in mind that the rapid decline of Arctic ice volume and extent has never been observed before, so we do not know if the trend-line to an ice free Arctic in approx. 2016 will continue without positive feedbacks kicking in.

Hans Gunnstaddar

Sorry, I meant negative feedbacks. Too bad there isn't a way to modify a post once posted.


it's taken a while but it has finally arrived.

I expect things to pick up on that coast now.



most of the remote sensing products A-team mentions are shown and updated daily here:


Remko Kampen

A-Team | June 17, 2013 at 23:22 - What can I say? 'Thank you for that sickening illustration of my comment re underestimation of current events'? Wow. Thanx, A-Team.

Strikes me some people below your post still don't get it. Daily a day worse.


Yes Neil the daytime temperatures in the 50s have certainly hastened the surface melting. The radar animation is quite interesting too, up until last night the fast ice extended out to about 5km but overnight it has cracked about 2km offshore and started to move. Presumably one of the offshore pressure ridges that held it in place has melted enough to release.


Peter Ellis

There are no grounded ridges this year. From the SEARCH June outlook:

"We also have an 'on the ice' report from north of Barrow by Oliver Dyre Dammann and Hajo Eicken, University of Alaska. Shorefast ice conditions along the northeastern Chukchi Sea coast (Barrow to Wainwright) reflect the combination of late freeze-up after the 2012 record minimum summer ice extent and persistent westerly flow advecting warm air throughout fall and early winter. The lack of major storms from the western/northern sector resulted in a shorefast ice cover that is potentially unstable due to the lack of grounded pressure ridges. The complete absence of multiyear sea ice in the region, confirmed by thickness surveys and local observations, is a first for the region in the past several decades. In early March, shorefast ice broke out close to the beach along a >100km stretch of coast between Barrow and Wainwright. These conditions have left shorefast ice vulnerable to break-out events. Hunters in the region have had to traverse thin, unstable ice as reflected in ice thickness surveys along trails in the Barrow region. With lack of multiyear ice, a normal or slightly below-normal thickness offshore ice cover (based on ice thickness flights earlier in the season) and coastal ice vulnerable to early break-up, ice conditions would favor a normal or somewhat early seasonal ice retreat."


Examination of 79 Glacier, Greenland posted today, including Landsat imagery from 6/17/2013. 79 Glacier


My guesstimation is for 2.6 MKm^2

My thinking is that last year produced the largest amount of first year ice on record. I'm suggesting that the energy budget for the area is the same as last year plus a small additional amount of energy from the increasing global warming. It therefore seems likely that at least the same amount of ice will melt this year with some energy left over to melt more. With all the cracking and thinness I'm suggesting another 1 off last years result to give the 2.6 number.


OldLeatherNeck is referring to the element of 'hubris' in some of the predictions.

There is a tendency for putting aside the SEARCH scientists as unrealistic optimists. But referring to a single year (2012) does not prove their incompetence.

It's a bold leap to assume that the increase in ice volume and area we are seeing this year, is suddenly going to disappear by September. No evidence to support that. Only gut feeling.

2012 is far ahead in every way. Including the melt pond formation that some of the above commenters are referring to.


The determination of sea ice concentration, area, extent, volume etc are all ultimately derived from an underlying remote sensing image (or two or three). This not being done manually, any artifact in the imagery will be carried along by the algorithm into the product. It no longer measures what you thought it was measuring.

For much of the year, no problem -- there aren't significant unmanageable artifacts. However in late spring, the cloud cover can become thicker, warmer, wetter, more persistent and more pervasive. Melt ponds -- initially fresh water -- form on and in the ice, altering dielectric constant, electromagnetic emissions and visible reflectivity in a manner than can come to resemble open sea water.

The clouds and melt extents on a particular day do not match from year to year, rendering comparisons problematic to the extent algorithmic misinterpretation is carried through to final products. This is not Greenland, where something is actually measured (eg mass balance by GRACE). Nothing is measured in the Arctic Ocean. There is no concurrent ground validation of any of the products.

The previous post looked at cloud and emissivity artifacts creeping into higher frequency emissive microwave imagery and products as melt season progresses. The animation below investigates artifacts in 5.3 Ghz backscatter microwave (Ascat) for the 25 days ending on 13 Jun 13 (day 164).

Each frame takes 3 consecutive days and makes an RGB image out of the three grayscales. For example, the final frame is (day 164, day 163, day 162) = (R, G, B) and the previous one is (day 163, day 162, day 161) etc etc. The ice is moving underneath the whole time causing small color shifts vaguely analogous to a 3D movie but the predominant effects are on the periphery.

However this is day 168 and artifacts have moved in closer to the central Arctic. Judging by 2009-2012 Ascat, the useable fraction -- defined above from the mask-out provided by coloration -- will continue to decline. So the question becomes, with Modis visible and infrared so patchy and microwave diminishing, how are reliable ice products to be determined?

 photo 3dayColorShift2_zps26cf8831.gif

Kevin O'Neill

Melt pond formation peaks in late June. The algorithms that produce sea ice concentration deal poorly with melt ponds - resulting in SIC errors of 40% in areas with melt ponds (i.e., 'true' 90% concentration classified as 50% concentration).

In Melt ponds on Arctic sea ice determined from MODIS satellite data using an artificial neural network Rosel et al, describe a method to greatly reduce these errors. I don't know if this is an ongoing project producing regular data.


Old L -- and Lars earlier -- make an important point, that aerial photo interp has been researched and developed to the nth degree since the days of the Wright Brothers, both in information-theoretic academic journals and on the instrumentation side but primarily for land and air, the primary consumers being ecologists, agriculture, military intelligence, and operational safety.

On the blog to date, we have not used 0.1% of the image processing options available as open source Gimp scripts. The fraction of it that I previously knew came from helping various institutions compile and interpret aerial photo histories of nature preserves -- land and esturaries.

Of course in the Arctic, software can't make a silk purse out of a sow's ear -- white snow on white ice in white cap water viewed through white clouds. Still, we have not done nearly what we could have.

The US military handed over its meteorological satellites to NOAA a while back and we are free to rummage around on the Navy Hycom server. I'm not aware of them having magical sea ice sensors nor do I see an interest in reading license plates on Arctic amphibious vehicles, monitoring underwater barking of seals or capturing transmissions home from ice camps. Although militarization/industrialization of the Arctic could lead to higher resolution melt and floe products, that would be after-the-fact for summer ice pack loss and consequent climate change.

Arctic ice monitoring is mostly hand-me-down imagery from land satellites like Terra Modis. Whatever the orbit, the satellite can only spend a fraction of its day over the Arctic so inevitably they are multi-purposed. The biggest development really has come from Japan with Jaxa-Ijis imagery, the best of which is the color microwave. However the earliest imagery there is 24 July 12, precluding multi-year comparison from a fixed sensor.

I just leave a firefox browser window open with the tabs below (and let the newspapers jam up a safari window), refresh in late afternoon, add as a layer to an ongoing GIMP image for a quick comparative animation or blow-up, and save to archive (because most don't allow bulk ftp download). There are many other excellent sites, as mentioned by Arcticio / on Neven's best links page, but not better primary images as far as I know.

Since we are in unchartered waters given the 2012 preconditioning of the ice pack, you really want to be looking at primary imagery rather than traditionally derived products (eg as noted above by Kevin). Not that anything we could do will affect the season's outcome.

Jaxa RGB
Modis (equally WorldView)
89 Ghz
AVHRR Beaufort, Ellesmere
SLP, wind, quick ice product overlays
Navy animations


It was very hard to come up with a prediction since the ice is thin so weather will make a big impact to the minimum.

So instead I'll guess that if we have ice-favorable conditions the rest of the year the minimum in extent will be 4.6 million sq km.

If we have awful conditions for the ice the minimum will be 3.0 million sq km.

Average those two predictions and we get 3.8 million sq km. So considering we have average ice conditions my final guess for the minimum in extent this year is 3.8 million sq km.



You have one of the best and scariest posts I have seen so far. Seeing the blues show up off Ellesmere I ran off to check MODIS.

http://rapidfire.sci.gsfc.nasa.gov/cgi-bin/imagery/single.cgi? image=crefl1_143.A2013169054000-2013169054500.500m.jpg

With the scary bad ice turning blue off the last ice refuge of Ellesmere and the flow of multi year ice being pushed over Greenland, I am revising my September projection from 3.0 +0.5/-1.0 to less than 1.0


Paul Beckwith


You may want to revisit the navy movie for sea ice thickness for the entire last year. Even throughout the depth of winter, MYI has been sliding out of the Arctic basin over Greenland. Welcome to the < 1.0 group, zeros like myself are happy to welcome you to our group...

Hans Gunnstaddar

Mignonette stated: 2012 is far ahead in every way. Including the melt pond formation that some of the above commenters are referring to.

I agree, and find it fascinating in spite of incoming data so many are absolutely certain 2013 will be greatly less than 2012, with many adjusting their predictions downward. It could be a case of 'Groupthink' in which the need to follow the group is holding on to the idea of a huge drop in Arctic ice, but isn't there by any measure on the graphs.

Nature has it's ebbs and flows and this year definitely looks like a rebound. Remember folks it took five years after 2007 to get a new all time low.



No group think here. Quite to the contrary. My original estimate was based highly on the ice volume trends adjusted against PIOMAS' tendency to overestimate the thickness of thin ice, and on the binned ice volume/thickness behavior trends, then upwardly adjusted based on the extended cold this spring from the anticyclone.

However, in looking at the changes in the last several days and the horribly shattered condition of the ice near Ellesmere, and the greatly increased flux of MYI ice out of the arctic and off of Ellesmere, I cannot justify that any longer. Unlike previous years, it appears that the central arctic will melt out early freeing the ice near Ellesmere to move. With the extensive fracturing and thinning of the ice, it is hard to see how that can withstand the actions of wind and wave and the mixing of the surface layers. Next up, ice on both sides of the arctic should melt. This melt season looks like it will be quite without precedent.

What strikes me more is the group think that seems to be at play in reverse and the cognitive lock up that goes with it.



No group think for me either. I don't follow the predictions much and thought that I was being rediculously low compared to others. It was the Piomass graph that was my primary and pretty much only outside influence. After that was my 'intuition' which I don't trust a tuppence.


I want to lower my prediction to 3.75 million km2, if it's not too late, so that my prediction here and my entry on the forum poll ('between 3.5 and 3.75 million km2') are in line.

BTW, last day of voting on that poll and the CT SIA daily minimum poll.

Remko Kampen

Hans, are you serious about 'groupthink' referring to a group that just grew from 3 or so to 4 or so? Forget it - or please show how e.g. A-Team suffers from groupthink.

Look for yourself. Like anyone you too can see how 2013 will smash through everything in 2012 like this week or next the latest.


On the groupthink subject, it would be interesting to me, as an absolute laymen, to have some idea of the self judged ability of the forecasters in relation to their estimates.

This isn't an amateurs against professionals question. It just would be nice to be able to seperate out distributions and so on by group. Maybe years of experience would also be interesting. So, for example, me:

2 years


0 - occasioanlly read articles
1 - read blogs, follow arguments in season
2 - actually get into the math, amateur models
3 - very passionate but still amateur
4 - turn out serious products
5 - professional study of phenomena

I guess a lot of you guys know each other and your work. But for an amateur like me, I don't know what I am reading when I see the scores.


On scale, that would be a 1 for me.


For me that would be 16 years and a scale of 1.


Remko Kampen, A-Team's imagery is not immune to artifacts like melt pond formation. The ice pack is still around 2 meters thick including the pond regions.

A-Team is also a master in using contrast-enhancing techniques. Reference images to compare with 2012 would be welcome.

Some commenters are betting a lot of money on the Wipneus' trend for PIOMAS. But Wipneus himself is warning about the huge uncertainly range. Even a little upward nudge in the early melting season, may cause a temporary rebound to the 2010-2012 range. We'll see.

L. Hamilton

It seems to me that the numbers in text " from 3.4 to 6 million km2, with a median of 4.1." do not match the errorbars in Figure 1. Is there any kind of unit conversions that explain the difference? Am I missing something? Thanks.

The text you quote describes the median and IQR of SEARCH Sea Ice Outlook contributions. Figure 1, on the other hand, shows the median and IQR of Arctic Sea Ice blog readers' predictions. Farther along in the post, Figure 3 graphs blog and SEARCH distributions together.

Remko Kampen

Mignonette, check A-Team's imagery with e.g. terra modis true color. Pay special attention to rounded floes. They slosh around and collide in open water, up to very near the NP.
What 'artifacts', again?

Remko Kampen

Fufufunknknk, 32 years level 3 from the start :)
The first quarter century was quite boring...

Peter Ellis

For me, about 5 years of experience following Arctic sea ice. I don't get deeply into the mathematics or do any modelling myself, so I count as a 1 in your system.

However, I think you're missing something here, which is that not all mathematicians, models and arguments are equal. I am a professional scientist, my main field being molecular biology, and the reason I don't do the modelling myself is because I know the limits of my ability. Regrettably, I have seen several extensive "modelling" posts on this blog and others by those with no such self-awareness, proposing quite unphysical models.

The Dunning-Kruger effect applies to fake-skeptics and alarmists alike.

Oyvind Johnsen

I am an amateur at this, have been following this blog for a year. If there is a groupthink effect from reading and participating here, it must be about what will happen in the medium-short term. Almost everybody seems to expect an "ice-free" (=< 1 Mkm2) Arctic Sea Ice Minimum before the end of the decade.
In the short term (this year`s melt) it seems to me that the viewpoints vary quite a lot. There are a few who think we will surely see a new record low, but the majority seems to me to mean that the weather will finally decide.
For my part, I voted for no new record, and I think (guess) there will be a rebound in both SIA and SIE this year. I have no feeling of going against "what the group here thinks" in voting so. To the contrary, I find support for my guess from many of the informed posts here at the blog. support for m


I wasn't really expecting my system to last. That it made more than 5 posts is past expectations. It took about as long to think it up as it did to write. i assumed a better model would have been posted fairly quickly.

But I did want to leave room for amateurs at the top, thinking of people like Nevin whose opinion I trust. But also, for the spread of the numbers, I didn't want amateurs clumped at 0 and 1.

And I agree about modeling, even I thought once or twice about throwing togather a, uh, ahem, Netlogo model. (But for me it was more about the understanding that comes from creating a model rather than believing my model.)

But without making more categories to report, I thyink it has to stay with self reported ability. If someone with tea leaves (from ice tea, no doubt) wants to call themselves a 4, that is their choice. I was curious about who we are and how much experience affects the judgement (whatever the experience).

Angela Marchbank

I am definitely a 1 on your scale, I read and follow most of the posts and comments but do not have enough knowledge to make helpful comments. I will make a prediction for this year based on what I have read now and over the past few years of 3.9. I will base this on the feeling that it would be amazing if another record was broken after last year and the season seems to be slow so far so may run out of time to break another record. I do not think it will be much higher as I think a surprising amount of ice will still melt. Thank you to Fufufunknk for being inclusive for us interested parties and of course to Neven and all of the knowledgeable posters for making this an enjoyable of illuminating forum.

Kevin McKinney

Mignonette, I'm interested in the context of your comment about the ice being thicker than 2 m. Did you intend a pan-Arctic mean, or a specific area? I'm thinking about Neven's CAPIE measure--crude as it admittedly is--showing about 1.6 m at the end of May:

Kevin McKinney

HTML issues, second try:

Mignonette, I'm interested in the context of your comment about the ice being thicker than 2 m. Did you intend a pan-Arctic mean, or a specific area?

I'm thinking about Neven's CAPIE measure--crude as it admittedly is--showing about 1.6 m at the end of May:

CAPIE graph

If that is approximately correct, then do your conclusions about melt change?


Kevin McKinney, the 2 meter value is just an extremely rough estimate excluding the outermost periphery zones. I try to keep my posts short and simple. If that leads to oversimplification, so be it.

Remko Kampen, let's not put things out of context. Where did you see me suggesting that the events close to the North Pole are an 'artifact'? Check again. I am relying on the analysis by werther and others about these events.


@ Fufufunknknk,
About 10 years and a 1.


3.98 Mkm^2 Once again just a guess, but up from the last guess since extent is still just declining slowly.

Ian Allen

Mignonette, no-one has ever "looked" at an arctic so shattered. No-one. If you weren't new at watching paint dry/ice melt you would have noticed that shattering is a prelude to destruction. Many of us have been watching since before 2007 and realise the pan-crumbly arctic ice is a new thing. It wasn't like this in 2007 nor last year in June. Do not underestimate the power of July.


As far as deceptive contrast changes go:

I don't recall ever using the contrast change tool -- it's way too primitive a form of image enhancement and gives terribly sub-optimal outcomes.

The two images on this page are unretouched as is and furthermore at native resolution (resizing subtly alters the data).

In any event, the links to the original online source images have been posted here many many times, most recently yesterday; you should have been consulting these daily to begin with, all the more so if you wish to understand more advanced enhancement products.

Almost every product I post composes strictly 'scientific-grade' enhancement functions, meaning bijections, meaning invertible, meaning the original can be losslessly recovered, meaning no information has been added or lost.

Because only 3-4 people here are following along in Photoshop, Gimp or ImageJ, I no longer give replicative command line sequences. However Gimp .xcf file format does all of the instructions necessary to stepwise revert the image to what is on the satellite server.

Keep in mind wearing reading glasses at the computer or polaroid sunglasses while driving make rather serious and quite non-intuitive image enhancements.

Then there's the 12% of blog readership has some form of uncorrectable colorblindness.

I really can't grasp why people of minimal experience are seeking to jawbone the September minimum up. Nature isn't listening.

Even if you could persuade someone here of some piece of rubbish, September will be here soon enough, so what would be accomplished?

Chris Biscan

2013 isn't going to blow by 2012 this week or next week.


Still below freezing near the pole.

The Beaufort is still full of ice.

The Kara is still way behind 2012, 2011, and 2010.

The pattern is still not great for ice loss.


@Chris - probably not going to blow past 2012 this week or next, but that may not matter.

Slightly lower temps at the pole won't offset broken ice and increased albedo.

Kara, while still having extent, is in very poor shape, and should start vanishing before our eyes before July.

The Beaufort has more time to heat up, and even now appears to be racing to catch up.

We are a long way from being able to reliably predict a "bounce".

Paul Beckwith

About 10 years on a scale of 3 for the first 7 years tilting to a 5 for the last 3; these numbers are not for sea ice per say but are for the climate/meteorological system and elements.

Interesting how the NSIDC extent has no downturn:

Sea ice is not acting as a unit, but as an ensemble of independent fractured chunks with an enormous variety of shapes and sizes, all moving at the mercy of local wind and waves. Ice that melts at the perimeters is being replaced by that within the pack, at the expense of pack thickness, and this seems to be presently responsible for maintaining a large extent. We saw this effect also when the ice filled in "holes" in the MYI within a day or two of the cyclone movement elsewhere.

This is a definite negative feedback since a high albedo is being maintained over a larger area; it is also keeping the air temperature above the ice < 0 C. I think that the high extent will be maintained for a while longer and then quickly vertical line downward; I am not ready to change my zero to 7 million km**2 yet...


Blast it...*decreased* Albedo.... I hate not being able to edit...


Ah yes ... I'm a 1 to 2 off and on going back 40 years, and a 3 for the last couple.


A-Team, I withdraw my argument on contrast changing. Not the other arguments though. I don't like your insulting tone. Different people have different areas of expertise. The settled science and statistics cannot be dismissed lightly. Otherwise you would have no defense against the 'Global warming stopped in 1998' and 'The ice is recovering' kind of nonsense.

The same argument (fracturing) is used over and over again. It might well be the dominating feedback. But if instead we will see a 'rebound' year, then the fake sceptic blogs will not hesitate to use this official submission to SEARCH to strike against this 'alarmist' blog.

I raised my concerns. That should be enough. Back to business.

L. Hamilton

Regarding whether results from this cycle should form an official prediction contributed to SEARCH -- I think that remains an open question. I am happy to write up and send off a July contribution on behalf of this blog, if the sentiment here seems to favor that; or not, if people prefer otherwise. It seems there are both pros and cons. What do you think?

Either way I'll post a summary here sharing the results for discussion, and continue recording the data. Looking ahead to October, results from this crowd-source experiment could prove interesting enough to draft a research paper, with community input welcome on that as well.

Kevin O'Neill

Larry - I'd like to see it put out there. We shouldn't fear being officially on record. Each of us has an hypothesis - the data will test that hypothesis. There's no prize for being correct and no penalty for being wrong -- other than pride, self-esteem, and public accolades or ignominy.

While the numbers may be available here for anyone to see -- putting them on SEARCH makes it more readily available to anyone that might seriously be interested. Regardless of whether we strike out or hit it out of the ballpark, the data itself may make a researcher somewhere happy :)

Kevin O'Neill

Self-rating: 15+ years following arctic sea ice. Level of commitment has grown over time to a solid 3.


I rate myself a 2 on the given scale, since 2006.

Chris Biscan

I give myself a 4.

I have looked at the arctic weather everyday since early 2011. Followed others since 2008.

Spent 500-1000 hours at least reading literature on the hardware(remote sensing, ITP's, computer algorithms) Countless hours reading papers. Countless hours scouring the WHOI site and past buoy data.

I have learned that

weather supersedes all.

Albedo is next...

ice thickness is next.

I predicted 4.5 mil km2 preseason for 2011 min.

And busted with 4.3 mil preseason for 2012.

The only region I have ever read or seen that can melt out ice above 5CM per day for any length of time is the Beaufort where Bottom ice melt + Top melt can = 20CM+ per day.

Just because the ice is fractured doesn't mean bottom ice melt in the central arctic basin is going to be enormous.

So far I have seen an incredible utter lack of respect for the weathers impacts on solar insolation.

So the cyclones/cold atmosphere have wasted thousands of W/m2 of incoming solar energy this melt season.

The Beaufort has barely even begun to melt. Snow is finally melted and it's already peak insolation.

This buoy shows no ice melt in the Beaufort:


This one shows top melt and a tiny bit of bottom has started:


North Pole has upwards of 20-30CM of fresh snow still

no ice melt


Chris Biscan

To match 2012 we will need some epic weather and epic melt rates going forward.

To go past 2012 double up the epic-ness.

To get to the 2-2.9 mil Sept min we will need a all out 1030HP Dipole parked over the Beafort/CAB/Central Basin for 40-50 days at least.

To get under 2 million call in an act of God. Or a few small thermonuclear fission explosions within the ice pack to incinerate and obliterate it.

Doug Lofland

I am going to stick with 1.8 million km2, and judging by the open water and melt rings (for lack of a better word)that are starting to form around the pole, this will probably be the year of the North Pole Gyre, which will be a game changer.


This is a serious hobby for me. 35+ years as a 2 or 3 although the first ten were really uninformed. University courses in meteorology, hydrology, glaciology and atmospheric chemistry. Several amateur and overly simplistic GCM models usually using slab oceans. On going climate communications with government representatives.


@ChrisB - All 'Nukes would do is shatter the ice more... no where near enough energy to melt more than a few hundred thousand tons of it.

Take a look at 2012J.

Looks like about 70CM melt since June 1, Even with snow cover.


87N, 24E.

Cold and clouds are not preserving the ice, at least not at that location.


Wrong link. 2012L is in the Beaufort.

Look here:



Is it still possible to revise one's prediction?
I'm considering to revise my prediction up from 3.0 to 4.0...
Things seem to go slow... but on the other hand, all could change pretty quick.
Temperatures seem to be lower, than they ever have been on that record: http://ocean.dmi.dk/arctic/meant80n.uk.php
Does anyone have an explanation for that?


Posted by: L. Hamilton | June 20, 2013 at 01:04

"Regarding whether results from this cycle should form an official prediction contributed to SEARCH --- I think that remains an open question. --- What do you think?"

I think it would be O.K. whether you did or didn't. Could be useful.

I would like to see the top and bottom fifteen (15%) percent eliminated and the remaining 70% averaged.

Chris Biscan

That buoy is wrong. The other ones next to it do not show that


When it comes to comparing 2012 with 2013, one should not forget that about 800-900k of negative anomaly last June came from the regions of Kara, Baffin and Hudson, regions that are quite unsignificant for the final result because there is little export to these regions from CAB and because they will melt out either way. There is also a lot of talk about Beaufort, but has people forgoten about Laptev and East Siberia? These regions are both well below 2012, so much in fact that measured directly in SIA it compensates for the lack of loss in Beafort. The melt this year might have been weakened by PAC, but in my opinion the overall situation looks just as bad this year, compared to 2012, especially when looking at the holes in the pack close to the pole. Come July, and extent will crash brutally.

And Chris B, thermonuclear is a term usually associated with fusion reactions, not fission:)

Account Deleted

@ Chris

"Just because the ice is fractured doesn't mean bottom ice melt in the central arctic basin is going to be enormous...", but sun radiation warms up open ocean, and heat transfer from ocean to ice is proportional to contact surface area, so, a few sunny days... just educated guess.

"So far I have seen an incredible utter lack of respect for the weathers impacts on solar insolation", yes but same respect is due to the mechanical effect of cyclon on ice, the divergence on the ice cap drift field ( div(v) > 0 ) subsequent ice breaking, and ocean upwelling to fill up the void of the diverging ocean/ice surface. This mechanical effect becomes more important when ice is thin... just repeating what others say.

"So the cyclones/cold atmosphere have wasted thousands of W/m2 of incoming solar energy this melt season..." yes it seems so. But do not neglect the effect of years of increased albedo on the Arctic ice, see PIOMAS graph. Compared to that, a few days of snow is... negligible.

My opinion: this year, the local albedo effect on the pieces of the broken Central Arctic Ice can be determinant. If it is sunny for a while...

I keep my estimate of 2.0 M km2

Jai Mitchell

I emailed the people responsible for the buoys, noticed the temperature sensor for 2012J

I received word back about 2012J and the temperature sensor is definitely out but they have a backup. The temperature sensor provides inputs to the snow level sensor so that value will change when they begin using the new temp data.

The respondent also mentioned subsurface melt. There are also 2 sensors for this and they both indicate bottom melt. They don't seem to know why this is happening.

The levels of bottom melt on 2012J is at least twice the amount found at a nearby location (2012E) If (estimated) error sensitivities and time lag are taken into account.

However, 2012E does show subsurface melt and it is unclear why that would be happening. . .

2012J IS located much closer (really on the border) of a major melt patch (is there a name for this yet? surely not a polynya) that is located just to the north and east of Svalbard.


any ideas from the peanut gallery? I have a pet theory. . .


Account Deleted

... oh, and cyclones are ultimately the result of many W/m2 of solar energy, and their number and strength might be increased by greenhouse effect. That said, whether Arctic cyclones become a positive or negative feedback on the evolution of arctic ice volume... scientists will tell us in a few years


@ wanderer: I mean, it depends what kind of explanation you want. The graph you link to shows the sudden cold snap starting around day 120 that first took temps below the average; the easiest response is that that's just weather. Then temps rose rapidly, albeit staying slightly below average, and we can partly blame the PAC for the fact that temps have stayed cold the last few weeks. If you think this relatively long sequence of colder than usual weather is weird or unexpected, either handwave it away as a statistical fluke, add it to the growing list of unusual NH weather, or see it as a possible precursor of a new Arctic summer regime in which the battered icepack finally gets some respite. If what you really want is an explanation based on long-term climate oscillations, forcings, et al., well good luck. Ask the UK Met Office to call another conference, maybe :)

Remko Kampen

Mignonette, on 'artifacts' - I responded to your remark on A-Team's imagery in your post June 19, 2013 at 15:01 .


Don't count out the PAC weather just yet. So long as temps stay enough over 0C to stop things melting, rainwater will do way more damage than the sun. It will take away all the snow, faster than it fell and it will melt the ice from the top too. Faster than from the sun (initially at least).

It was raining on the north pole cams today. There was some weak sun too.

As has been pointed out many times, water carries far more energy than the air. Whilst the sun will warm pools of water on the surface, promoting melt, rain will make those pools faster than the sun will. There is no albedo for rain, it just melts by direct contact. Just like the sea is melting it from the bottom. Direct heat energy transport.

I suspect there is a huge amount of energy raining down on the ice from the PAC. So long as it does not freeze, then low temps are not so much of a big thing. Well that's how I see it.


I think the PAC is still cold enough to be dropping more snow rather than rain. There's no evidence of rain there at the moment. The droplets on the North Pole webcam are more likely snow or sleet melted by the camera's own heat. You can see from the marker poles (and the colocated buoy data) that there's been no snow melt there in the last few days.


3.8 million square kilometres.

And I am a 1.5 for the last 5 years, 3 here... I can follow some of the math, and I watch it year-round (for that matter, is there an out-season any more, especially considering this last year?)

I did not post an estimate in May, but if I had, it would have been 4.0 - I have revised it downward due to 2 impressions:

a) I was impressed that, despite cold temps, 2013 maintained almost the same track as 2012 right into the first week of June.
b) PAC-2013, and the many who expressed concern that the broken ice would tend to melt out quickly later in the summer.

After all I've read here, I've come to the conclusion that extent is least important as a measure of total arctic ice and what it will take to melt the remainder, compared to volume or area. Volume is the closest, but it seems thickness is somewhat overestimated in FYI, so how reliable is the volume estimate? I don't see how minimum volume won't return to setting new records (in spite of the pause it seems to be taking) - has there been any substantive increase in negative feedback, or decrease in positive feedback?

I expect area will set a record as well; I'm not really sure why extent is given more press than area or volume - isn't it reported more for navigational purposes than anything else?

Has anyone actually modeled what is expected to happen during 'the last days', WRT break-up, divergence, top/bottom/edge melting, etc? Would it be a surprise to see extent flatten or even increase just before the ice disappears completely (slushee will spread out more than ice cubes will)?

The bottom line is, I think, if we have a cliff like last year, even extent will set a new record; if there is no special event, it will be second-lowest on record.

Just another amateur's guesstimate, of course.

Ultimately, though, I think Maslowski will be validated - 2016 +/- 3 years is the beginning of the end of arctic ice. Then, the next experiment begins - when will we need to change all the titles to 'Arctic Sea Water'?


Actually it looked like snow that had been rained on to me. Blurred edges, nothing sharp. Slumped around the other camera in view.

Which is why I looked at it as rain.


I’ll revert to my original May estimate of 1.1. mio. sq. km. The reason is, that I discovered a simple error in my calculations. With an estimated 1 mio. cu. km volume left by the end of September, and an average ice thickness of 1.1 m, the area should go below 1 mio. sq. km., whereas the extent should exceed 1 mio. sq. km.

Inspired by some of the recent thinking on this blog, that we might have isolated patches of sea ice spread out all over the Arctic, it seems reasonable to keep the mean September extent in the order of 1.1. mio. sq. km.


On your scale, I would say >20 years and a 5’er. I also noticed, that the early summer cyclone(s) depicted by Neven on another thread, clearly shifted from Eurasia to North America over this past month. In my view, it was driven by sublimation of snow - first over Eurasia and then by sublimation of snow over Alaska. Judging the various satellite products, the only remaining snow cover is over NE Arctic Canada, which fits perfectly with the weather forecasts of a persistent early-Summer Arctic cyclone over the Baffin Bay these next few weeks.

What happens after the last terrestrial snow has gone, remains an open question. If the glacial snow over Greenland will sublimate as well, we should see a high pressure develop over the ice sheet – with surrounding lows fed by the latent heat. Just a few of the models have been able to pick up this pattern, but I have seen examples throughout the last week of a high pressure over Greenland and a low pressure over Svalbard. As predicted last year, this configuration will empty the Arctic basin of MYI, and we will only have patches of first year ice surviving until the “End Game” next year.

Please keep an eye open for the convergence of wind fields north of Greenland instead of all this radiation balance stuff…

PS If you wish to compare estimates of extent with experience, you may simply multiply years in the game by knowledge, so I would end around 100.


Remko Kampen, the remark you refer to states: 'A-Team's imagery is not immune to artifacts like melt pond formation'. There is nothing wrong about that statement. Anyone can see melt ponds appearing and disappearing in some of the coastal regions in A-Team's animation of June 17 at 23:22. This has clearly nothing to do with the cyclone region close to the North Pole where divergence and fragmentation have occurred.

You are causing me to intervene with this thread, while I was only planning to 'lurk' from now on. Stop doing this, please.



It seems as if your "friend" Mignonette in Brussels has deserted you. If she works within the NATO Headquarters, it's fine with me, but please make sure you choose your friends more carefully next time.


Just to be clear on my last post. I am not questioning at all that the divergence and fracturing events in the cyclone region are real, and that they will have a very big impact. However, Remko Kampen was suggesting that I called them an 'artefact' in the images. My use of the word 'artefact' referred instead to the melt ponds at the coastal regions. That's what the post was about.

I will stop commenting now on this blog. This is getting too personal.


Yes, stop misunderstanding each other.

Chris Biscan

So most of you expect all of the FYI to melt and at least half of the MYI?

all in roughly 80-85 days?

Even though SIV is higher than 2012 and 2011 and the weather so far has been much better for ice preservation?

Why didn't this happen in 2011 or 2012 during more ideal melting conditions?

Account Deleted

There's a very nice plot posted recently by user bornfromthevoid in the forum that represents the "dispersion" of ice (extent - area)/(extent+area). 2007 and 2012 values are way over the rest of years, and 2013 shows the same exceptional values. That parameter seems to me a remarkable measure of how broken the ice is. But maybe 2013 does not go the way of those years in minimum extent, who knows.

R. Gates

The thick yellow line roughly shows where the September 2013 minimum will fall:

Everything outside of this, except for some minor tidbits out the Fram is...toast.

Chris Biscan

Open water all the way to Greenland is a big commitment.

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

Your Information

(Name is required. Email address will not be displayed with the comment.)