PIOMAS April 2013 - extra update

The Polar Science Center has released some extra PIOMAS gridded data that allows smart bunnies like Wipneus and Chris Reynolds to show how ice thickness is distributed around the Arctic. Here's a thickness distribution map made by Wipneus that shows the difference between March 2012 and March this year:

It seems the situation has basically flip-flopped, with thicker ice now off the Siberian coast and thinner ice in the Beaufort and Chukchi Seas. This is most probably due to those big highs that caused a very strong Beaufort Gyre that pulled at the ice pack so hard it cracked all over the Beaufort Sea and beyond, but at the same time made the ice compact against the Siberian coast. This thicker ice along the Siberian coast is at a relatively low latitude and is bound to melt away anyhow, although it will be interesting to see when the Northern Sea Route opens this year.

Last year I was expecting thicker ice in the Beaufort and Chukchi Seas to protect the centre of the ice pack like had happened in 2010 and 2011, but I was completely wrong on that. The ice just disappeared at a rapid pace, an amazing sight. If the above map is correct we don't even have thick ice there now, and so I'm with Chris Reynolds when he says: "I expect an aggressive melt from the first week of June onwards, when the melt really starts within the Arctic Ocean."

Speaking of which: Chris has produced this thickness distribution map for March 2013, with next to that a table that shows the division between ice that is thinner than 2 metres and ice that is thicker than 2 metres since 1978:

According to the PIOMAS model there's even less thick ice than in the previous two years, which makes sense, of course, after last year's record smasher. There's now more first-year ice than there has ever been since the record started. Some real weird weather is needed to prevent this coming melting season from becoming a serious pretender to the throne.

To top this update off I present to you Andy Lee Robinson's latest PIOMAS vid:

 

Comments

[Craig Merry]

Distribution of ice thickness through the years is disturbing.

Nice work contributors!

[A-Team]

Wow .. to Andy Lee Robinson. More people will get what they need to understand about the Arctic from this than ever would from dry underlying excel tables.

Neven, I'm not quite understanding your post.

Are you saying that Piomas has gone on for ten years *not* releasing nor journal-publishing gridable data?

That's shocking -- this is inherently geospatial data that never should have been dumbed down to a single summary ice volume digit in the first place.

Are you saying Chris and Wipneus played an important role in getting gridded data out into public view? If so, that's a remarkable achievement.

Now we just need a proper non-dithered image in stereographic projection -- and we are off to unchartered scientific territory.

[Kevin McKinney]

Pretty cool, all right. And maybe flash the name of the month as a title while the years go by to form the curve for that month? That'd be more intuitive than referring to the color key.

[Erimaassa.blogspot.com]

what's up, doc, with the lagomorphic rhetorics in the beginning? :-). Andy Lee's spiral reminds me of some water turbine configurations. might even be somewhat efficient if one figured out the places for the axle and stator.

[A-Team]

Hmmm, that March difference map is rather interesting. Sounds like Wipneus has lots of years sitting there neatly in R, ready to png.

So it wouldn't be too hard to animate the 2012-13 difference map since freeze-up, or for that matter, 2012-2011, 2012-2011 and so forth, whatever people thought was interesting.

Also, along the lines of what Chris did, animate straight ice thicknesses, say back to 2005 (to provide context for 2007). That would take some 96 frames, not a big file at all because of the small palette and common masking.

As it stands, Wipneus has quite a pretty palette. However a scientific image is not a final goal, only a soccer ball we kick along to the next person to take farther downfield.

Here we would next want to cross-correlate ice thickness with ice-penetrating radar (aka dielectric constant, melt moisture, salinity), probably using Ascat grayscale since we have 4 years of data from the same instrument and QuikSat after that to 2002.

Scientific palettes are supposed to drop down to a ramped grayscale. (Usually people start with a grayscale gradient and add their favorite hues -- or just take a screenshot of one they like online.) The dozens of commonly used arithmetic and logic operations on image pairs (beyond simple differencing) are most easily and intelligably implemented once there in grayscale.

Looking at the sub 5-bit palette above with the three common methods of discarding color (lightness, luminosity, average rgb), this doesn't work out as a monotonic ramp. So it would take 24 replacements x number of months to manually straighten out the key.

[Greg Wellman]

A-Team, a couple of years ago the trouble with PIOMAS for "civilians" was that it was only gridded - it was popular demand that led them to put out a daily single summary number. I don't know what "extra" information was released that Wipneus and Chris used. Perhaps one of them can explain what it was that was beyond the normal PIOMAS output. Or maybe it was just the normal output, but the latest batch for March.

Erimaassa, maybe Neven is a fan of Eli. (http://rabett.blogspot.com/) :-)

[Chris Reynolds]

"Are you saying Chris and Wipneus played an important role in getting gridded data out into public view? If so, that's a remarkable achievement."

No, as Gregg notes the gridded data precedes our individual strands of work. However the PIOMAS team have noticed what is being done with their data and have been good enough to release the thickness data during this year rather than waiting until next January.

Gregg,

The data we've used is merely PIOMAS gridded data which dates back to being additional material for a 2004 paper.
http://psc.apl.washington.edu/zhang/IDAO/data_piomas.html

[Neven]

Neven, I'm not quite understanding your post.

Are you saying that Piomas has gone on for ten years *not* releasing nor journal-publishing gridable data?

A-Team, all I know is that PIOMAS has come increasingly into public view since 2010, also or perhaps mostly because of work done via the ASIB (by FrankD and Wipneus at first). Throughout this time the PSC has been most cooperative and very accessible.

If I remember well, I believe it was Chris Randles who found the gridded data and got permission to use it. He then shared it with the others. I forgot the details because I'm not capable of doing anything with the gridded data, but I believe this is how it went.

[John Christensen]

Wow, great work by Wipneus and Chris!

Neven,

Also helping to explain the difference from last year should be the consolidation of ice along the CAD and Alaskan north coast that happened in the winter of 2011/12, which is then contrasted by the strong Beaufort gyre this year.

If anything positive could be said, it should be that while SIA right now is equal to 2012, the SIE (according to ROOS) is less than last year, meaning the ice pack is relatively more compact this year, with high latitude areas faring better than last year - even if cracked.

East Siberian Sea only melted out completely in '07, '08 and '12, and this year stands a chance of retaining a minimum ice layer to help reduce the heating of Arctic surface waters during the main summer months.

But as many have commented, the ice is broken, and may just melt out at an unprecedented rate this year..

[Wipneus]

A-Team,

A March 2013 thickness map with polar stereographic mapping (south is 40 degW) here:

https://sites.google.com/site/arctischepinguin/home/piomas/grf/pg_3_2013.png

Yet another palette. I must admit that choosing colors is not among the best talents that I have. Finding something acceptable keeps causing lot of headaches again and again.

In this case I used colors from http://colorbrewer2.org/

These are supposed to be high quality, hand picked colors suited for maps.
I used "Spectral" for the march_2013-2012 map, and "RdYlGn" for the march 2013 one.

PS. A movie of all PIOMAS thickness maps can be viewed here:
https://www.youtube.com/watch?v=G1TLzgSlGtQ

Needs to be updated beyond 2011.

[Steve Bloom]

This very interesting Mail Online article from yesterday just came across my screen. Apparently UK Met Office chief scientist Julia Slingo is calling a big international meeting to discuss the implications of the recent jet stream changes. Jennifer Francis is mentioned.

It's interesting to see the Mail playing this basically straight. Apparently enough funny weather tends to focus the mind.

The contrast with the just-released NOAA report on the summer 2012 drought in the northern U.S. plains could not be sharper, although I suppose there's no reason to expect that the process of a new scientific paradigm taking hold would be very pretty.

And until now Julia herself was in the business of tamping down public discussion of this. How times do change.

[Artful Dodger]

Hi folks,

Dr. Axel Schweiger, Chair of the Polar Science Center at the University of Washington and a principal investigator for the PIOMAS model, has followed our efforts at the ASI blog from nearly the start, making his first comment here on Aug 6, 2010, just 2 months after Neven's first blog post.

Axel drops in infrequently to make a suggestion, help steer our reasoning, or to provide an update on activities at the APL-UW and the PSC. He has also been extremely gracious in making PIOMAS numerical results available for our discussions.

We are extremely lucky to have his support, and that of many, many other members of the polar science community.

Thank you all. You are AWESOME!
--
Cheers,
Lodger

[Steve Bloom]

Apparently ITV was first with this news. Article and video are here, including footage of Julia Slingo. She seems nearly alarmed. :)

ITV has been running a series of related reports this week, so be sure to have a look at those.

[A-Team]

Indeed Lodger, we are all in this together.

Here is a quick conceptual overview comparing March 2013 Piomas with the 31 Mar 13 Ascat radar. There is good but not total agreement (maybe I should have used some other March date for Ascat?).

I had to upsize the Ascat by 105.03% and rotate by 4.64º. These oddities arise from some choices made long ago in choosing the Piomass curvilinear coordinate system. Normally things are aligned, like Ascat, Jaxa, Modis etc to some lossless variant on the Greenwich meridian.

The top image posterizes (8 ice classes) the enhanced (equalized, colorized) bottom Ascat image. Now that I have the masking and registration down, a lot more can be done quickly.

[A-Team]

I wish ITV, Julia Slingo and the other scientists my best in getting the message out. The people in my circles just do not want to hear it:

[Robertscribbler.wordpress.com]

@ A-Team...

Too much effort for most people. And it's somewhat understandable when you realize that we're mostly captives to a massive political and industrial resistance to positive change. Too many predatory, self-interested institutions. Too many predatory, self-interested individuals. Both force a bad outcome. It's a lesson that crops up again and again and again in human history. But, try though we might, it seems we've been unable to remove the chimp from the man.

Sorry for the rant...

Best,

-R

[Chris Reynolds]

I've stressed that I don't want the PIOMAS team to put themselves out too much. But there is a good chance we'll have a gridded PIOMAS ice thickness update in August, to cover up to July, and In October to cover up to September. The team are not financed to provide us this data, but Dr Schweiger said it should be possible to provide that data.

I chose to ask for up to July because June and July are the months with the biggest ice volume loss, and conditions in July arguably set the scene for the late summer melt. And I asked for up to September because that enables PIOMAS data to be included when we look back at this melt season, rather than adding the gridded PIOMAS data afterwards.

I can confirm what Lodger recounts, this data has been released because it has been noticed that good use is being made of it.

Thanks for that Steve,

In studying the Arctic I think we're in on the greatest science story of the early part of the 21st century at an early stage. I've had loads of hobbies, this one blows the rest out of the water for challenge and excitement.

[David vun Kannon]

Neven, your comment "It seems the situation has basically flip-flopped, with thicker ice now off the Siberian coast and thinner ice in the Beaufort and Chukchi Seas." is, I think, mistaken. Since the image shows change in thickness, what it is saying is that the ice off Siberia has thickened the most since last March, while there has not been much change in the MYI north of Greenland.

[Espen Olsen]

Chris;

I had a chat with one of my personal friends today, he is a member of a Doctors club containing some 200 since +/- 1800, although he has been shaking his head the last couple of years listening to me, he now starts to "comprehend" what I am talking about, so I understand your excitement!

[Neven]

David vun Kannon, I see what you mean, but what I wanted to allude to was the difference with last year, where the ice was really thin on the Atlantic side of the Arctic (Kara and Barentsz), and thick - or so we thought - on the Pacific side of the Arctic (Beaufort and Chukchi). North of Greenland and the CAA don't come into my flip-flop comparison.

This year it seems it's the other way around, and that's because the winds blew from Siberia to North-America, and this year, or at least since January, it's more or less the opposite. At least, that's what I think. I'll try to show it in my upcoming Winter analysis.

[A-Team]

Here I took March 2013 Piomas ice volume as imaged so nicely by Wipneus, converted it into staged grayscale (top) by color replacement, with six units of separation between grays (which fully utilizes the color legend). The only trick here is to reserve pure black and white for later (removable) use somewhere else, for example as text. This is now set up 'to do the math' over Ascat grayscale (later).

I then tried overlaying the 31 Mar 13 Ascat radar ice classes over color Piomas (middle), but it was hard to see what was going on, even if I dimmed down the Ascat.

It worked quite a bit better just to outline the Ascat ice class boundaries and overlay those on color Piomass (bottom). Here the satellite image had too much detail for this purpose so I applied a gaussian blur of 7 pixel radius so that classifier outcome (posterize) would be simpler.

This came out pretty well. It highlights missing on-the-ground resolution in Piomas, for example goat's head, a very distinctive multi-year ice feature that has been stable for years.

[Espen Olsen]

It must be pretty frustrating to a Ice Scientist now a days, with this site working 24/7!

[Dr Tskoul]

Amazing since armatures are providing peer pressure. I hope their products meet our expectations. All is healthy competition .

[Jim Hunt]

Following a long day at TEDxExeter, I'd just like to point out that I beat the Mail to the punch on the Julia Slingo story:

http://econnexus.org/met-office-admit-our-climate-is-being-disrupted-by-the-warming-of-the-arctic/

and here's how the Mail Online goes about manipulating constructive climate change comments:

http://econnexus.org/the-strange-tale-of-the-mail-and-the-snow-dragon/

[Espen Olsen]

Dr Tskoul;

I agree, competition in knowledge is always superior, and I hope we "amateurs" are setting some pace to the information coming from the "pros", because that is needed!

[Yvan Dutil]

Hot from the press: The latest prediction frm NOAA.

http://onlinelibrary.wiley.com/doi/10.1002/grl.50316/abstract

[A-Team]

Awesome video from 1978 to Dec 2011 by Wipneus, worth repeating that link:

https://www.youtube.com/watch?v=G1TLzgSlGtQ

Below I sliced out April the current month for the last 33 years for people wanting to make that comparison:

[A-Team]

"Hot from the press: The latest prediction from NOAA."

I read the full text of Overland and Wang from Seattle Noaa. It is well-written review chit-chat rather than new research per se.

I hadn't realized that data was a poor substitute for model ensembles because it is ruined by dubious 'natural variability' and that sea ice volume is an iffy measure compared to the demonstrated expertise of general climate modelers. Surely they're both the other way around.

The authors average out 2020, 2030, and 2040 to get "very likely timing for future sea ice loss [to 1 million sq km] to the first half of the 21st century.

That's right: the middle ground is 2030 but they can't bring themselves to admit it, I would assume because they were on the disasterous 2007 IPCC chapter.

I don't agonize over weighting approaches myself -- after manifestly total failure of the models with no end in sight, I'm comfortable assigning them zero weight. And of the three sea ice metrics, there can be no doubt that volume is the only way to go.

In summary, some senior modelers who got it dead wrong have now started the process of walking back their previous mistakes a couple decades at a time. I wonder what their shape-shifting story will sound like come September 2013.

http://onlinelibrary.wiley.com/doi/10.1002/grl.50316/abstract

[Robertscribbler.wordpress.com]

The modelers missed warm water upwelling, I think. Didn't count on local areas melting during winter. Air temp is not what's most important. The people on the ground know it.

[jdallen_wa]

Based on my understanding of what I'm seeing, as a systems analyst, even without a full and coherent understanding of the details of the models and environment, I believe we are seeing a system in "tip over" to a new energy state.

I've been watching the climate for over 40 years, as an enlightened amateur. The next few years are going to be spectacularly unsettled, climate wise. How soon it stabilizes (if it does...) is another question.

[Artful Dodger]

Hi Espen,

Freudian slips are bound to happen on the sea ice. ;^)

"Armateur" it is! since fore-warned is four-armed.

[Paul Beckwith]

@A-Team
I agree completely about applying zero weight to the models as they stand. However if you change the time-frame of the models, specifically change the year 2100 to 2020 or so then there is some validity to the prediction. And we must remember that these models are similar to those that NOAA is relying on in their erroneous claim that sea ice is not responsible for the US drought conditions.

I am glad that it is becoming perfectly clear to many people on this forum and elsewhere that what we are observing real-time is an abrupt change in the ice regime which is directly causing an abrupt change in the entire climate system. Something that I have been studying for my PhD work for the last 3 years.

Since November of 2011, a group of Brits formed the Arctic Methane Emergency Group (AMEG) www.ameg.me which I discovered and joined in Dec, 2011. We have hashed over, and over again the details of the abruptly changing Arctic situation and our small group has done everything in our power to inform the public, politicians, and scientists about the urgency/emergency facing humanity as a result of this abrupt change. Our meetings with members of Parliament in Britain were stymied by the Met. Office and others who blindly followed models and ignored the data; likewise in the Canadian Parliament.

Nobody really listened to us (AMEG)or at least it seemed that way, and now that the physical situation is clarifying as we have expected for at least the last year you may understand why AMEG has been proposing seemingly desperate geoengineering schemes.

I have had much internal conflict over whether to publish papers or just do rapid turnover blogs on my research. On moral grounds I decided, rightly or wrongly from a career point of view, to concentrate on blogs and advocacy to get information out there to the public. And to consider the whole system and avoid specializing on only one element of the system, like 99.9% of scientists do. Most scientists keep their work/ideas close to the sleeve until it has been written up in a journal, but there is a large time lag on this. I have felt over the last number of years that this lag was not morally justifiable, so I just blog on new ideas as quickly as they appear.

With the rapidity of the Arctic changes and knock-on effects around the planet writing peer reviewed science is just way too slow and cumbersome to understand the overall system and what is happening in the big picture. From what I see, the cutting edge of science/knowledge/understanding on the sea ice is right here, in Neven's blog.

I have a proposal for contributors here. How about writing papers from the data in this blog. Detailed observations, analysis, with numerous contributors. Similar to CERN and other nuclear physics papers that routinely have 30 or 40 or even more co-authors. Also similar to engineers around the world cooperating in the design of the new aircraft. If you are interested I can give a detailed outline of what I have in mind for the first paper. I only ask that I am first author (my group effort idea and outline and big-picture context and I will be the coordinator) for the first paper or two so that I can fulfill my requirements for my PhD thesis. I think that a paper a month or so is very achievable.

Who is in?

[Paul Beckwith]

@jdallen_wa
Agree completely. We have left our stable climate system that humans have enjoyed for over 10,000 years (400 generations or more) and are transitioning abruptly to a different state. There is great precedence in the paleorecords for such abrupt regime change. For example, the D-O oscillations involved transitions over Greenland of from 10 to 16 degrees C in less than a decade or two.

[Paul Beckwith]

@neven
I had a lengthy post just prior to the one above. It appeared, and then just vanished into thin air. Is there any way to recover it?

[TypePad has been working with a new spam filter system for a weeks now, and hopefully it will improve soon. Quite a few comments get stuck in the spam filter. I try to release asap.]

[Artful Dodger]

Hi A-Team,

Overland, James E., and Muyin Wang. "Future regional Arctic sea ice declines." Geophysical Research Letters 34.17 (2007).

From the Abstract:

"Based on the selection of a subset models that closely simulate observed regional ice concentrations for 1979-1999, we find considerable evidence for loss of sea ice area of greater than 40% by 2050 in summer for the marginal seas of the Arctic basin."

Hmm, just 5 year to a 50% decrease in SIE. It seems the De-Sanctification Process Has Begun!

[Werther]

The Overland/Wang letter ...

After reading the abstract and a considerable part on three approaches to the challenge, I’m left with the feeling that we’re quite alone in the face of imminent danger.

I may be an alarmist, a Cassandra case...

The biosphere can not be sliced up into neat, parallel systems. They don’t function in an independent state. In fact, we may have given them individual names, posed attributes to them, but do they ‘exist’ at all?
We view the world as a form, from which we carve out profit. Meanwhile, we have no idea of the meaning (I’m adressing myself, mostly...). Thus the representation accords to our confounded collective reality.

One of our attributes/representations is the supposed uniform sensitivity of the biosphere to forcing. It is a creation of our licking, probing, overrated scientific method. In a way, this misses what several of our group have been calling ‘black swans’, or what Asimov once literated as ‘The Mule’.
GCM’s are missing the intrinsic, immanent aspect of the Arctic and it’s sea ice. The process is inducing consequences; it shapes other forms and makes our uniform climate sensitivity useless for prognosis. Elegant mathematics are not going to get us out of the mess.

The date of summer ice free, or our self-nominated numeric treshold on extent or volume are completely irrelevant. Consequences shape up to the synchronistic nature of existence.
We witness them already; some gifted people (Dr. Francis comes to my mind, but there are more of course) hint at their interdependency.

But, we want to be sure, to be in control...

Dear friends, there’s no comfort in September extent staying above its numeric value last year.

Forgive me if I’m way beyond my personal limits. I’ll back off to the Jakobshavn Isbrae, the Laptev scintillation and lurking MODIS. That’s how I can create my own little bit of meaning to what’s going on.

[Jim Hunt]

It seems we're not entirely alone Werther:

My #TEDxExeter question to an English stranger is: Are you worried about the #Arctic #SeaIce?

— Jim Hunt (@jim_hunt) April 13, 2013

[R. Gates]

Werther,

I enjoyed your rant and think your wholistic approach is of course quite on target. It also gets to the reason why models will always be wrong, and the faster a whole system is changing, the more wrong models will be. Models carve out boundaries in "systems" when there are neither boundaries nor systems in actual reality but one continuum. It is our human brain that must impose these, and this becomes a major source of confirmation bias. But all this of course does not mean that models such as PIOMAS can't be very useful in describing bits of dynamics within the continuum, but we must always keep in mind that in regards to the whole, they are always just models and therefore doomed to be wrong in little and bigger ways.

[A-Team]

I've also been watching Nares, Jones and Lancaster Sound. The latter is the eastern end of the Perry Channel (and so key to the media's favorite glass half full -- slightly more profitable shipping for Maersk Lines).

These areas are not good proxies for the 2013 season. Lancaster didn't really froze up properly and has gone back and forth, Jones Sound has receded ... and you can see Hell Gate and Cardigan remained open water.

As you can see from the image below, the northern edge of Nares has not budged since the 1st of March. While it is noticably more advanced than the same dates in 2012 and 2011, the Nares was much farther along throughout 2010.

It wasn't worth an animation so I just drew in some boundaries on the best cloud-free days.

[Shared Humanity]

Wipneus....It is interesting that on your PIOMASS Monthly Arctic Average Ice Volume chart the months that show the most rapid decline are May through July. Is this a result of an earlier onset of melt? Why don't we see evidence of a later onset of freeze in the fall?

[A-Team]

Just to repeat our monthly reminder (Aaron has made this point over and over) -- it's a whole lot easier to see something with the laws of physics than with statistics. And in some areas of climate science, we can do just that with complete certainty.

Here we are talking strictly about 19th century physics ... Newtonian inertial mechanics, second term calculus, ideal gases, pv=nrt, gravity, maxwell eqnts, thermodynamics, rotating coordinate frames. The junior undergraduate curriculum.

Let's just say nothing here has been up for discussion for well over a century.

Read the 1896 paper by Arrhenius if you haven't already. It could be published again today with no one noticing anything odd about it.

[FrankD]

Thanks for the hat-tip, Neven. I must say, I laugh to look at my original efforts to find a bit more "texture" in PIOMAS results when I see the truly incredible work done by Wipneus and Chris (and quite a few others) since then.

PSC's helpfulness has greatly enhanced the quality of the analysis posted here and thus the standing of this blog. And I like to think this blog has helped advertise the quality of their work to the wider world. So five cheers all round!

[Shared Humanity]

Paul Beckwith....."And to consider the whole system and avoid specializing on only one element of the system, like 99.9% of scientists do."

First a caveat...I have been a lurker here for almost a year and am amazed by the work done here.

It is not at all surprising that scientists specialize and develop a deep understanding and perhaps breakthroughs in their specialty. They then publish to explain a perfectly valid perspective or understanding of the thing studied.

I am currently doing research on the "nature of knowledge" (understanding) and learning. All knowledge is subjective, influenced by the perspective of the person engaged in the study of a thing. Each of us will develop a unique understanding of this thing, again influenced by our perspective(prior knowledge and the manner in which our personal observation, using our senses, provides new information about the thing studied).

Breakthroughs in understanding (real learning...read Thomas S. Kuhn, "The Structure of Scientific Revolutions")of the thing studied occur through dialogue as the unique perspectives of well informed individuals are shared and a common, more complete, understanding of the thing studied is arrived at.

This community and others like it are uniquely positioned to contribute to these breakthroughs in understanding. This is not an indictment of the scientist. In fact this community depends on scientists to inform the dialogue that will deliver the breakthroughs. These breakthroughs are, in fact, a synthesis of scientific research.

I am commenting here because I believe the work done here advances our understanding in a way that the scientific community cannot do as readily. I encourage all of you to continue to engage in a lively and sometimes contentious dialogue that is informed by an understanding of relevant science. As I lack any real knowledge to contribute to the science, I will now go back to lurking although I will occasionally show up to ask stupid questions.

[Shared Humanity]

One last comment on the nature of learning. My avatar highlights the importance of reflective listening as a foundation for learning. Dialogue is not each of us, in turn, merely defending our understanding of the thing studied but, as active listeners, we need then to then reflect on the new information and construct a deeper understanding by using this information to build on (constructivism) a more accurate understanding.

[Wipneus]

SH:

Is this a result of an earlier onset of melt?

Yes, we first saw this in the downward "spikes" in the official PIOMAS graph:
http://psc.apl.washington.edu/wordpress/wp-content/uploads/schweiger/ice_volume/BPIOMASIceVolumeAnomalyCurrentV2.png

Why don't we see evidence of a later onset of freeze in the fall?

Spring and fall are not symetrical. Part of the reason of more spring melt is, no doubt, the lower albedo caused by thinner ice, more open water, less snow on the ice, contaminated ice, more leads, more ponds and perhaps more. You need sun energy for that, and there is more in spring than in fall.

[A-Team]

Paul, a bit of advice:

(1). Finish your doctorate in a climate-relevant field. Without a degree, you are pidgeonholed forever as an enthusiastic activist, from the reporter's perspective, equivalent to some untutored dufus at the Cato Institute.

The more your wikipedia bio looks like this, the better:

Romm, Joseph Jacob. "Applications of Normal Mode Analysis to Ocean Acoustic Tomography". American Doctoral Dissertations, Source code: X1987, Massachusetts Institute of Technology, 1987

(2) Nobody is going to do your homework. The dissertation must contain original work. Since you are impatient to get on with activism, the trick is doing both.

First, cut way back on the scope. Forget coral, drought, penguins: the Arctic alone is a very full plate: sea ice, methane, tundra, Greenland.

Second, pick a breaking event (Sept 2013 sea ice) -- not a thousand articles on it already, easier to be original.

Two-thirds of a dissertation is reviewing previous work (so your committee knows which part is original).

Here I would start by chasing down exactly what went wrong with the Arctic Ocean component of all those climate models. They're very cagey about disclosing assumptions. This will take real work -- it may even be buried in code. (The reason: the assumptions stink, the modelers know they stink.)

What went wrong is going to be a very hot topic come September. And there you are, on the day of Sept minimum, emailing off your thoroughly-documented polished review article for Cryosphere Today explaining exactly what went wrong, why it can't be fixed, and where do we go from here (the modelers will say, oh just give us the new initial conditons, models are still great).

The rest of thesis goes on to show the models aren't working any better on subsea methane, permafrost methane, tundra albedo, or Greenland melt. The advantage is, you are learning details all along the way that are useful later.

(3) Figure out how to maintain a recognizable institutional affiliation. An edu is preferable but I don't see anything wrong with Sierra Club or AMEG for that matter.

(4) It shouldn't matter but it does: people buy the book by the cover, wine by the label, clothes make the man, on and on. You're off to a great start with the perfect mug shot (is there a web site for these?) -- bookish, balding, friendly, engaging, square.

Last thing you want here is the activist look: dreds, nose rings, food stains on shirt, tattooed knuckles, fly open, holes at knees, filthy toenails sticking out from mid-winter jesus sandals. People won't get past that, regardless of content.

(5) Think longer term on reputational consistency, the internet tracks you forever. I've seen you post all over the map, from sensible well written reviews to ill-considered broadsides on nutter bioengineering that you really don't want to be associated with.

[Neven]

Shared Humanity, this is one of the reasons of this blog: to increase awareness and let knowledge concerning the Arctic become common in the collective consciousness. I'm publishing a post on the development of this phenomenon tomorrow.

[Neven]

Last thing you want here is the activist look: dreds, nose rings, food stains on shirt, tattooed knuckles, fly open, holes at knees, filthy toenails sticking out from mid-winter jesus sandals. People won't get past that, regardless of content.

Which is why I'm not showing pictures of myself. :-P

[Jim Williams]

I have one general criticism of A-Team's analysis of how to get ahead -- time. He'd be right if we had another 50 years of "nothing new under the Sun," but I really doubt that whatever it takes 10 years from now will at all resemble what it took 50 years ago. This will be true from the utter collapse of the copyrighted journals to the rejection of learning in schools.

[Chris Reynolds]

Wipneus,

"Part of the reason of more spring melt is, no doubt, the lower albedo caused by thinner ice, more open water, less snow on the ice, contaminated ice, more leads, more ponds and perhaps more."

Actually the anomalous spring melt is not driven by albedo, it ceases by around late June, so happens too late for albedo to play anything but a role in the last few weeks. Dr Schweiger has today confirmed that PIOMAS does not vary albedo for the age of ice, it is clear from the papers that there is no difference in PIOMAS between MYI and FYI, apart from the differences that arise from thickness. Furthermore ice contamination from, for example, black carbon is not modelled in PIOMAS, and has no presence in the NCEP/NCAR forcings used to drive the ice/ocean model.

The anomalous early spring melt occurs to a degree in post 2007 years, however it is most marked (highly anomalous) since the 2010 volume loss. The 2010 volume loss is found by Schweiger et al 2011 to be outside the uncertainty bounds for PIOMAS, and as I have argued on my blog there are other reasons for considering it to be a real event. So it may be reasonable to assume that the spring melt anomaly is happening in the real world and is the reason behind the change to summer area loss seen in CT Area anomalies for 2011 and 2012, which are more similar to the CT area anomaly for 2007 than other years.

Informed by suggestions from both Dr Schweiger and Dr Zhang I am looking again at the influence of thickness. The PIOMAS model thicknesses given in gridded data are 'effective thicknesses'. I had been thinking of them as average thickness. However because thickness is modelled using a probability density function, g(h), to simulate different thicknesses of ice in a grid box (allowing sub-grid parameterisation), the effective thickness seems to be the integral of this density function. In short I now suspect that the post 2010 thinning is allowing the lower tail of the density function to 'reveal' more open water, simulating leads at a sub grid level (Dr Zhang's suggestion). And it may be this that is leading to the more aggressive melt in the spring.

Really I need to read Thorndike 1975 "The thickness distribution of sea ice", but the damned paper's paywalled, I can't find a free copy anywhere.

And I don't want to disturb the moths in my wallet this early in the spring. ;)

[SATire]

I think Paul got an excellent advice from A-Team. The only thing I would like to add is: Convince your Prof. At least in my country that is most important (we call him "Doctor-father" for a reason). Some similar law or rule exists all over the world - so play according to your rules.

[HeisenIceBerg]

Chris Reynolds,

Here's a copy of Thorndike et a. 1975. If for some reason you can't get it through this link, let me know and I'll try something else.

https://www.dropbox.com/s/8v97j4q7ex4arjl/The%20thickness%20distribution%20of%20sea%20ice.pdf

This is why still being a university student is great. I have free access to every single journal that I've come across.

[Chris Reynolds]

Heisen,

Very, very, much appreciated! The moths send their thanks too, they can doze a while longer.

Now I know what I'm doing tomorrow morning.

[Fairfax Climate Watch]

here's a variation on the PIOMAS ice volume:

http://climatewatch.typepad.com/blog/2013/04/sea-ice-seems-the-same-on-the-surface.html

It's a daily figure for April 10th from 2013, 2012, and 2011, each broken down by latitude and thickness. The thickest ice is alarmingly low!

I included only the ice above 3 meters thick and above latitude 70N. A reasonable estimation of daily volume can be made with this and the numbers show what the eye has trouble seeing.

I've done a few other days of these (unpublished) and they are all similar so far.

[HeisenIceBerg]

@All

Just for the record, I would be more than happy to grab and share any publications which anybody wants to read but cannot access due to a pay wall. I don't know enough (about climate science, statistics, graphics, or just about anything else that you guys do amazing work with) to contribute much of anything in terms of analysis, but I would love to be able to contribute in other ways. If helping people get access to journal publications is what I can do, then I am open to any and all requests.

[A-Team]

The mid-September NSIDC minimum announcement may bring a rare teachable moment.

On the whole, I think people here are already well-prepared for the 'Arctic Ocean, what's that got to do with me' but there are other Whack-A-Moles out there.

First and foremost, the opportunity to take down GCM, their pretentious non-performing models, their immensely distracting side show on the drip drip drip of CO2, and their disasterously lengthened timeline for the onset of serious climate change effects.

Too many peoples are just playing daisyworld in their cubicles, draining resources. We need more people on the ice, more gear on the ground and more instruments in the air.

The knock-on damage is tremendous -- who is going to believe a scientist from here on out?

For example, take those folks with their exceedingly complex Arctic botanical and soil models, piled on top of twenty 2050 GCM model ensembles -- a house of cards built on top of another house of cards.

I saw an even more preposterous article the other day, basically 'give us supercomputers, we can do finer grid cells, give you more accurate predictions'.

No, actually you can't. You have the Arctic physics dead wrong, so your models can only spew out more rubbish since everything hinges on the Arctic.

Come September, will the climate modeling community fall down on their knees, apologize profusely to humanity, admit to gross errors, retract their papers, refund the grant money out of their own pockets, pull the plug on their computers, and go back to grad school to learn how science is done?

No. That's why we have to get organized -- to give this whole bullsh*t industry a knockout punch come September.

[R. Gates]

Here's an interesting comparison between the rate of melt in the southern side of Nares Straight this year in April versus last year in June:

April 11th Nares Straight:

http://tiny.cc/drwhvw

June 11, 2012:

http://tiny.cc/ywwhvw

One of the areas I'll continue to monitor closely this summer will be where the MacKenzie river meets the Arctic ocean. Here's an image from June of last year:

http://tiny.cc/jpxhvw

It is here (and at other large river discharge points into the Arctic) that we are seeing huge water temperature anomalies in the Arctic, which is understandable, as the entire permafrost under structure that the MacKenzie passes through on the way to the Arctic is melting:

http://www.climatehotmap.org/global-warming-locations/mackenzie-river-basin-canada.html

[R. Gates]

A-Team,

You're right about the house of cards, though a bit harsh on the men and women who built that house, as it takes a certain kind of genius to see a bigger picture and how their models are so limited. I think, you'll know the house has fully collapsed when this man:

http://2.bp.blogspot.com/-DABm_s1V29I/UE95smVuDhI/AAAAAAAACjA/FWl5K2vIvTw/s1600/Peter+Wadhams.jpg

Gives a major speech to the United Nations, as I believe he will within a few years at most.

[Chris Reynolds]

HeisenIceBerg,

Thank you for that helpful and useful offer. Very much appreciated. But one suggestion - don't leave them up indefinitely could incur copyright problems. When I've shared papers via GoogleDrive I tend to leave them up for a couple of days to give people a chance to get them, then remove.

[Werther]

Matt Owens, hi,

I’ve just followed your link and saw the post on West Greenland snow cover.
As I stated above, I intend to stick to concrete subjects(….). I just happened to collect 12 April MODIS images r02c02 around Kangerlussuaq and Sermeq Kujalleq. Couldn’t agree more; the sit around there is very, very different from former “MODIS”- years. Baffin Bay ice boundary shows lots of floes with clear melt-tinges, fjords show the blue-glassy aspect we know from the CAA/Laptev end of May.
On a prelude to melt, things couldn’t look worse for West Greenland now.

[Steve Bloom]

Let's not get too carried away with that, A-Team. They after all are our only means of getting a handle on the likely pace and extent of future changes. Yes, modelers are prone to believing their models, but I think there is wide awareness among them that Arctic amplification is the biggest defect (especially, as I keep pointing out, with regard to extreme Pliocene warmth).

As you know, as a general matter even where they have the physics right the models don't really do abrupt change, i.e. they want to smooth things out. So the sort of transient events we're seeing now should be taken as an indication that the models point in the right direction but in many critical areas will be much too conservative as to the timing. To the extent that lots of people currently believe the opposite (a consequence of early trends being both hard to detect and relatively slow on the scale of human lifetimes), let's focus on using forthcoming events in the Arctic to educate them on that point, not to trash the models.

Another thing to bear in mind is that mid- and late-career climate scientists all began in the business with an assumption (entirely valid given what was known at the time) that warming would proceed relatively slowly and smoothly. Add to that the way they get pilloried for being wrong about the slightest thing (e.g. the Met Office's attempt at seasonal forecasting, or Viner's remark about UK snow becoming a rare thing) and it becomes much easier to understand why people like Marty Hoerling take such a conservative approach. The professional consequences of overstating things are far worse than those of understating them.

[A-Team]

Case in point: NSIDC (to their credit) forthrightly retracted their Greenland melt day story on 18 Mar 13.

Read the article and be astonished at the utter lack of ground-truthing. In 2013?

Melt estimates are based on little more than someone sitting in their cubicle counting pixels from a single channel of the Jaxa radar, the 37H.

I had long suspected this from the visual match of their map to the imagery.

Meanwhile, the really horrific news got lost in the shuffle -- melt water in the snow from last summer never refroze. What happens with that when the summer comes on?

http://nsidc.org/greenland-today/

[HeisenIceBerg]

Chris Reynolds,

Thanks, that's a very good point. I think that I'll start a thread on the ASIF where I'll take requests and post links, so that it's easier to keep track of. Would "Developers Corner" be the appropriate board to start that in?

Regards,
HeisenIceBerg

[A-Team]

R Gates notes "huge water temperature anomalies in the Arctic at large river discharges, which is understandable, as the entire permafrost understructure of the .."

Important point. I was just puzzling the other day (to no effect) about what drives much earlier breakup on both sides of Banks Island, seems like the mouth of the MacKenzie mouth is more 'logical' (see that earlier animation).

The MacKenzie drains one-fifth of the total land area of Canada -- and 75% of the basin is underlain by permafrost so we can expect some major effects on the Arctic Ocean as that melts down.

In addition to the excellent link you provided, AK Betts et al 2003 write that "streamflow at the mouth of the Mackenzie remains very low until May and peaks in June" though their graphic shows some flow beginning April 1st.

I'm not seeing any attributable MacKenzie outflow effect on Modis or AVHRR as of Apr 13, the weather has finally cleared a little.

[Boa05att]

With regards the wrongness of models, and taking account of Werther's Hollistic approach,is it time to look a little further afield?

Such as in this blog post on how:

"The Antarctic Half of the Global Thermohaline Circulation Is Faltering"
http://www.dailykos.com/story/2013/04/10/1200602/-The-Antarctic-Half-of-the-Global-Thermohaline-Circulation-Is-Faltering

Which concludes that:

"increased melting of Arctic sea ice may be related to declines in Antarctic bottom water formation. Likewise, the cool Pacific, warm Atlantic pattern causing increased U.S. droughts and storminess in the north Atlantic may be tied to these changes in ocean circulation patterns. Paleoclimate studies have consistently shown oscillations between Antarctic and north Atlantic bottom water formation and between relative coolness around Antarctica and north Atlantic warmth.

The Arctic melt down that is far exceeding model predictions is connected to the slow down in Antarctic bottom water formation. Climate modelers will be challenged to model the connections and the details. The cooling waters around Antarctica, while apparently good news, are not. The rapid melting of the Arctic will be enhanced."

[Fairfax Climate Watch]

Hi Werther,

yes, and as a matter of fact, there was snow cover on March 14th of this year for that whole area. Then clouds rolled over and when they left on March 17th, all the snow cover was gone, with conditions like you describe. It seems like heavy rain must have fallen.

[R. Gates]

The total input of warm water to the Arctic from river discharges in the late spring and summer is significant. To understand a bit of the dynamics, here's a great reference article focused on the MacKenzie:

http://www.sciencedirect.com/science/article/pii/S0272771410002301

[Alais Elena]

Climate Nexus has a conference call with Rear Admiral Titley, Wieslaw Maslowski, Walter Meier, and Dr. Vavrus

The audio (minus the first 15 min) can be found at this link:

There is a smooth but convoluted process to getting to the audio file.

http://bit.ly/16ftRqf

The most interesting thing I heard was near the end when Maslowski said that the models (and I will paraphrase badly here) do not include the warming of the sea water from the summer in the winter. He said all the models do this.

This would go a long way toward explaining why they are so off.

[Alais Elena]

Sorry, I had to sign in using facebook -- for some reason twitter just refuses.

And I have not the slightest idea why it says I am logged in as "Alais Elena" which is my daughter's name.

Tenney Naumer

[Steve Bloom]

I would think either "Arctic Background" or "The Rest," HeisenIceBerg, but in any event if Neven thinks there's a better location he'll just move it.

Anyway, I have a request which I'll just go ahead and make here since the forum topic doesn't seem to have been set up yet. I already asked the first author of these papers to respond, and she seems to have declined despite James Annan having vouched for me. Her being French, I suppose my chances would have improved had he specified a lack of association with either elderberries or hamsters on the part of my parents. :)

Anyway, the papers are the three listed here.

Thanks so much for your kind offer to provide this service.

[Rob Dekker]

A-team, thank you so much for the link to the most recent Overland and Wang paper.

I hear you when you state "I hadn't realized that data was a poor substitute for model ensembles because it is ruined by dubious 'natural variability' and that sea ice volume is an iffy measure compared to the demonstrated expertise of general climate modelers."

It can be frustrating to see that scientists are reluctant to admit that climate models have been underestimating Arctic sea ice decline.

However, the paper itself makes some extraordinary admissions.
This one for example, from the abstract :

The observed rapid loss of thick, multi-year sea ice over the last seven years and September 2012 Arctic sea ice extent reduction of 49 % relative to the 1979-2000 climatology are inconsistent with projections of a nearly sea ice free summer Arctic from model estimates of 2070 and beyond made just a few years ago.

Really, for a scientist to state this is quite extraordinary.
But it gets better :

Observations and citations support the conclusion that most Global Climate Models results in the CMIP5 archive are too conservative in their sea ice projections. Recent data and expert opinion should be considered in addition to model results to advance the very likely timing for future sea ice loss to the first half of the 21st century, with a possibility of major loss within a decade or two.

Really, for a climate scientist to make this statement is truely amazing in my opinion.

Seems to me that these guys (for good reason) just through OUT the models, and are willing to look at the data and any expert reasoning to make an assessment of what the heck is going on in the Arctic.

[A-Team]

Greg writes, "a couple of years ago the trouble with PIOMAS for "civilians" was that it was only gridded - it was popular demand that led them to put out a daily single summary number.

To summarize, (1) radar backscatter is used to estimate sea ice concentration/area/extent, (2) which Piomas takes as its primary empirical input (not being an ab initio calculator of anything) and (3) outputs a gridded data table of ice thickness which understandably proves too geeky for end users so (4) only releases the summary single volume number on a monthly basis and (5) the gridded data on an annual bases and (6) never once in 108 months of operation ever converts gridded thickness data out of its goofy coordinate system back into the world standard polar stereographic coordinate system of the original radar image so (7) when Wipneus and Chris first did this, (8) making the first sea ice thickness class colored maps, it represented the first time that the ice thickness output map could be directly visually compared to the original rader input map which (9) lead within ten minutes to my initial discrepancy maps between the high resolution radar input and necessarily somewhat blurry Piomass output which (10) are measuring slightly different things, radar signal return and ice thickness but (11) could very well lead to a very accurate correlation function between the two which (12) would allow direct read-off of better resolution ice thickness directly off the radar (13) and a daily ice thickness report for seasonal prediction purposes without waiting for or relying on the eventual lower resolution Piomas image which has served its purpose once it has provided radar imagery thickness calibration.

Too bad we're just realizing this in April 2013 because the shelf life of ice volume prediction-ology, like all that we do, is going to be very limited with the multi-year ice tanking so fast.

Mid-September is only five months off, so predictions don't add much value to the wait-and-see option if this is the last year (ie the heat budget disruption is about maxed out).

On the bright side, the main significance of what we are doing overall with sea ice may come into play (decades or centuries later) in monitoring early stages of sea ice recovery.

So maybe burn a CD of the blog periodically in the hope that the dominant life form at that time (cockroaches?) will be technologically savvy and find value in what we did.

[Neven]

A-Team, I'm carving all blog posts and comments in a cave wall. ;-)

[Wipneus]

Chris Reynolds:

The question was spring vs autumn, not spring vs "after June", aka "summer".

Yes, suggesting that all these parameters are directly in PIOMAS was wrong on my part. Yet some of them are, snow on ice is one. And concentration is another. Thinner ice in PIOMAS comes with lower concentrations:

https://sites.google.com/site/arctischepinguin/home/piomas/grf/pm_thick_conc_6_31-35.png

I would think the concentration is the most important aspect of the grid cell thickness distribution.

[Chris Reynolds]

A Team,

No, actually you can't. You have the Arctic physics dead wrong, so your models can only spew out more rubbish since everything hinges on the Arctic.

Actually they have the broad physics pretty right. For example the Tietsche et al paper reveals why 2008 was a rebound, and what is going on in the Autumn. Model papers are an endless source of illumination on the current state of the pack.

The Arctic is replete with feedbacks and interactions which amplify any slight differences between reality and assumption. Furthermore the atmosphere is acting in tandem with ice and snow in ways that are not yet understood.

I hadn't realized that data was a poor substitute for model ensembles because it is ruined by dubious 'natural variability'[1] and that sea ice volume is an iffy measure compared to the demonstrated expertise of general climate modelers[2].

This in my opinion is a total misunderstanding of what the authors actually say, followed by a misrepresentation not supported by the text.

Point [1]

The question, however, is not as straight forward as simply comparing data timeseries with model results. Global Climate Models (GCMs) are often run several times, referred to as ensemble members, with slightly different initial conditions to simulate a possible range of natural variability in addition to steady increasing greenhouse gas forcing. Data, in contrast, is a single realization of a range of possible climate states. Observations confound signal (global warming forcing) and noise (natural variability). Thus it is not completely valid to compare the ensemble mean of a model or several models, which could be considered the expected value of the climate state, with the single data realization. A better approach is to look at the range of ensemble members and to determine if the data timeseries could be considered a possible member of the population of ensemble members. Unfortunately, there are seldom enough ensemble members to test this hypothesis. The science question becomes: is the observed rapid loss of sea ice in the real world consistent with model ensemble members with the fastest rate of loss? Multiple Groups (AMAP, WCRP, various national programs), as well as the climate research community and the general public, are interested in this question for adaptation planning and as a popular indicator of climate change.

So they're not saying data is problematic, they're explaining how it is best to compare data and model output.

Point [2], I cannot find any support for that being in the paper. For example, from their conclusion:

Available evidence suggests that scientists have been conservative in their climate projections, with a late bias in dates for change (Brysse et al. 2012).

[Chris Reynolds]

HeisenIceBerg,

Not sure where the best place would be. I don't think Developer's corner is the best place. Scientific papers aren't really the issue there. I'd just catch them when you see it. I've already given people the opportunity of getting papers they need via my Google Docs over on the forum, just on the relevant thread. So perhaps just on the relevant thread on an ad hoc basis.

I'll keep on keeping my eyes open, At present I have 260 papers in my Arctic folder and sub folders. I frequently find that I've found a free copy in the past which has since disappeared from the net.

[Rob Dekker]

Personally, I think that the Arctic is in a bifurcation state, as described by Eisenman 2010.

Simplistically speaking, if you imagine an Arctic (such as currently) with a lot of summer ice, then most of the summer insolation will be reflected back to space.

Now imagine an Arctic without any ice during the entire melting season.
The albedo difference suggests that an ice free Arctic will absorb some 160 W/m^2 EXTRA energy during the melting season, and thus represent a completely different state of the Arctic.

The question is, how do we get from one to the other state ?

Will it be 'gradual', or will it be 'sudden' and highly non-linear ?

GCM's suggest that the transition will be 'gradual', but actual data suggests that there is a highly non-linear factor involved, to the point where we may state that the Arctic at this point is simply "instable", and may be in the process of 'flipping' over to a new (ice free) state, just as Eisenman et al suggested.

http://eisenman.ucsd.edu/reprints/Eisenman-2012.pdf

[Jim Williams]

Rob, I think that it is the models themselves which are gradual. That is to say, the Science used to be that climate change is slow and smooth so the models where made to reflect this belief. They do not toss the old model and start over every year because it's too expensive (even if they should), so the models slowly change in the direction of newer understanding. (Note my use of newer and older, not better or worse.)

It's a simple case of Lamarkian Evolution.

[Artful Dodger]

A-Team wrote | April 13, 2013 at 22:22

"Too many peoples are just playing daisyworld in their cubicles"

Hi folks,

Sometimes, it's hard to keep up with A+Team, and he's bound to slip a high fast one past the best of us at times. But let's not be deterred, there's gold in them thar nils!

So, what is 'Daisyworld' to a climate modeler? (click this image to study GeoSci at Penn State U. ;^)

Daisyworld is a simplified model used to calculate the heat budget of the Earth. State your assumptions, apply your formulae, do the math. Compare to empirical observations, revise the model and/or parameters. Rince, repeat, that's how GCMs work. And that's kinda the goal of this discussion.

Fun, huh? (and to think you almost missed it ;^)
--
Cheers,
Lodger

[Artful Dodger]

Hi Jim,

We're not looking at actual global climate model outputs. Instead, we're given an ensemble result from numerous model runs.

That's the heavy black line in the famous fig.4 from Stroeve et al., 2007. The thin, faint lines are the individual model runs of the ensemble members. The thick red line is historical observations: (click the image to get a PDF of the original paper)

Notice that at least one of the models matches pretty well with the sea ice history. However, since that one is weighted the same as all other members of the ensemble, we say the prediction is wrong.

To me, that just confirms only one model run will be the closest. Duh! But our criticism is based on the desire that the Ensemble Mean also match what happens in the real world (more in my next comment explaining why that can NEVER happen).

I say NO! As long as the modeler has chosen a sufficiently wide range of parameters so that at least one ensemble member gets it right, then the modeler has done their job.

What needs to be done next is to examine the parameters in detail, and choose the next set of models to form an ensemble mean close to the last successful model.

Notice that in the 2007 projection, there are NO ensemble members that get close to zero SIE before 2050? Time to recompute!

That's my take on what has happened, and how to proceed in the future. Still it takes courage and academic honesty to do this well.

And I applaud the scientists who are doing it.
--
Cheers,
Lodger

[Jim Williams]

"I say NO! As long as the modeler has chosen a sufficiently wide range of parameters so that at least one ensemble member gets it right, then the modeler has done their job."

It's not clear to me what "their job" means. I guess Climatology Modelling is just doomed to be another dismal science along with Economics. We certainly Don't have anything like Newton's Laws here.

I might applaud the scientists doing the modelling while I shift funding dollars to obtaining actual data.

[Artful Dodger]

In my last comment regarding Arctic sea ice models, I said:

our criticism is based on the desire that the Ensemble Mean also match what happens in the real world (more in my next comment explaining why that can NEVER happen).

Most models of arctic sea ice produce a non-normal distribution for predicted SIE vs time, ie: an exponential or even an S-shaped function. That is, their mean and variance of the function does not follow the classic bell-shape of the normal distribution.

But a funny thing happens when we take those non-normal distributions and combine then in an Ensemble Mean. Let me tell you why.

In probability theory, the central limit theorem tells us that the mean of a large number of independent random variables, each with own non-normal distribution, will be normally distributed.

This quirk of statistics acts to smooth out the resulting curve of the Ensemble Mean, even in the case where every single member of the ensemble shows a steep, non-linear collapse of SIE.

That 'thick black line' in Strove 2007, the Ensemble Mean, will always show a smooth and gradual decrease, when when every single model member is screaming TIMBER! Look out below! :^O

And it's that black curve, the Ensemble Mean, that gets reported as the 'prediction'. Don't be fooled! It can not happen.

Remember, as Duncan McLeod famously said, "there can be only one*" ;^)
--
Cheers,
Lodger

* I suppose they could always report the median curve as the best guess for a prediction.

[Artful Dodger]

Jim Williams wrote | April 14, 2013 at 16:10

"I guess Climatology Modelling is just doomed to be another dismal science along with Economics. We certainly Don't have anything like Newton's Laws here."

Very close, Jim! It's more like Quantum Mechanics, where you can describe the rules of the games but there is an irreducible amount of random chance in every particular event. Chaos is a part of the universe, even though our human brains don't deal with it well.

So here's an better analogy: Craps

• do you expect to win every time you roll the dice?
  
• would you play if you knew the dice were loaded?
  
• do you realize that if you keep rolling the dice, eventually you'll lose?
  
• do you blame the stickman when you crap out?

Predicting Arctic sea ice decline due to climate change is easy: Soon, and dramatic. Predicting weather is just the craps.

Cheers,
Lodger

[Artful Dodger]

"Predicting weather is just the craps."

... by which I mean FYI is the random noise in the system. Watch the MYI. That's the measure of the climate. ;^)

Cheers,
Lodger

[Dan P.]

Artful Dodger: "I say NO! As long as the modeler has chosen a sufficiently wide range of parameters so that at least one ensemble member gets it right, then the modeler has done their job."

I agree, as long as "getting it right" also means "getting it right for the right reasons." For example, in Stroeve et al (2012) they point out that one of the models currently doing OK on September ice extent started out the 2000s with winter modeled extents correct but thicknesses far too thin, meaning that it's still not capturing volume loss properly.

I was pleased on re-reading to see that they're planning to broadly analyze the CMIP5 models for ice thickness comparisons to observation, which should be illuminating.

I understand the reasons to use a model ensemble, but with the wide variation in quality of the models and the different physics that goes into them it's hard to see much use to taking the mean of the whole lot. Better as you say to look at the envelope and see what factors are causing the variations.

I also totally agree with your well-taken point that the ensemble prediction will never show a sudden dropoff even if each model does manage to correctly capture that behavior.

[Chris Reynolds]

Daisyworld is interesting; a model needn't have a close correlation with reality, either in terms of leaving out second order factors, or in terms of match to real world progression of the process for it to tell you interesting and useful things that can be used to inform examination the real world.

AD, DanP,

Years ago Connelly was making the point that the real test of a model is that the real world observations compared to the model be statistically indistinguishable from the set of model ensemble members. From what I've seen the non assimilating models fail even that test. I'd not use the models for prognostication, but I maintain they're of use in diagnosis.

Given that I view area/extent as epiphenomenal of the real driver: volume. With open water formation efficiency (and weather) translating volume loss into area/extent loss*. I am bound to be biased, but it seems to area/extent orientation of virtually every modelling study I've read is missing the major factor.

* Of course area/extent play a role in ice/ocean albedo feedback and autumn ice growth, but volume losses/gains are the spine upon which the whole shebang hangs.

[Chris Reynolds]

Correction - wasn't Connelly, was James Annan.

[Chris Reynolds]

...and another post goes missing...

[Susan Anderson]

Amazing discussion, thanks all. Particularly liked the quantum-craps bit.

I don't think it's fair to blame the whole organization for Hoerling's work, though sadly he appears to have a platform there.

Gorgeous towers of cards too.

Now I'm going to stop wasting y'all's time, but thanks again.

[crandles]

Definitely wasn't 'Connelly' ;) . Yes, Annan and Hargreaves were saying aim for/test data being statistically indistinguishable from ensemble members. Dr Connolley was saying there is a big variation in quality between the models so you need to discard the rubbish ones based on some objective measure of whether the models are any good.

[wayne]

When do you think the melting started? Say 75 degrees North (not East of Greenland)? with temperatures below -11 C. Make a guess and read my latest work a culmination of 4 years of observations.

[wayne]

yes on my blog: http://eh2r.blogspot.ca/

[HeisenIceBerg]

I created a thread in Arctic Background board of the ASIF to take requests for access to publications; it has been stickied, so will remain at the top.

Chris Reynold's,
The paper I got for you has been moved to a slightly different location, so the link I gave before no longer works. You can find a new link here.

For those who want to read the full article (Dynamics of the Mackenzie River plume on the inner Beaufort shelf during an open water period in summer) which R. Gates posted a link to, but who do not have access, you can find a link to it here.

Steve Bloom,
I'll have yours in a few minutes.

Regards,
HeisenIceBerg

[HeisenIceBerg]

Steve Bloom,

I got the tree papers you requested. You can find links to the copies here.

Regards,
HeisenIceBerg

[HeisenIceBerg]

Steve Bloom,

It appears my post with links was eaten. You can find links over on the ASIF in the Arctic Background board.

Regards,
HeisenIceBerg

[Bosbas]

Lodger,
Thanks for explaining the Daisyworld. I was completely lost when I stumbled on that term.

[Llosmith57]

Wayne, Is there evidence for what is driving NW passage ice volume loss? Pacific water? Atlantic water? Dust from Asian power plants and desertification? Has anyone collected any ice samples to determine source origin? If the Antarctic Bottom Water formation continues in slow mode then the Arctic ice is going to have a very difficult time making a reappearance once it disappears. That would definitely be a game changer!

[Paul Beckwith]

@A+-Team
I greatly appreciate your advice and will consider it carefully when I have more time after I get through the next hoop for my PhD (oral comprehensive exams) in a few weeks.

How do you propose people here "get organized" to counter the observationally-challenged reliance on broken models?

I thought having some community projects to publish papers on observations would be a great way to get organized and have some goals and direction. It is a shame to not have such great work here get more exposure to the world.

The sea ice is telling us when she will vanish, we just need to ask the right questions and look in the correct places.

For example, it is clear that MYI has been leaking out of the Fram Strait year-round, even in the dead of winter. How does this compare to previous years?

Can someone estimate the total surface area of the sea-ice including cracks? This is clearly very important to melt rates.

Has someone generated albedo time series for sea ice during the melt season over the last few years? Has anyone here analyzed the sea-ice behavior during the early August/2012 cyclone? There is a peer-reviewed paper that claimed the 0.8 million km2 sea-ice lost over that 8 days or so would have happened anyway without the storm? Doesn't make any sense to me, those storms are death to the ice.

In Apollo 13 the 3 astronauts made it safely back to earth in a busted lunar orbiter only due to the actions of a team of great propeller-heads under strong leadership. I see sea-ice as the busted orbiter and people here as the team, but a lot more is at stake here then 3 lives. For example, numbers are not all in yet, but the morgues in the UK were counting huge numbers of bodies in excess of normal death rates over the past few months...

[Jim Hunt]

@Paul - "Some community projects.... would be a great way to get organized and have some goals and direction"

How about The Distributed Arctic Sea Ice Model?

[Shared Humanity]

Jim----"I thought having some community projects to publish papers on observations would be a great way to get organized and have some goals and direction. It is a shame to not have such great work here get more exposure to the world."

The idea of community projects is a great one. I think papers, however, are more an output from doctoral candidates like yourself.

I believe educating the public and advocacy are projects more appropriate to this community. For example, this community is taking current research on arctic impacts on climate and weather and arriving at accurate (I believe) conclusions around specific "weird weather" impacts. Making the public aware of these links will build the understanding and political will to drive policy. Scientists are reticent to play this role. We should not be.

If this community were effective in explaining the specific links behind the arctic and the drought in the breadbasket of the U.S., it would definitely drive activism in those states dependent on the agricultural economy.

If more effort was made to educate the citizens of Great Britain about the effect of ice loss on the unbelievable wet weather you are getting, perhaps we can take effective action before Wales washes into the Celtic Sea.