Editorial board member pissed off over a paper on “the pause”
Story submitted by WUWT reader Duane Oldsen
WUWT readers may remember Dr. Syun Akasofu as the source of a graph tracking the Pacific Multidecadal Oscillation with sine wave shifts in global temperature up and down. 
Dr. Akasofu’s recent submission to the first issue of the new journal “Climate,” a submission in this same vein of analysis, provoked one of the journal’s editorial board to resign in protest.
Dr. Asasofu’s submission was entitled “On the present halting of global warming,” and Dr. Chris Brierley of University College London declared the work to be of such insufficient quality for publication that his resignation in protest was requisite.
Dr. Brierley cites computer models and insufficient evidence in the paper as his reason for rejecting Dr. Akasofu’s submission to ‘Climate’ and thus provoking his resignation from the journal’s editorial board, despite crediting Dr. Akasofu’s hypothesis as valid and reputation as “deserved.”
Dr. Brierley specifically cites a lack of testing of Dr. Akasofu’s assertions in the submitted paper, which Dr. Brierley presents as an extreme abuse of the scientific method.
Dr. Brierley lists extensive critiques of the quality (i.e. lack thereof) of Dr. Akasofu’s work in the submitted paper. If accurate, this would be an effective indictment of Dr. Akasofu’s previous work as well. So both Dr. Akasofu’s source article and Dr. Brierley’s critique deserve attention.
===============================================================
Here is the paper:
On the Present Halting of Global Warming
Syun-Ichi Akasofu
International Arctic Research Center, University of Alaska Fairbanks, USA
Received: 28 January 2013; in revised form: 15 April 2013 / Accepted: 15 April 2013 / Published: 3 May 2013
Download PDF Full-Text [810 KB, uploaded 3 May 2013 14:45 CEST]
Abstract:
The rise in global average temperature over the last century has halted since roughly the year 2000, despite the fact that the release of CO2 into the atmosphere is still increasing. It is suggested here that this interruption has been caused by the suspension of the near linear (+ 0.5 °C/100 years or 0.05 °C/10 years) temperature increase over the last two centuries, due to recovery from the Little Ice Age, by a superposed multi-decadal oscillation of a 0.2 °C amplitude and a 50~60 year period, which reached its positive peak in about the year 2000—a halting similar to those that occurred around 1880 and 1940. Because both the near linear change and the multi-decadal oscillation are likely to be natural changes (the recovery from the Little Ice Age (LIA) and an oscillation related to the Pacific Decadal Oscillation (PDO), respectively), they must be carefully subtracted from temperature data before estimating the effects of CO2
To answer my own question as to whether or not Dr. Brierley (finally have it spelled right) has a “drank-the-coolaid” bias or “jerkness” towards the null hypothesis related to climate change, I have been reading some of his work, which is predominantly about oceanic systems. This is the first one I read:
http://earth.geology.yale.edu/~avf5/publications_pdf/Manucharyan.Interm.JGR.2011.pdf
The above linked study reminds me of the work we see posted here. The SST wake of a hurricane is quite cool. This paper discusses heat transport due to tropical cyclone activity. The first citation in the introduction is to observed changes (imagine that, he thinks observations are a key part of an introduction) and cites a paper I also just read:
http://tao-tc.ucsd.edu/WEB_DATA/PUBLICATIONS/DAsaro_cold.wake.frances_GRL_2007aug.pdf
Brierley’s research looks into two models of this observed behavior of tropical cyclones, a complicated one, and a simple one. I learned something. The observed and modeled SST wake is cooler. We’ve had a post about that wake here at WUWT. But does a hurricane suck the warm out of the ocean and send it through the eye up to the sky? Nope. It sends it downward and mixes it with colder layers beneath the surface.
I find the paper well done, meets peer review standards I have linked to previously, and does not mention AGW. Not even tacked on at the end. I will continue the search.
I just found his Ph.D. thesis. I also discovered that this thesis apparently was the beginning of more articles with co-authors on the subject later on that I can only find abstracts for.
First his thesis:
http://www.met.rdg.ac.uk/phdtheses/Ocean%20model%20uncertainty%20and%20time-dependent%20climate%20projections.pdf
This is a related publication later which is behind a paywall. But the abstract and the next page of the article is available to read at Springer:
http://link.springer.com/article/10.1007%2Fs00382-008-0486-3
So what do we collectively think? (oh I CAN be cheeky sometimes)
From what I can tell so far, Brierley’s work on anthropogenic climate change (which is only one part of his general concentration on oceanic systems.) seems to focus on why models are so crappy [high degree of uncertainty, low degree of “probabilisticness”]. He narrowly looks at ocean parameters [IE oceanic GIGO – dials, data, and mathematical formulas] as a source of that crappiness, and narrows even that down some more.
Two things I took particular note on:
1) He stated that for the models as a whole, atmospheric uncertainty encompasses the value for oceanic uncertainty, leading him to wonder if there is a cause/effect relationship with atmospheric model components playing a stronger driving role or at least a precursor but larger role on oceanic model components [makes me wonder about the real world cause/effect relationship too]. 2) He also noted that pre-industrial model uncertainty spread is greater than observations, which begs the question, if you can’t get the null hypothesis model to work, how do you know the anthropogenic version is any good?
In summary, his thesis appears to be a substantial amount of work and at a very high level of intellectual rigor and questioning ability. He seems to be an unusually curious tinkerer of models and exhaustively follows his nose about as well as any dog I have had the pleasure of hunting with.
Pamela Gray says:
September 11, 2013 at 4:05 pm
Thanks, Pamela. Sorry for my lack of clarity. It was an example. I was referring to your claim that the only thing worth discussing is the written paper and the written response, and that everything else is off-limits. Or in your terms, “side issues” and “emotional issues”
If we had applied your criterion to other papers, then we would have very little to discuss … but sadly, in climate science, what happens outside the ambit of just the papers themselves is often as important as the papers themselves … as is clearly revealed in the Climategate emails.
Take a read of Bishop Hill’s story of Caspar and the Jesus Paper, for example … and then try re-reading the paper while striking out references to everything but the paper itself.
You may get a sense of what I mean—often, the papers themselves, far from being the only topic worthy of discussion as you repeatedly have claimed, are the least important part of the story.
Now, if you don’t want to discuss those important issues, that’s fine.
But what you are doing, trying over and over to convince people that we’re all just being “emotional” and pursuing “side issues”?
Sorry, but that won’t fly.
w.
u.k.(us) says:
September 11, 2013 at 7:17 pm
Well, since (as you say) you don’t have a clue what the topic of the discussion might be, you might try re-reading Pamela’s and my comments including the one just above. Check Richard’s comments above as well. All of that might help.
The part I don’t understand?
Why someone who doesn’t know what the discussion is about seems to have such a strong view about it …
w.
Pamela Gray says:
September 11, 2013 at 5:47 pm
Good God, is that what passes for an argument in your world? Make up any fantastic thing that crosses your mind, and pretend that is what your opponents are saying? No one has said anything remotely resembling that rant.
Pamela, I know you are capable of high quality thought and argument, your succeeding comments alone are enough to prove it … but that’s grade-school nonsense. When I say that there are things beyond the paper itself that are worth considering, you pretend that I’ve said that anything and everything is worth consideration … ummm …. Pamela … I never said anything remotely resembling that, and you’re just making a fool of yourself with that pretense.
w.
Pamela, thanks for the link to Brierly’s PhD thesis. Folks like Gavin Schmidt have claimed that there are few “tuned” parameters in the climate models, and thus the models are “physics based”.
In the thesis, Brierly provided a list of just some of the parameters in the ocean, viz:
I count no less than nineteen parameters … elephants‘ trunks, anyone?
w.
Willis, Solar calculations are equally complicated. Climate models, ocean models, physics models, they are all complicated and many parameters. And it took a while to nail them down to any degree. Brierley appears to have accurately stated that in their current state, climate models have components that are wrong. And it appears to me that Brierley is interested in tinkering with the models because they are complicated and therefore ripe ground to find out why they don’t work very well.
I find that better than those who think their models are right and the observations are wrong. Unless we are talking about sensor data becuz those observations are right except for ocean data becuz those observations are wrong. And maybe the sensor data is wrong recently because we have missing sensors and missing people who look at them. But we can fix that by creating a model to make them all re-appear and be counted.
So far, at least when judged by the academic rigor of his published work, I don’t see the kind of researcher we love to pan here at WUWT. This guy actually reminds me a bit like Leif in that regard. Only his gig are the oceans while Leif’s is Solar-focused.
As to Brierley’s public review, I wish we had public reviews more often! I wish we had an open-accessed journal just for the purpose of publishing reviews!
Willis, one more point of clarity. There were many additional areas of concern listed in this thread by several commentators, many of them fueled with emotional feelings. But few having to do with the topic clearly stated by the reporter. I was not singling you out. However, I do think the gut response in several commentators was to pan an ocean climate researcher not because he openly reviewed a piece of work but because that piece of work claimed solar drivers as well as intrinsic drivers, while disagreeing with anthropogenic CO2 drivers. The gut reaction of us versus them came into play. So instead of asking yourself if the review was spot on, the immediate response was to denigrate the man because of that gut-reaction. Reasoning and thinking was far behind.
Folks, if we want to be taken seriously by “them”, I suggest we reason and think. That includes putting a filter on the from-the-gut reactions and name calling. This was an excellent opportunity to reason and think. Did we miss the mark? Yep.
Pamela Gray says:
September 14, 2013 at 9:21 am
Thanks, Pamela, I’m sorry, but this is totally unclear. Which solar complications are a complicated as what? Are you saying that solar calculations have 19 tunable parameters in just a portion of the model of solar dynamics? If so … please enumerate them.
The problem is not complexity … it is tunable parameters.
w.
Pamela Gray says:
September 14, 2013 at 9:37 am
I don’t understand your unending insistence that anything not “clearly stated by the reporter” is some how off-topic. On my own threads, often some of the most perceptive comments are regarding things I haven’t mentioned … and that is the beauty of the blog. People bring up aspects of things that I never even considered, and often they are important issues.
For example, I did a little research and found out that Brierly doesn’t even understand what the “Hockeystick” is … how is that not relevant to the discussion? It gives us great insight into his lack of knowledge of the subject matter.
The same is true of where he chose to publish his objections. If you truly object to bad science, you don’t post that objection on a website that is infamous for supporting bad science. And if you do that, it is obvious that your objection is not to bad science, otherwise you’d publish elsewhere … and again, how is that not relevant to the question of the validity of his objections?
If you look at my very first post on the subject, I said that I find the Akasufo piece to be greatly lacking. I also find Brierly’s claims to be equally unconvincing. As one problem among others, he has put up an impossibly high standard, one that is not used in the field … why do you find that off-topic?
Now, is any of this “fueled with emotional feelings”? Perhaps for others, but for me, heck, no. I never even heard of Brierly before this, and near as I can tell, deservedly so—he thinks the Hockeystick is the last hundred years of data, why should I have ever heard of such a clueless noob?
But concerning your continued attempts to convince people that only and uniquely the things “clearly stated by the reporter” are worth discussing?
Well, given your unending insistence on that claim without a single logical reason to back it up, THAT certainly might be “fuelled by emotional feelings” …
Does Brierly do good work? Yes and no, I’d say. He wrote his PhD thesis about a model with 19 tunable parameters. And while he did discuss the effects of changing those parameters, and did a good workmanlike job of it as you note … he didn’t say one word about the underlying validity of a model with 19 tunable parameters. Surely at least a mention of Dyson’s warning about tunable parameters would have been in order, no? And that is a huge hole, in my opinion, one that reveals a lot about his mindset.
Best regards, and as always, your ideas and contributions are most welcome.
w.
I have never said you could not discuss whatever you wanted. But will I join in a discussion of some of the issues brought up in comments? Not unless it interests me. So please do what you wish to do. I will do what I wish to do. If you want to discuss the paper and his critique with me, I am interested. If you want to discuss the resignation with me, I am interested. If you want to discuss his choice to publicly announce his resignation and post his critique, I am interested. I am not interested in discussing his choice of using the Skeptical Science blog. I really don’t care where he posted it. The fact that he posted it is interesting. However, the context of his reasons helps me to understand his decision to bring this to the attention of the greater audience.
Two points to consider: 1) WUWT has posted research that was pretty bad. Does that bother you? And 2) yes models of things like solar-related dynamics and models of natural global climate circulation without anthropogenic parameters have many dials. Does that bother you? Models of everyday mundane things have lots of tuneable dials. Is that bad? If you build a computer model of oceanic parameters, which would require lot more tunable dials, how does that make it a bad model? I guess you could build many hundreds of them with glued down parameters but I am betting it would be pretty costly and a poor use of tax dollars. Tunable dials make more sense to me. Is that you don’t think any of this should be translated into a model? How come?
The issue with making the trunk wriggle has to do with the difference between a “trained” climate model output, which at that moment in time matches observations, (the real elephant is wriggling his trunk and you made the model do the same thing by adjusting the dials) and what happens going forward. Right now, most climate models are still making the trunk wriggle. But the real elephant is taking a nap and not wriggling his trunk. So something in the model isn’t right. There is nothing bad or good about modeling trunk wriggling in and of itself. The bad part is when the modeler doubts the observations instead of doubting the model. Brierley doubts the model, as far as I can tell.
Pamela Gray says:
September 14, 2013 at 1:57 pm
As I have said several times, that’s fine, do what you want. My objection was and is to your claim that anything but what you want to discuss is a “side issue” or an “emotional issue” … which kind of defeats your claim that I should “do what I want to do”. If you don’t wish to discuss it, fine—I’m objecting to your attempts to poison the well regarding something you repeatedly claim you’re not interested in. Not interested is fine. Claiming that what you’re not interested in is a “side issue” or an “emotional issue” is not fine.
Not in the slightest. The nature of WUWT is that each post is judged on its own merits, and part of the purpose of WUWT is to allow bad science to be publicly exposed and falsified.
Only when it is claimed that the models ability to hindcast reality means that they are correct … which is all the time.
Sorry, too vague to reply to. Which models of which mundane things have tunable parameters?
It depends on what you want to use the model for. If you want to use it to convince me that we should restructure the worlds economy based on the model, yes, it’s a very bad model for that. And as above … that happens all the time.
Again, the question is really too vague to answer. It depends on the use to which the model will be put, and the claims about the model. If you claim that your model with 19 parameters is correct because it fits reality, that’s a logical mistake.
Here’s an example. The climate models are tuned to hindcast the past global surface temperature, provided they are given a certain specified set of inputs.
However, then the claim is made that because when you pull out the human-related inputs the model performs very poorly, this shows that the human-related inputs are a real and necessary part of the total climate equation …
I’m sure you can see the circular logic in that claim, and yet despite that, the IPCC makes that very claim and repeats it with each new Assessment Report.
Finally, I’ve shown that the temperature output of the climate models can be emulated to a ~98% accuracy by a simple, one-line equation … in other words, their output is a trivially simple lagged and resized version of their input. As a result, their claims about climate sensitivity are meaningless.
Is this a good use of tax dollars, a million-dollar, million-line, million-man-hour model that just lags the input and declares that as the output? Is this a valuable model? Can we conclude anything at all from such a model?
Best regards,
w.
Willis Eschenbach says:
September 13, 2013 at 8:06 am
“Why someone who doesn’t know what the discussion is about seems to have such a strong view about it … ”
=======
It was just the tone of the discussion, between two familiar commenters.
Seemed a waste of words, entertaining as it was.
The disrespect brought on the “strong view”.
Your turn…..
Pamela Gray says:
September 14, 2013 at 2:17 pm
Thanks as always, Pamela. Brierly seems totally unaware that the temperature output of the GCMs is a simple lagged version of the inputs. I see no evidence that he doubts the model for the right reasons or in the right amount … which is lots.
As to whether “There is nothing bad or good about modeling trunk wriggling in and of itself, as I explain just above, that depends critically on what you want to do with the model. For some things it doesn’t matter at all, and for others it is a death sentence.
w.
u.k.(us) says:
September 14, 2013 at 4:45 pm
Thanks, u.k. So if I understand you, you don’t know what the discussion was about, but you don’t like the tone. And somehow the tone makes it a “waste of words”?
I’m still not clear how that works. If you don’t know what’s at stake, how would you know what tone is appropriate? And does a wrong tone automatically mean that nothing good will come of it?
I’m a passionate guy, u.k. And sometimes, that comes out more strongly than I might prefer in a perfect world … but why on earth would the fact that I’m passionate, or even over-passionate, make a discussion into a “waste of words”? It certainly doesn’t seem to have done that here, the discussion has continued, with clarifications and greater understandings on both sides. And it doesn’t seem to have done that in general, the threads to my posts are very productive, I learn heaps from them regardless of my tone.
From my perspective, the problem is not that I care too much or am too passionate about these issues of the trashing of science itself by the CAGW fraternity—it’s that other people care too little and are not passionate enough about them. When Brierly takes what is a valid complaint about bad science and posts it on an anti-scientific website, we all lose, but the reputation of science itself loses the most … and yes, that’s important to me.
Look, I could pretend that this stuff never bothers me, and that I’m always on a perfectly even keel. I’m a pretty good wordsmith, I could write nothing but sweetness and light regardless of my passions … but I’m not capable of that, to me that’s lying. If I’m happy I write happy, and if I’m angry, I write that way. And it has a huge advantage, in that I’m never fooling anyone, and everyone is quite clear where I stand on any given matter.
Could I do it better and still tell the truth? I sure hope so, and as a result it’s an ongoing project with me. But at no stage in that process will I pretend to be someone I’m not.
All the best,
w.
Willis Eschenbach says:
September 14, 2013 at 5:05 pm
“If you don’t know what’s at stake,….”
================
My “Wasted words” comment was stupid/badly worded.
Who knows what’s at stake anymore, all we hear are lies.
Our best effort at combating the willy-nilly run towards CO2 regulations is to provide a well-reasoned voice regarding modeling the climate. Our best effort at combating the will-nilly run towards “it’s the Sun stupid” is to provide a well-reasoned voice regarding modeling the climate.
To that end, I like the idea of trying to model the climate. It intrigues me. I’m curious about it. I like the challenge of, “Can it be done?” It’s like going to the moon and back with collected rocks in my pocket no less (I sure wish I had been on that ride). Can it be done? Don’t tell me no. It just makes me work harder.
I am also intrigued by folks who are trying to get the climate models to work. Between the two combatants here, Brierley and Akasofus, I find the work of Brierley worth something, Akasofus not so much. And much is at stake here. We don’t need the half-hearted attempt by Akasofus. His effort will not focus serious attention on the models. Anybody with half a brain will dismiss his hastily put together article. We need the rigorous curiosity of someone who says there is something wrong with the models and we need to focus on them. Brierley seems qualified to tell the modelers, “Look here. There is something wrong with this part”.
I also think Brierley is saying that the effort is worth far more than what Akasofus put out there and his public pronouncement gave that thought visibility. I wonder if he was also saying that about the journal itself. I would agree with him on both counts.
Another question that equally intrigues me is why he did not come here. Maybe that question has just as much merit as why he went to Skeptical Science. Yes?
I would put a challenge out to posters, readers, and commentators. Find rigorous work that challenges the models and offers ways to improve them. I think that is where the money is at.
Pamela Gray says:
September 14, 2013 at 6:09 pm
Been there, done that. I’ve shown, quite rigorously, and for both individual models and groups of models, that regarding the future evolution of the global temperature the global climate models can be replaced by a simple one-line equation, viz:
T2 = T1 + lambda (F2-F1) (1-a) + a (T1-T0)
This does far more than simply challenge the models. It establishes, again quite rigorously, that for the purpose that they are being used for, the entire current generation of climate models are worthless junk. They are useless Tinkertoy™ models, mindlessly and endlessly churning out a trivial lagged, resized version of the inputs.
So I’m not sure what else you are looking for in the way of a challenge to the models. Note that they have done this by adjusting parameters, rather than by replicating the basic physics.
As to ways to improve them, the problem is that they use parameters rather than using physics. They do not contain any emergent climate phenomena. No thunderstorms. No cyclones. No PDO. No El Nino/La Nina. As a result, they don’t have a thermostat.
As to your comment that “that’s where the money is” … I cracked up when I read that.
Pamela, the money is not in accurate climate models. The money is in alarmism. The old saying that “Bad news sells newspapers” is true in spades for climate science.
Anything else you’d like to know? I’m on a roll this lovely Edinburgh morning, where it’s pouring down rain.
w.
Willis, is your model capable of a projections (and how far out) and is it accurate at hindcasting (and how far back)? Or is it solely dependent on observations and can only project a few months ahead? Explain how your model would serve humanity to prepare for severe decadal conditions or instead make investments in agriculture during a growing conducive decade. Would it have been able to project the probability of the dust bowl? What a devastating time in history. Would it have been able to project conditions that spawn such things as the killer nor’easter that snuffed out the lives of scores of military men? And explain how the scale of your model is better. Does it allow us to see that some regional climates are less sensitive to oscillations while others are more sensitive? Inquiring minds would want to know. Not to downplay your effort but I do have my doubts, as I well should.
I do not like the current set of climate models. I think they put the cart before the horse. They put into place a warmer ocean due to CO2. Yet SWIR is the real muscle, and clouds are the real source of how much SWIR gets to the surface. The good news is that these intrinsic equatorial conditions can be modeled on a decadal scale using a suite of ENSO models, instead of a suite of CO2 models.
I would sit up and take notice of a suite of several versions of models based on long term oceanic-atmospheric oscillation scenarios. This is where Brierley’s work would be used, clearly. He and others should be at the forefront of modeling. With oscillation scenarios as the main driver with models that use a combination of statistical and dynamical calculations, one can also then examine cloud and storm conditions. If anthropogenic CO2 is still an issue, plug that in and see what amount it takes to changes anything and under what conditions. Maybe some oscillations are more susceptible to anthropogenic warming and others are not. Maybe some oscillations produce additional CO2 while others do not.
The beauty of a suite of intrinsic decadal ENSO models that together provides the entire range of more SWIR, less SWIR, and more neutral SWIR with the component cloud and storm conditions isn’t a competition to see which one is best. They allow us to see which decadal pattern we are likely in and to act accordingly. They would also allow us to see which ones are sensitive to pollution (and which kind) and which ones are not, if any of them indeed take notice of pollution.
Does your model allow us to project all that? Or is it just something that tells us what we already know. Its raining today, take an umbrella.
Finally, in what way is Brierley’s work alarmist?
Pamela Gray says:
September 15, 2013 at 7:31 am
Thanks, Pamela. My model is a very simple model, which emulates the global temperature results of the climate models with extremely high fidelity. Regarding the global temperature, it can do everything that the climate models can do … which, as my model shows, is nothing.
But then, what would you expect from a model (either mine or a GCM) which merely outputs a lagged, resized version of the inputs?
So no, it won’t do any of the things you mention, not one … but its value is that it also demonstrates that the GCMs can’t do any of the things you mention either.
It seems to me that you’ve missed the point of my model. It does not have any independent predictive values. It just shows the limitations of the GCMs, which turn out to be very simplistic. They just take the inputs, scale them, and lag them … bozo, huh? Millions of lines of code and the output can be totally emulated by a one-line equation. Go figure.
w.
Pamela Gray says:
September 15, 2013 at 7:35 am
Where did I say that Brierley’s work was alarmist? I’ve searched the comments and can’t find it. I did say “the money is in alarmism”, but that was in response to a comment of yours, not regarding Brierley.
w.