Do some of you remember Steig et al 2009, a pre-Climategate Nature cover story? Like so many Team efforts, it applied a little-known statistical method, the properties of which were poorly known, to supposedly derive an important empirical result. In the case of Steig et al 2009, the key empirical claim was that strong Antarctic warming was not localized to the Antarctic Peninsula (a prominent antecedent position), but was also very pronounced in West Antarctic. Their claims are set out firmly in the opening sentences of their abstract as follows:
Assessments of Antarctic temperature change have emphasized the contrast between strong warming of the Antarctic Peninsula and slight cooling of the Antarctic continental interior in recent decades. This pattern of temperature change has been attributed to the increased strength of the circumpolar westerlies, largely in response to changes in stratospheric ozone. This picture, however, is substantially incomplete owing to the sparseness and short duration of the observations. Here we show that significant warming extends well beyond the Antarctic Peninsula to cover most of West Antarctica, an area of warming much larger than previously reported. West Antarctic warming exceeds 0.1 deg C per decade over the past 50 years, and is strongest in winter and spring. Although this is partly offset by autumn cooling in East Antarctica, the continent-wide average near-surface temperature trend is positive.
Their claims were illustrated on the cover of Nature as shown below:
Again, to be very clear about this, the “novelty” of Steig et al 2009 were their results for West Antarctica – the location of the West Antarctic Ice Sheet. Otherwise, there was nothing in their article that warranted an article in Nature, let alone a cover.
There were a variety of problems with their analysis, many of which were documented at CA and tAV at the time. The remarkable episode of Gavin, International Man of Mystery, plagiarizing a small point on a CA thread arose in the context of Steig et al. At an early stage, it seemed that Steig’s rather complicated method was spreading results from the Antarctic Peninsula into other parts of the Antarctic, a surmise that proved correct.
After an abusive peer review process in which the Team were evidently involved, an article has been accepted by Journal of Climate (O’Donnell [Ryan O], Lewis [Nic L], McIntyre and Condon [Jeff Id]) refuting the West Antarctic claims of Steig et al 2009.
Their results proved to be sensitive to the number of retained principal components, with Steig et al selecting only 3 PCs, using arguments that contradicted Wahl and Ammann’s efforts to retain bristlecones in MBH. Steig et al themselves did not appear to realize that Chladni patterns emerge from spatially autocorrelated time series, a peril known for many years (“Castles in the Air”) but apparently forgotten by Steig et al and the Nature reviewers.
Using a more rational (though not magical) policy on retained PCs, a very different picture emerged as illustrated below. The “traditional” picture of very strong warming on the Antarctic peninsula re-emerged, together with large areas of cooling on the continent. Instead of the West Antarctic being a location of anomalous warming, as Steig et al had purported to show, parts of it were actually cooling.
The gauntlet that had to be run shows that practices in climate science journals remain unchanged despite Climategate. Horton’s essay for Muir Russell noted that conflicts of interest were not simply financial. In this case, the Journal of Climate appointed a reviewer – or shall we say a representative of a Team of reviewers – whose energy in attempting to suppress the article went far beyond an unconflicted reviewer. Ultimately, the reviews and responses totalled 88 pages! And Andy Revkin and others blame critical authors for not running such gauntlets. The reviewers for Ross’ and my comment at IJC on Santer et al 2008 were even worse. All too often, in the case of Santer et al, after Team reviewers sabotaged a straightforward and correct comment, Santer and other Team members criticized us for not publishing in the PRL after getting data and continued to put their results forward as unblemished despite knowing of unrebutted criticisms. Getting this article accepted is entirely to the credit of Ryan O’Donnell who did more than the lion’s share of the work.
Climategate documents show the asymmetry between the puffball “pal reviews” that Jones submitted for Mann or Schmidt or Wahl and Ammann, as compared to the Team “going to town” on criticism. Unfortunately, the Muir Russell “inquiry” did not investigate the peer review incidents evidenced in the emails (not even the “going to town” incident) nor did it comment on pal review nor did it make the slightest effort to see if there were other peer review incidents in other CRU documents (part of their terms of reference.)
The gauntlet that was created in this particular incident had nothing to do with additional due diligence occasioned by perhaps overturning a well established result. Steig’s results, showing West Antarctic as a particular locus of warming, were themselves novel and, if anything, contradicted prior views of Peninsula warming. Our results were straightforward – the 88 pages of review and response were nothing more than obstruction, “going to town” on the comment rather than the original article.
Substantively, what is actually left of the signature results about the West Antarctic, which were:
Assessments of Antarctic temperature change have emphasized the contrast between strong warming of the Antarctic Peninsula and slight cooling of the Antarctic continental interior in recent decades… Here we show that significant warming extends well beyond the Antarctic Peninsula to cover most of West Antarctica, an area of warming much larger than previously reported. West Antarctic warming exceeds 0.1 deg C per decade over the past 50 years, and is strongest in winter and spring.
Nothing. Steig’s West Antarctic warming results from a spreading of warming in the Peninsula to the West Antarctic through choices made in their principal components. Different choices – ones more plausible in the circumstances – lead to opposite results.
Also see blog posts at Jeff Id here and WUWT here.
Update Dec 7: Online SI here http://www.climateaudit.info/data/odonnell/
Climate Audit 
337 Comments
Congratulations!
All of you authors are to be congratulated for the publication of your paper and for the courage and tenacity to challenge Nature’s pre-Climategate news story.
Steve, Steig himself seems to be a decent sort of a cove, who’s maybe fallen in with a bad lot. Did anyone contact him during the production of this paper to discuss the potential results with him. Or have any of the authors spoken to him since acceptance for publication? The reason I ask this is given his apparent difficulties and given the respect which most of the authors of this paper seem to have for him it would be nice to see he has been consulted and agrees with the findings of the paper, or at least has differences that he can articulate.
Steve: I expressed a similar view at CA at the beginning – using, as I recall, the same expression “fallen in with a bad lot.” I sent a pleasant email to him asking for data not in the original SI and responded pleasantly. He suddenly turned and made abusive comments about me both here and at realclimate, making wild and false claims that I’d accused him of fraud. Jeff Id posted on Steig at the time and used harsher language, but, for some reason, I seemed to be the particular target of Steig’s opprobrium.
There’s no question in my mind that Steig was part of the Team that constituted “Reviewer A” and that he participated actively in trying to suppress our article.
Keep pushing, Steve, Jeff and all – Shining light on each of the incomplete, erroneous, obfuscatory or misleading cornerstones supporting the alarmist positions has brought tremendous clarity to the entire issue, including the part having socio-political content. Congratulations!
Typically a reviewer must get permission from an editor to bring in outside experts to help with a review. If the original “Reviewer A” did not solicit and obtain editorial permission to include a circle of critics, then Reviewer A is likely in breach of professional ethics. If the manuscript editor was aware that Reviewer A had included others, without having made a formal request, then the editor was also ethically remiss.
This surmise need not remain in the domain of speculation. The 88 pages could be subject to computer textual analysis.. It is a significant corpus, perhaps sufficient to indicate the number of (sub-)reviewers involved. The analysis could be further extended to determine the likely identity of the (sub-)reviewers.
Not to be coy, but we do not require such sophisticated analysis to determine these things. 😉
Re: Ryan O (Dec 3 13:20), as in people who forget about meta data in files, for example.
Or as in people who use the exact same wording in reviews as they do in emails.
Re: Ryan O (Dec 5 10:08),
or, maybe the real reviewer plagarized the email.
I think eric doesnt understand the tenacity of the engineering mind set.
We have a saying in marketing. time to shoot the engineers.
hehe. bet you havent heard that
Kind of like Rainman with OCD.
By the way, as the review .pdfs are all rendered by the AMS software, the metadata only contains information from the rendering. Metadata from the source file is not copied or included. 😦 Perhaps AMS thought of that one! 😀
Re: Ryan O (Dec 5 10:23),
ha, once I had a sales guy mail me a offer in newer version of word. Upon opening it in a different version I was able to recover all his edits. hehe. Too bad they succeeded in bleaching the metadata.
Once again, with no evidence you make a claim about someone. This is a habit of yours.
Except for the reviews themselves as well as Steig’s original arguments against the work they were doing… those certainly amount to evidence. Is the inability to see this, the obvious, a habit of yours?
Mark
So you don’t actually have evidence it was Steig, other than guesswork, as I said. But don’t let that stop the chance for the demonization of a scientist pass by.
Re: stereo (Dec 3 06:49),
The important point is, given McSteve’s previous engagement with Steig previously and the team et-al passim, given that he’s actually seen the 88 pages and responded to them, given his undoubted expertise at parsing sometimes complex and obscure text and eloquently highlighting differences and / or similarities, is it plausible that he has reasonable evidence that “Reviewer A” was a group effort, and that Steig was part of the group.
The answer, of course, is yes. Now he certainly hasn’t presented anything specific, but that’s a different matter. The fact is that he’s in a position to see data that would allow him to make that call, and certainly has form for making these calls accurately in the past.
That “mark t” or anyone else on the thread *doesn’t* claim either the data or expertise to interpret that data wrt “Reviewer A” is of no consequence to McSteve’s statement and your subsequent pointless and content-free niggle in response to it.
Why don’t you head off and ask Steig directly if he was part of the “Reviewer A” collective? See if you can get a denial.
If he *wasn’t* involved, it should be plainly stated and rapidly forthcoming, yes?.
Then you can trot back here and we can put it on record, and I’m sure McSteve will modify his comment if it’s warranted.
Guilty until proven innocent. Yes, I have seen that type of logic before.
Guilty of what, stereo?
Do you consider the involvement of Steig as a reviewer to be a crime of some sort? Or do you consider that the peer-review process as applied to this paper to be unreasonable and / or unethical in some way?
Do tell.
I didn’t say the reviewer did anything wrong. I’m just working with the guess that Steig was even a reviewer. Nice try at changing the subject.
For what it’s worth, I did not originate the belief that Steig and/or his associates were Reviewer A – Ryan held the view first. Immediately on reading the first review.
Exactly. You are guessing.
Nuff said.
You’re the one who employed the term “guilty”, not me.
I repeat, McSteve is in a position to make a judgement on the likelihood of Steigs involvement.
You, however, are not, and yet you raise an objection based on total ignorance.
You can address that ignorance by trotting off and asking a simple question, but you choose to dodge around, introducing your own notions of guilt and innocence.
If there is no wrongdoing, there can be no guilt. The inference I draw is that you are not yourself convinced of the fairness of the process, but will not admit it.
So, trot off and ask Steig.
As in, within the first few paragraphs, it was pretty clear.
From Ryan O’s comment: “Or as in people who use the exact same wording in reviews as they do in emails.”
Kinda hard to do if you aren’t the same person.
Mark
Cool thing is that he’s mostly right in his guesses. For example, check out ‘Gavin, International Man of Mystery.’
Guesses. Exactly.
Re: stereo (Dec 3 18:09),
If you hold that position, go and ask Steig. Then come back and tell us.
stereo: Just because he hasn’t presented the evidence, doesn’t mean he doesn’t have the evidence.
Perhaps that makes you the person making a claim about someone “with no evidence.”
Re: stereo
“But don’t let that stop the chance for the demonization of a scientist pass by.”
Considering Steve’s past history with him(below) this isn’t a random shot in the dark you portray it to be.
“He suddenly turned and made abusive comments about me both here and at realclimate, making wild and false claims that I’d accused him of fraud.
Think about the character of someone that would do this, then post on Jeff’s site about why Steve cannot be believed, all while telling Jeff he’s looking forward to a preprint of the article. Really, the gall something like that would take is amazing.
Mark
I’ve had several reasonable email conversations with Dr. Steig since our initial blog discussions. The whole thing started because when I asked for the data and code he made a smart comment about me taking his matlab class. So naturally, having my Irish temper, I let rip on tAV for a while. It was rather entertaining but in the end it worked into a nice learning experience. Ryan really, and I mean really!, deeply dug into the math of these methods. It was quite entertaining, like a crossword puzzle and was in the end probably my favorite blog experience.
Like Steve wrote above, I don’t think our conclusions were particularly welcome, but we could tell very early on in the process that the total trend for the Antarctic should sit around 0.05 – 0.07C/Decade so there was something wrong. One of the points which seemed to get lost in the complexity of the paper (but is still key) was the fact that the peninsula was heavily oversampled, having a high density of temperature stations. This is one of the main reasons that it was so easy to become blended across the Antarctic continent. Higher PC’s and different methods did a much better job constraining (but not fully eliminating) this oversampling effect.
Anyway, over a year of work went into this correction so it’s fantastic that Ryan was able to keep at it in the face of a contentious review. It’s not like he got paid for this after all.
One gets an oversampling effect in MBH data with all the bristlecones. This also applies to some of the data sets used in McShane-Wyner vairations e.g. Gavin Schmidt’s comment.
Bristlecones are functionally similar to the Peninsula – without any evidence/proof that the strip bark phenomena are not merely mechanical artifacts.
You are free to disagree with people without making continuous and boorish comments on their personalities, competence and motives. The review process for publishing a scientific paper is tough. Who would have thought it? I just hear all the time about how easy it is to get published. They must be singling you out for special treatment.
“I just hear all the time…”. Hearsay? Gossip?
Tougher for some more than others it is fair to say…
stereo:
I agree with this. Of course, one could hardly say that RealClimate or OpenMind is innocent of this behavior too.
Do you apply this standard equally, or only to people you disagree with?
Where has Steve made boorish comments on personalities, competence, and motives? Please provide links.
You haven’t read any of the CRU emails have you. There is blatant pal review enumerated therein, as well as blatant obstructionism.
So Stereo, I do not see you questioning why Steig et al 2009 got published without a rigorous “peer” review, that did not take 10 months considering all the errors they made. Did you ask Steig why he said Steve was accusing him of fraud?
But Steig et al 2009 was a year between reception and publication….
Dear Steve,
I already asked this question over at Jeff’s blog, where it seemed to have been drowned under a lot of congratulations . .
But, does this publication and the math in it changes the way to look at the Mannian method? How difficult is it to take this math and use it for example on the Mann08 dataset?
Like Jeff said: Ryan seemed to have dived pertty deep into the math . .
All the best regards,
LoN
The use of RegEM in S09 is a bit different than the use in the paleo reconstructions. The S09 method holds much closer to the original purpose of RegEM – which is to infill missing data – than the paleo stuff. In S09, the real reconstruction is the infilling of the missing ground station data. Since the ground station information has a known scale (i.e., it is already temperature data), many of the considerations are different. While S09 also uses satellite data, the satellite data merely provides icing on the cake: yielding pretty, gridded pictures. Fundamentally, it is the ground station data that makes up the reconstruction.
The paleo stuff is different. In the paleo reconstructions, RegEM is used as an iterative regression tool where the scale of the predictors (proxies) and their relationship to temperature is not a priori known. RegEM was not originally designed for this purpose. Not having spent a great deal of time on the paleo stuff, I do not know how well-suited the algorithm is for this type of use.
Steve: this distinction was made in my original posts on RegEM. One of the benefits to the Steig commentary was that we managed to achieve a very precise benchmarking and reconciliation of the awkward Mann algorithm to our own more coherent algorithm in R – something that was reported on in early 2009.
This permits us to revisit M08 and analyse it – something that I’ve thought about from time to time. However, as I’ve said on many occasions, I do not regard the primary issue in this field as the absence of a new and better multivariate method, but the need to reconcile apparently inconsistent “proxies”. I will stipulate that bristlecones and/or Yamal plus random data yield hockey sticks using a variety of methods.
What? You guys were not richly funded by big oil (or coal) interests? Hard to believe. What other motive could you have except the desire for personal monetary gain… and to do evil in general?
I had forgotten about the Matlab class comment. Its hard not to smile about that now. Hopefully Steig has gotten an education into the perils of elitism in an interconnected world. Someone made an interesting comment recently about how no single person knows how to manufacture an iPhone (there are people who understand how to build tools to drill an oil well, and there are people who know how to turn petroleum into plastic, and there are people who know how to make silicon waffers and people who know how to create chips using photolithography, etc. But there is no one person anywhere on the planet who knows how to do all of these things). Climate science suffers from people who have very narrow expertise who try to claim expertise in things they are not experts in (and refuse to acknowledge the limitations of their expertise). Its a real object lesson for how climate science needs to become much more multi-disciplinary and open.
Milton Friedman – The pencil
My assessment has been that many climate scientists tend to be jack of all trades sorts, master of none. They have a smattering of a wide variety of skills and assume, unfortunately and incorrectly, they are masters of them all. Few seem to realize that many aspects of their field are actually studied in detail by others that specialize in certain areas. Statistics, control theory, physics, etc.
Mark
That is a ridiculous comment. Think about it.
I did, that’s why I posted it. Climate science covers a very wide cross-section of unrelated, or loosely related, scientific, engineering, and mathematical fields. It is impossible to master them all.
You should think about it.
Mark
It has always seemed to me that the study of “climate” is statistics and only statistics. Physics, among other things, is necessary to understand the weather but we are told that weather is not climate (except when it is). “Climate” only exists as statistics.
No, it doesn’t. That’s why they can model it.
Uhmm, yeh, but not accurately. And BMcB is essentially right. Climate is the aggregate of weather. And how do we quantify the aggregate?……With numbers and ratios, etc. or what many call………Statistics.
You can model a statistic. What I think you are trying to say that this statistic can be modelled from physical assumptions but you can’t escape the conclusion that those models can only be tested or validated by more statistics. On the other hand plenty of science, including thermodynamics and perhaps even quantum mechanics, is essentially statistical.
Stereo
The mere fact that it is possible to create something which you can describe as a climate models does not seem persuasive. It is also possible to create financial models and other models which have nothing to do with physics. In any case, I doubt that there are many climate models which are based exclusively on known physical principles. As I understand it, the art of climate model making (like other model making) is to posit or assume relationships between physical phenomena. Once you have a result, you compare the results of the model to the thing you are trying to model. In this case, both the model and the thing you are modeling are really statistics. It doesn’t make sense to talk about it in other terms.
Nevertheless, I probably shouldn’t have said that climate exists only as statistics. I admit that there is a thing called climate which would exist even if statistics regarding it did not. What I meant was the study of climate ought to be considered a sub-field of the study of statistics.
I have never seen any climate scientist claim to have ‘mastered’ their field or specialty, and to suggest few seem to realise that there are specialists out there is laughable. How does one pass through the education system and working life without exposure or resort to specialist and foundation texts? You should really think about. Your ‘assessment’ is just pointless.
This raises the question then of why they do not seek specialist help. In the particular case of statistics, three have been major senior climate scientists reported here are dismissing the need to consult them.
SMc has dissected published methods and found redundant use of linear algebra operations on the data. Operations were undoing what previous ones did on multiple occasions. The reported method seemed to be a series of linear algebra operations enabled by Matlab or a similar program that was put together with little understanding of what they were doing with the data. SMc comments on the reliance of climate scientists on home made methods. it is surprising that for such a globally vital topic as AGW that elite mathematicians are not enlisted to vet these methods and identify their limitations.
SMc’s observations that the reported temperature fields in Antarctica are contaminated by Chladni patterns are a case in point. The Chladni patterns are a result of the physical shape of the continent We are all discussing the resulting field as if it were meaningful but it is really just a contamination of the data by a sampling window. Engineers are very familiar with such contamination in the sampling of data and use well established signal processing techniques to determine the contamination and to reduce it to an acceptable level. No such technique is reported here even though such contamination was acknowledged by previous workers in the field over 30 year ago.
What sort of statitician offers to help people while publicly pillorying them? Is that the kind of statistician you ask for help? Or the kind that knifes you in the back in a partisan report to Congress, which itself has errors. Is that the kind of statistician you ask for help?
If they do offer help the they are pilloried. Why would any statistician, mathematician etc. want to be associated with such a community? There is a lot more interesting topics to do research on than pointing out the errors in primitive techniques generated by non-specialists. The non-specialists will call it “total garbage” and “go to town on it”.
Just look at what SMc has to put up with and ask yourself why would anyone voluntarily subject themselves to it. We all owe SMc a huge debt.
You really don’t know, or choose to ignore, the history here. I’d suggest you go back to the early posts and catch up.
Stereo, welcome to climateaudit. You appear to be very new to this topic and quite uninformed. You should start by reading the Climategate emails to get a sense for just how unwelcomed outside help has been to the Team throughout the years. You should also read the Wegman report, where they describe the total lack of statistician involvement in the Team’s novel use of statistical techniques. Steve has pointed out on many occasions that the statistical techniques used by the Team are often unpublished, untested, and unknown in the statistical literature. Further, the Pal Reviewers often have no understanding of statistics and have proven to be quite un-inquisitive regarding the novel statistical methods invented by the Team.
Are you still replying to me? Do you actually understand my point? I replied to a comment that this is a very narrow field. It is not, rather it is very broad yet we are constantly told they are the experts/authorities. If you think climate scientists aren’t claiming this you are not paying attention. The Team’s arrogance regarding nearly any subject is monumental.
Mark
Well now that you’ve gotten taken his matlab course and published a rebuttal, should we expect Rasmus Benestad & co to now do a definitive rebuttal after learning about wavelets?
Thanks Steve, it was probably your phrase that I remembered from the earlier posts on Steig, but the decent cove bit came from his rapid response to the Tijander varve affair, which made him. look like a real scientist. Can a reviewer pull a team together to refute a paper? Isn’t the reviewer there to give their own opinion. Does sharing the paper with others not constitute a breach of trust with the journal?
Congratulations for a “team” effort. It’s the biased review process that, IMHO, is part of the ongoing behaviour that leads more and more independent scientists to become sceptical of establishment “climate science”.
Phillip Bratby,
I agree with your observation.
The more the behavior of biased climate scientists becomes publically known, the more I expect a trend were more independent scientists will become more involved in reaction against the bias.
I thank CA for bringing the audits of climate science to us.
John
Scince working in spite of the scientists.
Seriously, climate scientists have complained of lack of funding. However the Internet provides a new means of encouraging research. Here is an example of a crowd-sourced research that has passed peer review for a prestigious archival journal. If the data were freely available more such research would be possible.
Funding agencies could well take note that a good indicator of the worth of a piece of research is the use that is put with its data by other researchers. Researchers who choose a policy of restricting the availability of their data could well fund their work marked down in insignificance and attractiveness for further funding.
TAG, good comment. Apart from the funding issue, I think this paper represents a new way of scientific cooperation. I remember an earlier remark, although I cannot remember who made it, that blog postings actually hurt the progress of science because insights made on blogs cannot then be published in journals because the idea is no longer novel. It seems this paper proves just the opposite. Insights from science blogs can be drawn together to create a good quality paper. And co-authors rise to the occasion based on the quality of their insights and not their proximity to the lead researcher or because they did some grunt work.
Steve Mc has made offers of cooperation to ‘team’ members (I think it might have been Amman?) but they turned them down.
I found the whole “overfitting” issue raised at RC last year to be a tad ironic. In S09 much of the predictive power was held by the over-represented cluster of correlated stations on the peninsula, which is the most unrepresentative sample of the greater continent that one could conceive of. Yet the cut off at 3 PC’s actually made it impossible for the S09 model to express a regional pattern such as peninsula warming.
You state: Steig et al themselves did not appear to realize that Chladni patterns emerge from spatially autocorrelated time series, a peril known for many years (“Castles in the Air”) but apparently forgotten by Steig et al and the Nature reviewers.
However, given the timing of Steig 2009, it could well be that they were trying to influence Copenhagan 2009 and had not really forgotten it. After all, it took 9 months to get the rebuttal published with, as you state – the 88 pages of review and response were nothing more than obstruction – that ensured the rebuttal was published long after the damage would have been done.
The rebuttal is excellent, but I see more agenda than science in Steig 2009.
It seemed at the time that nature – as opposed to Nature – published a firm rebuttal more or less immediately, but of course one winter is merely weather.
What is the exact reference for the O’Donnell et al paper, and where one can get a digital version?
It’s not in print until the 7th I believe.
Will the 88 pages of comments and responses be available?
I think that December 7 is a very apropos date for publication. Interesting coincidence.
Congratulations to all concerned! And coming on the same day I read that the Japanese said stuff-it! to extending the Kyoto accord, things are looking up. Like others, I’m interested in seeing the finished paper, though I expect that it will be more a subset of the material on this and related blogs than a great advance, though I’d be happy to be wrong in that respect. New ideas are always welcome.
Can I just add my well done.
My admiration for you all grows every time I drop by here.
Steve: in this instance, admire Ryan, not me. I’ve gotten tired of dealing with obstructionist team reviews.
As an extreme layman, I don’t see how a simple regional analysis of The west Antarctic in itself would be sufficient to disprove Steig.
Anyway. Congratulations on your result. Hopefully a turning point in the publishing regime.
SteveL Steig’s principal finding pertained to the supposed identification of high trends in West Antarctic. This proved to be an artifact of his principal components which spread Peninsula warming into WA. And It’s not a “simple regional analysis” – it’s a technical analysis of methodology.
Like an old dog with a bone I expect the team to continue flogging the west antarctic warming the same way they pursue bristlecones. Nevertheless, these results are welcome not necessarily because of their results, but because of the application of the appropriate math to the problem.
Ryan O.: Big-time congrats! Obviously, you have had to be super-persistent and dedicated and tireless. I hope we are building more like you.
I’ll be honest. In my particular field, and especially in my institutional training, suppression of new entrants and abuse was the norm, so much so, that I think now that it is simply a part of how science operates.
This, of course, is a different issue from the special-case obstructionism, which is presumably at work here (?).
But if we do not separate the two, people are going to assume that amateur skeptic scientists and authors do not know the difference. I would request the authors to come out with their carefully considered opinion of the review process itself perhaps at a later point, so it can serve as a guide to others, rather than forget all about it in the joy of publication. High degrees of obstructionism is found in science fields everywhere, so characterizing its nature is important and would be helpful.
Congratulations, particularly to Ryan O’Donnell for this sterling achievement.
Obtaining peer review in the current climate (no pun intended) is a very significant step (already used achievement)towards pulling some of the teams teeth. Although i cannot appreciate the science, i am reassured by the certanty that your data and methods will be available to anyone who would wish to replicate your paper.
That is “Science”
Sorry, my last question was wrongly phrased. What I meant was – would not a simple regional analysis of West Antarctica (excluding the peninsula) have been enough to disprove Steig?
Afterall, his chart shows warming unsupported by the stations in the actual area…
Probably a stupid question, but it seems somehow logic.
The lack of sufficient surface station data in West Antartica, particularly long record data, means that carrying out a “simple regional analysis” of the area is, unfortunately, not practicable. The mathematical methods that we have used in the paper ameliorate the problems arising from a paucity of station data in West Antartica.
Are you able to comment further on this at this stage, please ?
As a geologist who has tried many, many times to make useful maps from very poorly distributed areal data, it is of intense interest to me
In a nutshell:
In this case, we had a secondary source available: the AVHRR data. Using that, one can empirically construct the covariance map for the entire area of concern, and then use the point data (in this case, the station data) to extrapolate that forward or backward in time.
Without the secondary source, the method cannot be performed. So if all you have is point data and no way to analytically generate a theoretical “field”, then you are reduced to traditional interpolation techniques such as kriging.
Kriging is modern. We were taught isopleth mapping in geology by a retired meteorologist. His preferred technique was a paper map with measured points plotted. Contours were interpolated with a three-sided engineer’s scale, a slide rule, and a very sharp pencil. I think he did weather forecasts for the Army Air Force during WWII.
Thanks for the answer, Ryan
It’s what I expected, but had naive, forlorn hopes for something else.
Geo’s are the “Kings of Krig” 🙂
Even in this reconstruction, there is some question of the history in the West. While it is vastly improved due to Ryan and Nic’s work, the lack of local data and opportunity for dense peninsula data to fill in the gap due to the close proximity of so many stations creates a problem which the four of us have discussed several times. The implication in employing these methods is that in this particular region, the uncertainty from methodology (not included in statistical uncertainty by variance) is somewhat greater than elsewhere in the reconstruction. Personally, I believe that due to the extreme warming in the peninsula created by circumpolar currents, that the West trend is very slightly increased over actual.
Compliments to Ryan O’Donnell for maintaining a throughly professional demeanor throughout all this.
And compliments to Jeff for sticking it to . . . .
never mind.:>)
Many congratulations for Ryan, Jeff and you Steve. It’s always a good show to see honest people managing to break the barrier of stubburn curmudgeon scientists.
I have attempted to post a link to this thread on Andy Revkin’s NYTimes DotEarth blog with a recommendation that he give it a thread of its own there. I will not be making book re the probability of getting through let alone his putting up a discussion thread.
I too would like to see more detailed disclosure and discussion of the peer review process at work in this instance, especially as it applies to a scientific journal. Over the several years this blog has existed, there have been numerous comments at ClimateAudit regarding the alleged pitfalls of the journal peer review process, and the counter-arguments relating to its supposed infallibility. Peer review is an issue that could be very interesting in the current context of web publication and especially interesting in the context of the existence of ClimateAudit.
Well done to all the authors. In addition to a rationally described investigation, showing a previous study could be significantly improved. Thank you for braving the review process which I am sure was tedious. One benefit of that tedium is a detailed documenting of the obstruction. Some editors and colluders know the lack of honesty is close to the surface.
Weird, your comments don’t sound at all like this:
“Overall, we find that the Steig reconstruction overestimated the continental trends and underestimated the Peninsula – though our analysis found that the trend in West Antarctica was, indeed, statistically significant.”
…which is from the abstract of the paper itself.
Gotta be a pony in here someplace…
They confirm Steig’s result. Isn’t that a pony?
Compare the two pictures. Would the one on the left have made it onto the cover of Nature? Would Steig’s own article with his own correction made it into Nature?
And maybe I missed it, but what is the temporal trend? In other words, does it still hold that most of whatever warming that’s there occurred in the first ten years of the record?
Good question. I would like to know as well. What is the trend?
“Would the one on the left have made it onto the cover of Nature? Would Steig’s own article with his own correction made it into Nature?”
Dunno. But the point is, was Steig essentially correct? And the answer appears to be, from this paper, yes.
Steve: Steig’s main novelty was: “Here we show that significant warming extends well beyond the Antarctic Peninsula to cover most of West Antarctica, an area of warming much larger than previously reported.” Using a more plausible variation on their method, this is untrue. Much of West Antarctica actually shows cooling.
“essentially correct” = “not entirely wrong” in “bcl speak”
Cooling always looks like warming when viewed through heavily tinted warmista glasses. If you just take yours off you will easily see what Steve says – much of Western Antarctica shows cooling, pulls the rug out from under Steig’s outrageous claims to the contrary.
But you can believe anything you want. Your choice. Your consequences.
Red tinted glasses ?
For varying values of ‘much’. Much in this case, from an eyeball check, gives me about 20%, while the peninsula is much worse than Steig’s analysis. Steig is therefore substantially correct, while underestimating the peninsula warming.
Jeez Stereo, Steig’s main claim is that the whole Antarctic is warming. It isn’t. Steig’s conclusions were and are “Substantially incorrect”.
What an ironic nick you choose, for one who’s comments are so “one-eyed” !
Re: Ed Snack (Dec 4 15:32),
Not only that.
As long as incorrect / inappropriate / suboptimal choices smear the warming across the continent, the picture of a continent experiencing consistent warming is plausible.
When this is corrected, and the warming retreats and becomes ever more localised, this not only destroys / significantly reduces the level of plausibility, it’s an obvious “dig here” for further clarification on why it appears to be localised – whether that’s looking carefully at the quality of the underlying data – as opposed to it’s analysis – or looking at localised physical phenomena.
But I agree with your main thrust Ed; there’s a decent chance that if the anomaly graph on the peninsula had a single pixel at an apparent boiling 20C rise, stereo would still be arguing that the warming was dangerously underestimated, rather than considering there may be a thermometer somewhere sited above someone’s frying pan.
Not only that…
But some of these sensors get completely buried in snow for days, perhaps weeks on end. Thus not reporting the actual air temperatures, but the significantly higher 0c.
You’ve got to love it when after 88 pages of hard discussion you still get this kind of shallow post… 😉
Nature Cover: “Antarctic Warming”
Results for entire continent:
Steig Trend: 0.12 ± 0.09, statistically significant warming
RyanO Trend: 0.06 ± 0.08, NOT statistically significant.
However, if you limit the “result” to ONLY West Antarctica and ONLY concern yourself with “statistical significance” and ignore the magnitude of the trend, it appears that indeed, this “result” is confirmed.
bcl: What is there about the word “refutes” that you don’t understand?
Or perhaps you didn’t have the patience to read Steve’s post all the way to the end, where he summarizes the paper:
“Steig’s West Antarctic warming results from a spreading of warming in the Peninsula to the West Antarctic through choices made in their principal components. Different choices – ones more plausible in the circumstances – lead to opposite results.”
in BCL’s world opposite results are consistent with each other.
The exercise is to keep attempting different phrases until you get a right one to spread around continents-wide. I don’t know if “not entirely wrong” is good enough for that.
Secondly, O’Donnell was in response to a 2009 paper, published in 2010 and for much of the year it sat around while the 88 pages were being filled out? One only hopes substantial changes were introduced by that exercise.
Steve: little value was added by Reviewer A and “he” imposed unwarranted burdens. If 88 pages of review commentary is regarded as a good thing, then Nature should have required it before publishing Steig in the first place.
I realize you can’t post the paper here, but can you post the reviewer comments and your responses?
Steve, Ryan,
Any chance those 88 pages will be made public?
Ultimately, the reviews and responses totalled 88 pages!
It seems like they did indeed redefine peer review.
Congratulations to Ryan et al. The team opened itself up to an massive defeat – trying to defend a flawed method.
I don’t want to appear rude, but BigCityLib do you understand what statistical significance is?
Congratulations to you Jeff, Steve, Nic, and Ryan. I really appreciate your tireless work to get to the truth and am eagerly anticipating the release of your paper.
Steve, ask the editor if you can publish the 88-page “review” here.
Re: reviewer obstruction. With 127 peer-review publications, I perhaps have some perspective on obstruction. I have had dozens of papers accepted with no revisions, including several in geology/climate that were not on contentious issues. I have had papers rejected because reviewers missed the point or wanted me to do something other than what I did, or because I did a poor job explaining my results, and I have dealt with nit-picking. But I have only encountered angry and nasty reviewers in the climate field, especially when the issue touches on climate change.
For more on Antarctica, including why the peninsula warms, go to:
http://notrickszone.com/2010/11/10/a-wind-in-antarctica/
The ex post facto selection of retained PCs is quite a handy tool for the Team. Seems that any data set can be made to tell any story given the right manipulations of retained PCs. Nice to see a clear exposition of the chicanery.
Notice the sector of west Antarctica which changes from warming to cooling by varying the retained PCs to a more plausible selection of Chladni patterns. Ironically this is where the Harry station is located – the station that led to Gavin the International Man of Mystery – a different Harry than Harry ReadMe. It’s just a coincidence.
As I recall, in the end Steig’s introduction of AWS stations like Harry was just window dressing, and had nothing to do with their primary “TIR” reconstruction, which was based entirely on the manned stations plus the AVHRR satellite-derived file.
I think you’re basically right about the Chladni patterns, however. With 3 principal components derived from the AVHRR covariance matrix, there is just enough resolution to pick up the TranAntarctic mountains separating E from W Antarctica, but not enough to see the Peninsula, with the result that the numerous Peninsula stations get smeared across W. Ant. without reference to their actual location.
Six of the “Pensinula” stations, BTW, are not on the Peninsula at all, but are on K George Island in the S. Shetlands, 100 km offshore out in the circumpolar ocean currents. They aren’t even in the 50-km AVHRR grid, that represents the Antarctic Ice Sheet, O’Higgens being the closest actual Peninsula station within this grid. These stations do show a warming that could be important in reference to the ocean, but don’t have much to do with the continent itself.
With respect to the off-shore stations, exactly. We did not use those stations as predictors.
I wonder if JOC will do something unusual in publication, like publishing rebuttals from Steig or others at the same time as O’Donnell et al.
A reply by Steig would have been a more appropriate method of carrying out the debate than allowing (in my belief) Steig and/or his associates to attempt to block and delay publication through abusive peer review.
Except that you don’t know it is Steig, but you are going to attack him for it publicly anyway. Why doesn’t that surprise me.
Steve: I said that it was my belief that it was Stieg and/or his associates – that is my belief. I don’t “know” that it is, but it is my belief given the conduct of Reviewer A.
A pseudonym used to make a thinly veiled bashing of Steve, why am I not surprised? I wonder if a similar comment bashing Gavin at RC would last as long as yours has.
I am pointing out the not at all veiled bashing of Steig based on no more evidence than a ‘belief’.
The irony… it burns!
The evidence of Steig bashing exists right here in this thread. It’s not just my belief that it has happened, you can see for yourself that it is happening. No irony involved.
Steig will be judged by the company he keeps as well as by his own comments such as this below: https://climateaudit.org/2010/12/02/odonnell-et-al-2010-refutes-steig-et-al-2009/#comment-247102
His agenda is clear as are the tactics his team employs. It is not “Steig bashing” to document either, nor to refute bad science and those who publish and promote it.
So, will this put an end to the complaining by some persons that Steve never publishes in peer-reviewed journals? Of course, after 88 pages of reviews … will Lucia run a pool for the date it happens again?
Steig has made a few comments at Jeff Id’s. He makes some bitchy comments. He then makes a semi-substantive comment:
In the Nature article, the lurid blotch of red over the Ross Ice Shelf and the land area abutting it (where Harry is located) was prominently featured. Now Steig says that this doesn’t “matter”. What ‘matters” is the little knob of red, as a sort of extension of the peninsula.
However it’s not as though this knob derives from data. More likely it too is an artifact of retained eigenvectors – which in this case are merely Chladni patterns. Steig wrongly ascribing physical reality to Chladni patterns, something that Nature reviewers didn’t understand apparently and Steig still doesn’t admit.
Steig: “Nature has a funny way of conspiring against those that are trying to tell you that nothing interesting is happening!”
A classic unintentional double entendre.
He doesn’t get it. He is behaving like a 2-year old child who just had his toy taken away.
I loved his comment (at tAV) that he is looking forward to reading the paper. If he was part of the original review team, he’s seen it – weak attempt. If he wasn’t, can anyone expect that after his reaction to their initial analysis (brush up on your MATLAB, Jeff, thar be experts about!) including his temperature tantrum at RC that he is in any way looking forward to proof that he screwed up exactly as he was told he had?
Mark
He wants a pre-print, but why would the authors give him one, when all he wants to do is find something wrong with it ?
Sorry, but that was irresistible 🙂
Indeed. He still equates science with “trying to tell you” something. Perhaps science is trying to elucidate reality, in ways that are useful to society?
So your problem isn’t with the paper itself but the colour scheme of the magazine cover?
he I first came to blogs like this, the detais of the discussion were incomprehensible to me. Then I listened to the discussion and did some background reading. They became comprehensible and surprisingly to me quite within my capabilities. For example, the idea of retained eigen vectors and Chladni patterns are similar to factors in my work which can be issues in distortion. AGW is an important issue. The discussions here and elsewhere helped me understand important issues and the science behind them.
Now I am pretty good at making one line observations. However for this issue doing some wok to learn the issues is, I find, much more important.
bcl, seriously? Color scheme? That’s your defense of Steig? Don’t you understand that colors represent temps?
Let me break it down for you. Steig et al was much like MBH98. They tried to photoshop the previously understood temperature picture for Antarctica. And Nature bought it hook, line and sinker. When some real scientists and mathematicians (Ryan O, Steve Mc, Jeff Id) looked at it, they knew Steig was wrong. And they demonstrated it on the blogs. And now, finally, they have demonstrated it in a peer-reviewed journal.
This is a victory for science and a defeat for Nature. As long as Nature continues to push an agenda instead of science, they will continue to suffer embarrassments like this one.
Steve: it’s not like everything in Steig was”WRONG” but one point where Steig was unambiguously “wrong” was their misguided physical interpretations of Chladni patterns and their application of this misguided interpretation to select 3 eigenvectors (patterns).
So you would have used different colors? More yellow-y, perhaps? I like purple, but you see it so little in scientific discourse. How about purple?
bcl, you really don’t get it, do you? The “lurid blotch of red” on west Antarctica of Steig’s graphic represents temperature. It is not a color choice like a Jackson Pollock painting. Steig was trying to say temps there have been warming, but he was wrong.
On Steve’s graphic, the red is replaced by blue – meaning the area in blue has been cooling.
Geez… and I thought pictures were supposed to help!
Actually, he appears to have been right. The abstract says the warming in Western Antarctica is statistically significant, just not as much as in Steig’s paper. Ptonk! as Jerry Fodor might say.
Is there a point to any of this or just unending pettiness?
Layman – there is no point to BigCityLib’s posts. From observations over some time they never offer anything new, never raise interesting qustions, never demonstrate any understanding, never offer any humour, never show respect for other points of view and are indicative of a mentality stuck in the school playground. Quite the antithesis of the cosmopolitan liberal values his name appears to allude to.
I refer the respondent to Mark Steyn, the shagged sheep, BCL and google. A long read, but worth it.
Just got to the bit where Fallaci hurls her “stupid medieval rag” to the flood and her meeting with the Ayatollah is terminated. Slightly off topic for the finer points of Antarctic principal components, some would say – but BCL is clearly an all-rounder. Extraordinary world.
“The shagged sheep” No wonder he a smile from horn to horn.
Or did you mean shaggy sheep?
No it doesn’t. You keep trying to imply that the abstract stated that there was statistically significant warming in the ENTIRE Western Antarctic and that is not the case. Here is the actual sentence from the abstract:
now if you look up West Antarctica you will find that Marie Byrd Land is part of it:
http://en.wikipedia.org/wiki/West_Antarctica
Look at the picture from the O’Donnell paper and you see that in parts of the Western Antarctic there has been cooling. Now look at picture from the Steig et al 2009 paper and tell me how much cooling you see there? Let me help, the answer is none. According to Steig the ENTIRE Western Antarctic is significantly warming and not only that but the warming there is greater then the Peninsula. So no O’Donnel et al 2010 does NOT support Steig 2009.
boballob,
Put this way the differences amount to quibbling. Instead of the ENTIRE Western Antarctic warming, this result shows only that MOST of it is, and not as much Steig et al thought. Not that these aren’t distinctions worth arguing over, but since pre Steig the suggestion was that outside of the East the continent was cooling, it hardly is a refutation, though, as the abstract suggests, it might be a refinement.
Mind you that Nature cover…hideously red, that! A definite sign of conspiracy!
The real signs of conspiracy are the sheer obstructionism of “Reviewer A”, attempts such as yours at ‘spin’, and goalkeeping we continue to see in journals despite an invitation to ‘do it yourselves’.
Other engineers, scientists and mathematicians continue to find glaring errors in substance and magnitude in your precious bias ridden ‘peer reviewed literature’, yet you and others part of the team confirmation bias keep digging the hole.
There are good examples of retaining your pro-AGW view and addressing both the strengths and weaknesses – Judith Curry seems to have some of the best approaches to the credibility issues that the team has brought upon itself.
I think the obstructionism comment is somewhat misplaced, as it seems to be addressing a point that BCL did not make. In terms of the point that BCL is making, he seems to feel that unless we showed that everything in S09 was wrong that we largely agree with S09’s findings.
Since my earlier explanation to BCL has not clarified this for him, perhaps it is best to let this one lie.
The difference is that the warming of the Western Antarctica Peninsula was known and is easily explained through atmospheric circulation dynamics – regional warm air advection patterns. Not only the depth of depression there is increasing but the frequency of deeper depressions is also increasing! Therefore and in flagrant contradiction with Steig et al. 2009, catabatic winds and anticyclones coming out of Antarctica are stronger -in fact more snow precipitation there- thus regional warm air advections directed by the relief will also be stronger. That is hardly a sign of global warming… read Leroux “Dynamic Analysis of Weather and Climate” 2ed. Springer 2010
My recollection is that wind speed in West Antarctica had also increased over the period examined. Wind affects the reported near-ground temperature. Higher wind means greater mixing of the low-level air and, on average, higher temperatures due to that mixing.
No, he gets it. His game is trolling. Ignore comments that don’t speak to the point of the post and he’ll behave better.
No, I am merely pointing out what the abstract to Steve’s paper says vs. what he says. Why the disconnect?
Because you don’t understand the issues and make no attempt to do so
I believe the disconnect arises because your view is somewhat myopic.
The primary claim in S09 was not simply that the West Antarctic regional average was “statistically significant”. The claim was that the rate of warming was at least 0.1 deg C/decade and that it covered a much larger area than previously thought. Indulge me for a minute. 🙂
Claim a) The rate of warming exceeds 0.1 deg C/decade. In this context, “exceeds” in S09 refers to the lower 95% confidence interval. Our paper finds that the rate of warming exceeds 0.01 (actually, due to rounding, 0.006) deg C/decade for RLS and -0.01 deg C/decade for the E-W reconstructions. This is a fairly large difference in results. While both S09 and the RLS regional averages are statistically significant at the 5% level, they occupy very different ranges. Additionally, the difference between the two reconstructions is quite statistically significant (|p| < 5e-4). The claim in S09 was not simply that West Antarctica had statistically significant warming, but that the lower end of that warming was at least 0.1 deg C/decade. Our paper rebuts this claim.
Claim b) The area of warming is much larger than previously thought. In our reconstructions, statistically significant West Antarctic warming is limited to Ellsworth Land and the portion of Marie Byrd Land directly adjacent. This portion of West Antarctica was already expected to be warming. See Chapman and Walsh (2007), Fig. 12, Shuman and Stearns (2001) for examples. Lack of warming in this region was a key criticism of the Monaghan et al. (2008) reconstruction. So this statement in S09 cannot be referring to the Ellsworth region of West Antarctica, which is where ours finds the warming. The statement is referring to the warming over all of Marie Byrd Land and Ross in S09. We do not find statistically significant warming in this area; in fact, we find some evidence of cooling.
So if you restrict yourself to the question of whether “statistically significant warming” was found in West Antarctica and select only those portions of the results that agree with that criteria, then you are comparing the two papers using a criteria used by neither.
That is the reason for the disconnect.
I know your paper did not specifically address it, but are you aware if anyone is looking into the three weather stations that generated the data; data quality?
I’m not entirely sure I understand your question.
I think he means are there problems with station location that might produce misleading warming trends, or something like that.
Over the past year, I’ve read so much about manipulated or ‘adjusted’ temp data, that the same question pops into my mind. How reliable is the raw data?
No one has ever evaluated the systematic errors, such as albedo and wind speed effects, that certainly plague Antarctic temperature measurements, Tom. They’re probably not accurate to better than (+/-)0.5 C. If that, then one wonders about the physical meaning of any statistically extracted trend.
Your work has shown a very pronounced area of warming over the penisula. With your methods you have shown it is more local and not spread across all of western Antarctica as Steig did. I find this pattern hard to explain by increased CO2 globally. The next logical step seems to be a study of why the penisula is showing a pronounced warming trend. Is it UHI, Data problems, or something else. Is anyone looking into this?
The current theory about the Peninsula warming is an increase in the circumpolar westerlies. There doesn’t seem to be much argument in the literature on this. I do not have the required background to form a solid opinion one way or the other.
BCL,
1) Steve’s name is not on the paper you continue to call Steve’s paper. Improve your reading skills
2) the claim was that the Antarctic is warming. Even Steig’s paper showed fast warming in the first 10 years of the paper with 30 years of cooling to date. Improve your reading and comprehension skills
3) JeffId commented that due to the statistical methods utilized and lack of data, the slight warming shown is probably STILL an over estimation. Improve your reading and comprehension skills.
Looks like I also need to improve my reading skills. I see at TAV that Steve’s name IS on the paper!!
My comment on overestimation was addressed toward the West Antarctic region specifically due to lack of on site thermometers. Remember, the method looks for a blend of the most representative surface stations creating a region of influence based on covariance. Therefore two nearby points would be natural choices for the most influential info. The proximity of the peninsula to the West Antarctic region means that it is likely that the peninsula warming will blend inland somewhat and add to the trend in that region. I also believe that the effect is very small in this case.
One of the reconstructions we put in the SI was a closest station infilling. If data was missing from a point, we took the next closest station with data. The station anomalies were offset to match each other (no sudden steps). The procedure was done as a ‘sanity check’ on the result. Had Steig et al. done this, they would probably not have published their results. Our check got the same result as the more sophisticated method presented above and this is why I believe the effect of this spreading is very small. It is also why I am personally absolutely confident that the result is accurate.
Due to lack of data it could also be an UNDERESTIMATION.
You are still smearing data from where it is to where it isn’t. Claiming accuracy for these statistical methods only applies to your math skills.
I apologize for apparently misrepresenting your comment.
Kuhnkat,
Because of the relatively extreme uptrends on a dozen stations right next to the West the result is likely slightly too positive, perhaps 0.01, 0.02 or so. There isn’t a balancing negative nearby.
Also, you don’t need to apologize for anything. This is the internet after all.
2 out of three were wrong.
Anyone have any disagreements with my statement that the Steig paper showed approximately 30 years of cooling since the quick warming??
BCL.
Let’s put the two studies into medical terms, where AGW = a bomb that has exploded, sending shrapnel into a body, which would be the Earth’s climate. Dr. Steig and his team of doctors makes the diagnosis that, on the left hand, there is wide spread damage and the hand will have to be amputated (this is the part where everyone should be staring at the palm of their left hand, which, as luck would have it, sort of resembles the shape of the Antarctic). Dr O’Donnell et al takes a closer look and finds that there is some damage to the hand, but most is confined to the area on the thumb, just above the metacarpophalangeal joint (the joint that connects the thumb to the palm). The thumb might have to come off, but the hand is spared. Now, if I’m the patient trying to decide which doctors to trust, which, in this case, do you think is the one I should be listening to????
Oh, bad grammar on that last sentence. Oh well.
Guys, y’all don’t get it. BCL is a skeptic mole in the AGW community. His task is making the AGW community looking ridiculous with his trolling pro AGW comments. I must admit he’s doing a wonderful job.
Well I’d say that anyway the O’Donnell paper supports the AGW Team’s general view on Antarctica’s warming — though it refutes the particular pattern of Steig.
Skipping around threads to spread your misinformation?
I’ve refuted part of your claim elsewhere.
As for giult in authorship, it was not just Steig on that paper. There were several other team members – and they are equally guilty of misleading people. So stop trying to pretend the misinformation is coming from one person. It’s coming from a whole team.
On Antarctica, the climate community in general stands apart from the authors of this paper, who are:
Eric J. Steig
David P. Schneider
Scott D. Rutherford
Michael E. Mann
Josefino C. Comiso
Drew T. Shindell
It is a mistake to suppose there is one great consensus on Antarctic warming. These individuals comprise their own special camp. Unless any of them has publicly recanted …
Semi-substantive is correct. Actually there is evidence that the Ross Shelf is sensitive to warming. It has been retreating for much of the Holocene but the ice-front is apparently now pinned in its present position by Ross Island and Roosevelt Island. However the AND-1B drill core indicates that it has retreated further south during exceptionally warm interglacials, most recently probably during MIS 31 (though apparently not during MIS 5e or MIS 11 when the WAIS is often claimed to have melted).
Incidentally the cooling area in West Antarctca is mostly in Marie Byrd Land, not on the Ross Shelf.
Here is the link to Steig’s first comment on tAV:
http://noconsensus.wordpress.com/2010/12/01/doing-it-ourselves/#comment-41569
Notice the first sentences.
willard,
As I note in the same thread, I don’t think this happened quite in the way Eric wants to remember it. Eric was condescending at first, then lapsed into pique once it was clear that he was technically outgunned (fighting to the bitter end not to acknowledge the results presented in the paper) and finally shut down comments at RealClimate when it was clear he was on the losing end of the technical argument.
Artifex,
Thank you for repeating your interpretation of the events. That means a lot to me. A link would also be appreciated.
This was discussed in detail in many June/July 2009 Climate Audit/Air Vent posts and is from my melange of memories rather than a specific post. If you are seriously interested, I suggest you revisit the threads from this period and determine for yourself how well the arguments made hold up especially with this paper being published. The specific thread I took issue with is a June 2009 RealClimate thread (“On oversampling” .. I think). I tend to think it is bad form to mount a technical defense and then chop it when it is not going your way, but once again, that is an opinion and not one necessarily shared by the RC crowd.
I probably have a certain amount of bias here because at the time I was absorbing the contents of the debate/discussion raging at CA and tAV and it seemed pretty clear that the arguments against the S09 method were strong and easily understood and the arguments against were weak misdirection. This is why I am interested in seeing the reviewer comments. I am wondering if they were just another rehash of the weak blog level objections or if there was some technical merit apparent that was not discussed in the blog level review. (i.e. how well did the blog critique of the paper compare to what was done by the expert reviewers)
I think this is the link:
http://www.realclimate.org/?comments_popup=682
Ryan O’s ‘The wicket for publication is much more difficult’ was prophetic.
============================================
“finally shut down comments at RealClimate when it was clear he was on the losing end of the technical argument.”
^ This shows how serious the Team are about the truth. I’ve often found myself on the losing end of a technical argument with peers, but I’m more wedded to the truth, so I just admit they’re probably right (then do some work myself to verify what they said).
I’m not sure what it will take for the Team to lose reputation to the point that they lose editorial power in journals that they’re so clearly abusing. You would’ve figured there would be some sort of self-correction going on by now.
Actually, it would be hard to argue that there isn’t some sort of self-correction going on. The simple fact that this paper was published in JoC, and that what would appear to be a Team reviewer was essentially “replaced” is ample evidence that the field is warming to outside input. I do not think this paper would have been publishable a few years ago – not on merit alone, at least, and not in a prominent journal.
Mark
All important indicators of advance. Another is S. McIntyre listed as author. For some reason an inordinate number of people have spent vast amounts of energy since 2002 trying to prevent any kind of recognition being given to Steve. I’ve no idea why they bothered. But lack of understanding is no barrier to a delicious sense of glee that they’ve just been thwarted. If it matters so much to them, who are we to quibble? Another bottle of bubbly will do nicely, Jeeves.
I’m more inclined to think it’s due to pressure from the CRU emails. Rejecting this paper without REALLY good scientific/mathematical reasons would raise more eyebrows.
In all fairness, I prompted Steig to close the comments with my final one. I personally did not see anything more to be gained with the somewhat limited venue that blogs provide. Additionally, regardless of what I said, until something was actually in the peer-reviewed literature, the impact of anything I said or would have said was very limited. Having already made what I felt to be the salient points, continuing discussion would merely have degenerated into the RC sycophants hoisting strawmen and dragging the discussion off-topic.
You are a far better man than I Ryan, though I disagree that the effectiveness of anything you said would be limited.
To paraphrase one of Sun-Tzu’s precepts: When your opposition is saying stupid things in the permanent record (like blogs) it is folly to give them a reason to stop … The Ghost of Comments Past has a long memory and will come back and haunt them when they misbehave.
Congratulations to your paper, Steve! I’ve always known that your first co-author Christine O’Donnell is not a witch. 😉
😀
Well Gav I says “What do you think then?”
“It’s much worse than we thought in West Antarctica luv and there’s nothing new about the rest of Antarctica. The planet still needs saving and I’m the one to do it”
Gav? Do you always where your shirt tucked into your pants?
Mr McIntyre with regardd to your perserverence it must be Scottish blood:-
“The Bruce beheld a spider try
His filmy thread to fling
From beam to beam of that rude cot–
And well the insect’s toilsome lot
Taught Scotland’s future king.”
By Bernard Barton
Congratulations on finally getting the paper published. Too bad that ten months of review delayed publication until the end of this year’s climate conference.
Excellent work Steve. Well done…
Wow, you guys rock in so many ways. Congratulations.
Please. please, do not allow BCL to side track what can be an otherwise very interesting and informative thread. He appears to have a propensity that is too often seen by “true believers” in that he has no concept of context and looking at the bigger picture. Certainly he is not here to inform or to have anyone change or even adjust his mind – so what is the point in engaging him.
In my opinion the Steig paper had legs because the slipping of the West Antarctica ice into the ocean is a very much bigger and alarming deal for sea level rising than the melting of the Peninsula ice.
I see a major contradiction within the Steig paper: The authors made the statement in their paper about the “novel” change in view of their results, that the Peninsula is not so much warming compared to other publications, but instead the West Antarctica is warming more. Later in the same paper they imply that the less warming of the Peninsula in their results ( but not smearing it to the West Antarctica) could be an artifact of the retention of fewer PCs. How did that get by the reviewers? The preferential warming of the Antarctica Peninsula is a very big deal to climate science and that makes this contradiction also a very big deal.
Finally, I judge that the Steig authors were very much aware of showing warming trends that were statistically significant and in order to do that one would be motivated to reach far enough back in time to do that. The problem in the Antarctica is that the satellite and AWS temperature measurements only go back to the early 1980s, as I recall, and prior to that only sparse temperature data was available from surface stations. The data back to the start of the satellite measurements (with close to complete coverage of the Antarctica surface) would not have shown statistical trends (by Steig et al methods) and one would have to start 10 or so years after the 1957 historic limit to surface data to obtain one. With sparse data, statistical significance was still out of reach. To obtain the significance desired, required that correlations of the satellite and AWS data to surface stations be made in the early 1980s time and forward to present and then from that calibration project it backwards using the sparse surface stations post 1980s.
That process is what, I think, some considered novel, although I will stand corrected if others here can point to similar processes used in climate science.
“In my opinion the Steig paper had legs because the slipping of the West Antarctica ice into the ocean is a very much bigger and alarming deal for sea level rising than the melting of the Peninsula ice.”
My understanding is that the Watkins Ice shelf is breaking off becasue its being undermined due to wave action. Its seems like there is very little chance the land ice of west antartica would ever see temps above freezing.
Is there any way of seeing which papers have referenced S09 and made use of the findings that have been challenged by RO10? Does the new paper make a difference to the findings of any subsequent papers?
http://adsabs.harvard.edu/cgi-bin/nph-ref_query?bibcode=2009Natur.457..459S&refs=CITATIONS&db_key=GEN
Thanks.
I picked just one of these papers at (sort of) random: Defining dangerous anthropogenic interference written by Michael E. Mann, PNAS March 09. On S09 Mann writes:
“Indeed, findings that have emerged even in the time since Smith et al. provided their latest assessment suggest the possibility that even their reassessment of climate change risk might be conservative in some respects. This is particularly true with respect to the potential destabilization of the West
Antarctic ice sheet and concomitant future sea level rise. A larger part of Antarctica appears to be warming than was apparent at the time of the AR4 report (10) and, while not necessarily indicative of destabilization of grounded ice, the Wilkins ice shelf now appears ready to collapse in entirety
(11).”
(10) is S09, (11) is a British Antarctic Survey report from 2008.
.
While I’m not able to get into what AR4 said in detail, it appears that Steve is right – the significance of S09 as viewed by other scientists was that the warming was not localised in the peninsular but spread to the rest of West Antarctica.
Congratulations to all involved! Thanks, Steve, for a bit more insight into the process and the commentary on what this means for the primary conclusions of the Steig paper.
Congratulations, Ryan, Nic, Steve and Jeff! I’ve learned a lot about SVDs from you through all this!
In defense of Steig, he did release the crucial AVHRR file that allowed his results to be replicated and evaluated, though this was only after a month or so of outside prodding via the editors of Nature.
The importance of this rebuttal or whatever you want to call it becomes even more obvious when you re-visit the Guardian’s take on the Steig et al. article when it was published – “Research ‘kills off’ climate sceptic argument by showing average temperature across the continent has risen over the last 50 years.”
http://www.guardian.co.uk/environment/2009/jan/21/global-warming-antarctica
Presumably various other media outlets had similar reports.
Want to bet they do not run a similar story on the rebuttal?
Try posting with a copy of the O’Donnell et al abstract and see if this is deleted
I was a little surprised to see that the “kill off” phrase was provided by Steig, not the journalist. From the article:
“The [Antarctic cooling] issue, which had been highlighted by global warming sceptics, was an annoyance, said Steig, despite the science having been reasonably well understood. “But it has now been killed off,” he said.”
Many congrats to all for getting the paper published – it is clear the decks are still stacked against sceptic papers, but I hope papers like this and the events of climategate will make sure editors are increasingly objective about publication.
One observation: the team like their “heads I win, tails you lose” arguments. Dr Steig’s response reminds me a little of the classic response to the variability of paleoclimate reconstructions. Hockey stick shaped reconstructions “proved” something odd was happening in recent times, variable reconstructions indicated a high sensitivity – either way was “not inconsistent with” the narrative.
And here we have Dr Steig playing the same game. Irrespective of the pattern of warming, he can find some reason why it proves that things are even worse than we thought before.
Watch the pea!
Steve wrote:
What we need then is a world where O’Donnell, McIntyre et al now also get a Nature cover and a full and fair explanation of what they’ve shown within.
You may say I’m a dreamer, like John Lennon. So let me set this intermediate goal: would Eric Steig now campaign publicly for Nature to do this? Until there’s this level of integrity in climate science – which means integrity in individual climate scientists, whatever the cost – we will continue to have a problem.
Talk of conspiracy is beside the point BCL. Whatever so-called consensus arises from, the answer is always individual integrity, to the tune of challenging the tiniest bias even in top-level influencers like Nature. (And yes, we’re expecting higher standards of climate science than any other science – because it is asking far more of every user of energy on the planet, even the poorest. That should be made clear in the job description.)
And if this great change is not yet upon us, we all the more appreciate the example today of the unpaid O’Donnell, Lewis, McIntyre and Condon. The best known admits to being tired of ‘obstructionist team reviews’. But however he and they see fit to publish in future, a growing number of netizens is learning to follow. The truth will prevail, one way or the other. The world turned upside down, as the radicals of the English revolution used to say.
“And yes, we’re expecting higher standards of climate science that any other science”: Gosh, I would like to see standards in climate science be up there EQUAL to those in other sciences, such as my own (chemistry). If they ever got to equality, I think that would do the trick. It’s here at CA and similar sites where I can find those standards (thankfully).
Well said, Richard Drake.
Jockdownsouth reminded me of what happened in the press when Steig et al. came out.
http://www.guardian.co.uk/environment/2009/jan/21/global-warming-antarctica
Wow!
Good old Damian Carrington, never knowingly underscared. Two days before, my wiki tells me, I’d picked the story up from Richard Black of the BBC with New evidence on Antarctic warming. Also worth noting is how on the money Chris Booker was three days later in the Telegraph, compared to any other UK columnist, with Despite the hot air, the Antarctic is not warming up. One can only imagine how much superior, self-satisfied flak Booker got from other arts graduates for doubting the ‘peer-reviewed science’ at the time. It’s worth those people asking themselves now how he managed to get so much right.
IMO, It was precisely this sort of over-the-top reporting and science by press conference that fed the sustained interest on the climate blogs. Over night, the party line went from “Antarctica is cooling but it doesn’t matter” to “Antarctica is warming just like we thought all along”. It was just too much to stomach for many, and fed the Javert-like fascination with the paper.
If Dr. Steig had just simply published his results without fanfare, the paper would have probably fared much better. It would have gotten some attention on the blogs, but faded. It would have gone on to be cited numerous times as “settled science” without any serious challenges. Instead, they got greedy.
Everyone just ignore bcl and perhaps after a while he will learn something ( but I am not holding my breath waiting for that to happen). I have seen his schtick here and elsewhere and he chooses only to see what his preconceived notions tell him and then debate esoterical points which miss the real point
By the way, it’s nice to see this blog get back to discussions of science.
10 months! Academic sympathies for sure. But would love to know how long it took ‘Team’ Steig to get their paper accepted…
IIRC it took Steig et al about as long to get their paper accepted. But so what? They made a hash of the statistical methods they used.
This paper corrects that.
Isn’t that how science is meant to proceed?
Not disputing that! And tes, that is exactly how science should proceed. But what about all this nonsense about ‘The Team’ getting an easy ride (“pal reviews” etc) whereas our Noble Teamslayer doesn’t?
It’s not just the elapsed time, it’s a review process that involved having to write ten times more than was in the original paper. If this was also true for Steig, please let us all know.
No idea. Ask Steig.
To be fair, I am not aware of Steig whinging about how long it took and the hoops he had to jump through. So maybe you’re right and it was plane sailing (guided on no doubt by those pals in academia and at Nature).
Any bets on whether the paper gets cited in AR5? 😉
Congratulations to the authors Ryan, Nick, Steve and Jeff. The Steig ea. paper achieved significant exposure in the scientific world, making the cover of Nature, and in the media. I hope those who disseminated the flawed results will take steps to follow up with Ryan ea.’s newer results.
A question from a statistical novice (me):
In these techniques to extend the domain of the original dataset by interpolation, extrapolation or more complex means, I always expected that in the derived expanded dataset, all the locations present in the original dataset would still possess the same temperature time-series as the original dataset. (ie. the temperatures at locations with instrumental data should be invariant in the extension operation).
Is this a reasonable expectation, and do these techniques (including the ones used here) preserve the original series at their locations?
Principal components analysis (PCA) involves projecting a large-dimension space onto a smaller dimension subspace, and thus information (which here may be “signal” or “noise”) is “always” lost in the process. Thus the answer to your question about whether “…all the locations present in the original dataset would still possess the same temperature time-series as the original dataset…” is no. Actually, a major problem with Steig et al. is that, by projecting to a 3-D subspace, the details of the “signal” (the pattern of the temperature trends) simply vanishes.
TAC, thank you. I understand that PCA itself is a lossless re-expression of the data – information is only lost if PC’s are discarded. Why use a method of infilling missing data that doesn’t preserve the station data in the derived set – surely there are statistical tools that allow us to infill missing data without detriment to the existing station data?
“Why use a method of infilling missing data that doesn’t preserve the station data in the derived set – surely there are statistical tools that allow us to infill missing data without detriment to the existing station data?”
First: Yes, such tools exist.
The short answer to the larger question — why use tools that don’t preserve the original data — is most likely “incompetence”. 😉
There is a longer and better answer, however. Despite what you may think, it is not usually desirable to preserve all the patterns in station data. The role of Statistics is to remove patterns related to “noise” and extract “signal”. PCA, when used appropriately, can be a legitimate and powerful tool for doing this. However, care must be taken; PCA is not “robust”. Results are highly sensitive to apparently small things such as: Precisely which time series are employed; “minor” changes to data values; the number of PCs to retain; how data are “centered”; etc. Because many climate reconstructions are based on PCA, it is critical that climate scientists handle these things correctly. Unfortunately, as SteveM has so diligently documented, this does not always happen.
“Why use a method of infilling missing data that doesn’t preserve the station data in the derived set – surely there are statistical tools that allow us to infill missing data without detriment to the existing station data?”
Yes there are, and don’t call me Shirley.
Sorry, with Leslie Nielsen passing, I just had to use it.
From Steig et al. 2009 :
How much do 3 PC’s explain ? : the text refers Schneider, Steig and Comiso (2004), according to which the first PC explains 52% of monthly T_IR variance and the second 9%, so the first 3 PC’s presumably account for between 61-70% of the variance.
Why are the other PC’s discarded? Clearly the authors do not consider the variance as noise for the purposes of the analysis – Steig et al. describe the loss of variance as a “disadvantage” of having a cut-off of k=3.
The only reason which occurs to me for discarding PC’s here is limitations of computing resources, yet the input datasets are not prohibitively large, fitting easily in the RAM of an average 2008 PC. My guess, having failed to determine the expression by Googling, is that the RegEM algorithm doesn’t have big-O exponential to data-set size or iteration count, so the computation would have been do-able (I’d be grateful if anyone could confirm or correct this).
The reason provided in Steig et al., as far as I can tell, is that:
and that
These observations, while notable and encouraging, are not by my understanding arguments for discarding portions of our measurement dataset. What’s the advantage gained, apart from a lower computing cost?
Another question, sorry if trivial or already covered :
In S09, Fig. 2 (reconstructed AWS), the 95% confidence limits spans approximately +/- 1 C for both East and West Antarctica.
Comiso 2000, cited by S09 : “The surface temperatures derived from infrared data are coherent spatially and temporally and are shown to agree well with Antarctic station data with a correlation coefficient of 0.98 and a standard deviation of about 3 C.” [..] “In some station locations, the AVHRR temperatures (dotted lines) are higher than the station values while in other stations, the opposite is true.”
Comiso’s fig. 6 reveals no pattern connecting the variations of differences (station location is not plotted, however if a connection exists with location, no such analysis or result is mentioned in the text).
So, how is it that satellite data (with a 3C standard deviation in the difference from AWS stations measurements) are used to fill-in the missing AWS data, yet the AWS reconstruction has 95% confidence limits spanning only about +/- 1C ?
Also, why does East Antactica, having a sparser population of surface stations, not have wider confidence limits than East Antarctica?
A quote from Steig in a Nature puff piece entitled “Making the Paper”:
The original instincts were right.
The puff piece was before Nature published the correction:
http://www.nature.com/nature/journal/v460/n7256/full/nature08286.html
I guess the explanations weren’t detailed enough…
Will there be a press release for the new paper? It strikes me it would also be a good subject for a mini docu “making the paper” feature – AFAIK it is the first one which had it’s genesis in a blog discussion?
Agree it could make a good story. You may have given me a little idea.
Unbelievable.
I don’t think the new paper should be viewed as an “improvement” of Steig. Flat out refutation is more accurate — if not in the pure statistical sense (though certainly much there as well), then at least in terms of what Steig claimed his paper showed, what the Team thought it proved, and what the media happily lapped up as true.
Most things that “can’t be right” aren’t. It appears (though I haven’t seen the refutation myslef) that Steig failed to be sufficiently self-critical on this one.
Is this symptomatic of climate science in general? Let’s hope that Andrew Weaver’s endorsement signals a change in culture.
My questions would be: do these new results affect the yearly mean temperature grid calculations over the last 50 years? What does Antarctica contribute to the mean values?
Antartica would contribute a bit under 9.5% of the mean global land surface temperature and a bit under 2.8% of the mean global surface temperature, if I have got my data right. But, if I understand correctly, some at least of the global mean temperature series (e.g., HadCRUT3) give a much lower weight to data from Antartica than this, as grid boxes with no data in them are excluded from the mean calculation.
I don’t think these new results will in any case affect the yearly mean temperature grid calculations, as they depend on actual surface station temperature records – which both we and Steig et al. used – and not on reconstructed gridded data.
FYI . . . Steve has kindly given me space to host the reconstructions here:
http://www.climateaudit.info/data/odonnell/
Saved in ASCII format (white-space delimited), so whatever your number-cruncher of choice, you should be able to read them.
Could you post a file describing the format?
Yep! Coming soon . . .
Carrick, readme.txt is now uploaded, and it describes the format a bit better. Let me know if you have any questions or think something in the readme file should be clarified.
The online early release expected today or so is still not up.
No, unfortunately. I have run into an access problem on the new JoC manuscript portal thingy, so I can’t approve anything yet. Perhaps remnants of the hardware upgrade that went so poorly.
They just released 5 early release articles dated 12/8, but yours wasn’t among them.
The current status I have from JoC is that it is “ready for EOR”. Maybe tomorrow?
It looks like your article was such a hot potato that it blew up the whole AMS system!
http://www.ametsoc.org/au_upload/index.cfm
Don’t be shy about e-mailing the staff if the portal won’t work right for you.
I can’t legitimately say “congratulations” on the result without reading the paper first. But congratulations on your perseverance, Ryan, Steve et al. Hard to believe that climate scientists would not be aware of the problem of Chladni patterns. Richman and others were (though they didn’t call them that). Does the team not read?
If you are looking for Chladni patterns in the paper, you will be disappointed. For a number of reasons, that section did not make it to the final draft. It is better suited as a standalone paper, given the amount of discussion involved.
I’m not surprised. Just a little disappointed. Steig won’t learn from his mistake until he understands exactly what he did wrong to obtain the result that everyone told him “can’t be right”.
What did they call them, bender? I’ve never seen the term in signal processing literature, either (or I don’t recall seeing it.)
Mark
Chladni – wikipedia does the job.
In acoustics, concerning vibrational modes and patterns in plates.
In SP, and I’m wildly guessing, spectral elements associated with frequency-related interactions with media or interfaces thereof. Probably terms to be
tossed out as noise if Andy Viterbi isn’t looking.
Thanks, but I realize what thye are, just curious what term the Richman paper used. Seems like a lot of things it could be called. Signal processing literature, though largely statistical in nature, uses rather different terminology for many things.
Mark
Check out this presentation by Livezey:
http://www.cpc.noaa.gov/products/outreach/proceedings/cdw30_proceedings/Livezey_PCA_PSU.ppt
Watch for references to monopoles, dipoles and tripoles.
Slide 13 in particular, says, in reference to the monopoles and dipoles over the US:
“The sequence of patterns is seen repeatedly in other analyses and can be considered an artifact of the geometry of PCA”
i.e. The forced orthogonality of the oomponents introduces a false pattern because the source data (weather and climate data) are spatially correlated.
I’ll check it out. Thanks bendamon.
I’m trying to reconcile an analogy to the types of data I work with regularly, FWIW. Generally speaking, my data would be relatively co-located (phased array,) though there are examples in which the array is quite large. I’m wondering, then, if similar problems arise which would provide me with some text from “my world” to review.
Mark
Ryan & Co —
Is there any way to superimpose your colors on the topo-photo that appears on the cover of Nature and also (without the text) at http://earthobservatory.nasa.gov/IOTD/view.php?id=36736 ?
Perhaps NASA Earth Observatory would give you the data and/or the software.
Just a thought!
— Hu
You mean like this?
Hm. How about I try to get the years right:
Where’s BCL, to complain about your color scheme? Nice.
He’s still trying to figure out how to pronounce “chladni”.
B-E-A-U-tiful!
Where are you going to post this? (After your article is up, of course.)
And when you do post it, could you please post the pair side-by-side.
I put it in a post here
http://noconsensus.wordpress.com/2010/12/09/antarctic-trends-graphics-and-discussion/#more-10999
Did you first try to publish this in Nature? If so, what was the reaction?
We did not. There were a few reasons; not the least of which was that Nature’s length limit is far too restrictive. Our paper runs almost 7,500 words. Another major consideration is that Nature’s authors guide requires distilling discussions down to the point where a lay person with above-average interest in the topic would still be able to absorb and understand the discussion without extraordinary effort. In other words (my interpretation), they prefer papers that are easy to paraphrase for media outlets. This would have required even more words to accomplish. In short, Nature simply was not a good venue for this.
Implying that it was not a good venue for Steig’s article either. Agreed.
O’Donnell map is clearer, but I can’t see a significant difference. Where’s the beef?
I’ reading the famous explanation of Godel’s proof by Nagel and Newman. Godel was able to develop a method whereby he could generate self-referencing statements that would be of finite length. The above is another example of a self-referencing statement. Where’s the beef?
Toby:
The actual measured temperature increases are in the Antarctic peninsula. This was common knowledge before Steig 09.
Steig’s claim was that all of Antarctica was heating up (although not at the level of statistical significance) and that there was statistically significant heating in West Antarctica.
If you look at Steig’s map, the heating is actually moved FROM the peninsula to West Antarctica. This is the result of a statistical artifact (the chladni patterns that everyone is talking about). West Antarctica is shown as having MORE warming than the peninsula. This is simply wrong.
2 more early releases on the JC site today, but still no OLMC 😦
You don’t suppose they are waiting for Cancun to end?
Naw, it was explained very clearly on RC regarding the Dessler “clouds do strange things” paper that its release at this time was a sheer coincidence.
Yeah, that’s probably the explanation… 😉
I seem to get all the delays! Haha! Password is finally fixed, paper is properly entered into the manuscript tracking system, and the current status says “Ready for EOR”. Maybe today?
Hm. Another release today . . . no paper.
One thing I learned from the review process on this one is . . . well . . . patience is necessary. 😀
Ffs, McSteve, you put a caricature in the place of Steig’s map. There is a discrepancy between what you put there and what is on the front page of Nature.
Your map looks more detailed than Steig’s. So what did you do? Simply used up all sort of available data to produce a map as if there are precise weather stations all over Antarctica.
Your feud is so strong that you have become blind. Both maps show that a big chunk of Antarctica is heating up and melting away. But then, this whole thing has been about your feud, not climate science.
The two maps in the side-by-side graphic in the OP employ the same scale. They don’t show the same pattern of warming at all. And guess which is correct.
“But then, this whole thing is about your feud, not climate science.”
Thanks for playing.
I’d say that a more correct (albeit longer) title would be:
“O’Donnell et al 2010 confirms the general view of the Climate Team(*) while refuting the particular pattern of Steig et al 2009”
(*) Antarctica as a whole is not cooling as the Skeptic Team shouts; it is warming, with most of the warming concentrated to the West.
The statistically significant warming is restricted to the peninsula that juts way out from the continent. Thus it is misleading to talk about “Antacrtica as a whole”. You’re doing exactly what Steig did – taking the peninsular pattern and pretending it spans a larger areas – in your case “the West”.
Is that what GHG-warming theory predicts: a response restricted to the peninsula? Or might natural variability play a role in there?
Some elements of the Skeptic Team have been shouting that Antarctica (implied as a whole) is not warming, which contradicts the views of the Climate Team and proves that AGW theory is a hoax.
It seems to me that the position of the Climate Team has long been that, on the basis of atmospheric circulation models:
1-Most of the warming in Antarctica should be concentrated in the westermost part — though the westernmost part is to be identified more precisely
2-Taken as a whole in a average sense, Antarctica IS warming.
Which is why I’d say that O’Donnell et al 2010 confirms and adds some details to the general view of the Climate Team — while not in agreement the particular warming pattern of Steig et al 2009.
Why don’t you first try reading what the authors say about their results before offering yet another uninformed opinion.
I doubt that you understand which specific aspects of his own work Dr. Steig was trumpeting in the press as new and different when his original paper was published. Then, in this particular context, please inform us exactly how these points were addressed by the Odonnell et al. paper. Your “seems to me” arm-waving does not constitute an intellectual argument nor does it address the real issues involved.
nono has been asked two questions here. Will he reply? My prediction is nono.
In short, O’Donnell is not in agreement with the patterns and findings of Steig.
As I wrote.
Except fo the “confirms” bit, which you also erroneously wrote. Only a member of the team or a blind monkey following them would suggest such nonsense.
Mark
Again: What do the models predict?
Significantly less warming in Antarctica than the global average, and most of the warming concentrated to the Westernmost part.
citation?
FYI: technically the word “warming” is incorrect. Neither paper actually showed that. “Warmed” is a better word.
MMark
You allege a “feud” – however it is an entirely one-way “feud” on Steig’s part. Please review my posts on Steig et al from the start- they provided reasoned commentary on an article that was receiving wide international attention. On the other hand, Steig’s responses almost immediately degenerated into derogatory personal comments about me at realclimate. I regret Steig’s conduct but this was his decision not mine.
Steig’s decision might still have something to do with a sudden epiphany, a reading of the US federal policy’s definition of plagiarism, and a conversation with an International Man of Mystery.
Shall we discuss motives behind team decisions?
Shall we read the blog?
Good idea. Please stop trying to rewrite history.
Re: willard (Dec 11 10:36),
I was quite frankly baffled by Steig’s sudden vituperativeness – which preceded Gavin the Mystery Man incident by nearly a week. I had asked Steig for data on Jan 23 and received a pleasant enough response saying that they were working on it. On Jan 27, he made an angry comment here https://climateaudit.org/2009/01/24/4945/#comment-322363 . Lucia weighted in later in the thread observing to Steig that no accusations had been made at Climate Audit of the type that he had alleged. Steig proceeded to make a number of personal comments against at realclimate.
My own posts at that point had been very mild – observing that it seemed reasonably enough to me that Antarctica was warming along with the rest of the world.
I am sure to be baffled by Steig’s sudden vituperativeness, but the link does not seem to lead anywhere. I am unsure that paulw’s point depends on providing a timeline about who started what. Everyone’s decision to do what they do is their own to own, as always.
The most important sentence from paulw’s comment is this one:
> Both maps show that a big chunk of Antarctica is heating up and melting away.
To know if paulw’s interpretation is correct, there would need to be a discussion of the results for climate science. Granted the statistical shortcomings of Steig’s work, what do the colors tell us exactly?
“Both maps show that a big chunk of Antarctica is heating up and melting away.”
And a much bigger chunk is not.
Endless reframing.
“Both maps show that a big chunk of Antarctica is heating up and melting away.”
They don’t show a big chunk of Antarctica melting away.
They show a warming trend that, if continued for several centuries at least, may lead to a big chunk melting.
Why does this gang so resent being corrected? It’s very anti-scientific.
Shall we discuss motives behind team decisions?
why do you dodge?
The gang liked O’Donnell article. Even RC wrote it was a nice study.
So why don’t you inform yourself before you write? It’s very anti-scientific.
Links, please.
Here’s Steig’s post-scriptum:
> For those actually interested, yes, I’ll have more to say about O’Donnell et al., but overall, I like it.
Thanks for playing.
Get real.
“paper in the Journal of Climate shows a somewhat better way to look at the same data”
What a joke. O’ Donnell overturns Steig, to the point where the Nature paper should be retracted.
Whether Steig “likes it” “overall” is immaterial. False claims that exaggerate warming get hyped. Steig knew this and took advantage. And you can’t retract the hype, in the way you can retract a scientific paper, once it’s been generated.
For Steig to say the public response to O Donnell is “zzz” is a joke. The J. Climate paper isn’t even out yet.
Note the irony. “nono” wants me to “inform myself” – when the paper isn’t even out yet. What a joke. I am supposed to read RC when all they are doing is poisoning the well? I don’t think so. I’d prefer to read science than pre-emptive propaganda.
RC sinks to a new low allowing such illogic to frame a discussion.
I think I’ll wait for the paper than quibble with you trolls.
Again, only a blind monkey following the team would actually believe Steig’s comment was not just facetious c.y.a. after his many tantrums regarding the work they did prior to the JoC article. He was embarrassed. Now he spins it as support for his original conclusions. You two fools fell for it. How surprising.
Mark
> He [Steig] was embarrassed.
Links, please.
Here is the beginning of the abstract of the O’Donnell article:
> A detailed analysis is presented of a recently published Antarctic temperature reconstruction that combines satellite and ground information using a regularized expectation-maximization algorithm. Though the general reconstruction concept has merit, it is susceptible to spurious results for both temperature trends and patterns.
Source: http://noconsensus.wordpress.com/2010/12/01/doing-it-ourselves/
Here is the opinion of Bending Rodríguez :
> O’Donnell overturns Steig, to the point where the Nature paper should be retracted. […] I think I’ll wait for the paper.
An interesting reframing. Will it be an endless one?
Keep it up, this nonsense with your partial quotes. Pretend nothing has happened. Move along.
“O’Donnell overturns Steig, to the point where the Nature paper should be retracted. […] I think I’ll wait for the paper.”
Cute. Juxtaposition of two quotes, each out of context. Such integrity you have.
Let me explain. It’s clear to me what’s in the O Donnell work, even though I haven’t read the published paper. Whereas it’s not clear to me what’s in the Steig reply – because he chose to defer.
The abdursdity that I was pointing to is YOU turkeys insisting I read Steig’s unresponsive response at RC before the paper’s even out! Get a grip! Reading propaganda is not “research”. “no no” don’t know the difference.
In contradistinction to O’Donnell’s quote above, Bending Rodríguez considers that O’Donnell overturns Steig to the point where the **Nature** paper should be retracted. A paper it has not read the yet. In this context, our ellipsis seems warranted.
The reframing by Bending Rodríguez is even more obvious if we compare it with some sentences from **Doing It Ourselves**:
> In my opinion [Ryan O’Donnell, the lead author of the paper], the Steig reconstruction was quite clever, and the general concept was sound. A few of the choices made during implementation were incorrect; a few were suboptimal. Importantly, if those are corrected, some of the results change. Also importantly, some do not. Hopefully some of the cautions outlined in our paper are incorporated into other, future work. Time will tell!
Source: http://noconsensus.wordpress.com/2010/12/01/doing-it-ourselves/
Let the reader decide which reframing to choose to feed a feud.
Elliptical quoting certainly merits due dilligence.
O’ Donnell refutes Steig. If Steig and his defenders want to pretend nothing has happened, that’s their choice. It certainly will minimize the appearance of conflict. But the science is what it is. No amount of spin can change that the two results are diammetrically opposed: the centre of warming is the tip of the peninsula; the warming is not smeared across half the continent, loading heavily in the west away from the peninsula. It’s a local phenomenon, not continental. O Donnell right; Steig wrong. Bluff called. You lose. Chips, please.
True, I’ve not read O Donnell’s paper. But precious few have, since it’s still not out yet. However the core of the argument can be understood by anyone who’s read this blog. That’s how I know where the two papers stand in relation to one another.
Willard ever attempts to reframe, trying to make this about me. If we make a concientious attempt to return back to the facts of Antarctic climate, we see that there’s a good reason why Steig’s result was so surprising and novel that it seemed to merit a splash in Nature – it was wrong.
Just like the hockey stick. (Anyone see a pattern here?)
But let’s stop talking about the talkers and wait until the paper is out. Then we can talk about something of interest: Antarctic climate, and a systemtic pattern of flawed analysis that continually overestimates and over-hypes the rate and scale of regional warming trends.
Co-author Steve McIntyre felt obliged to point out that O’Donnell et al 2010 is both an improvement and a refutation of Steig et al 2009 :
> An improvement in the sense that the PC retention policy of Steig et al 2009 lacked any foundation and smeared Pensinsula warming into West Antactica. A refutation in the sense that the distinctive claims of Steig et al 2009 ( as compared to predecessor views of Monaghan for example) about West Antarctica are shown to be an artifact of their methodology.
That’s consistent with Ryan O’Donnell’s statement above and provides a bit more detail.
Not only Steig and his defenders (whoever that might be, certainly not us) but O’Donnell himself tries to minimize the “appearance” of conflict. No amount of spin, dodge, crosscheck, hook, mindread, and other tricks Bending Rodriguez is trying here will change the fact that this site’s jester-o-matic is framing to feed the feud, while O’Donnell is not.
Anyone see a pattern here?
Merely highlighting a contrast – to set the record straight as to who is right and who is wrong. If there’s a pattern here of setting the record straight, I’m soooo sorry.
Here’s another one for you: McKitrick is right and Schmidt is wrong. Schmidt, like Steig, could admit his error and move on. But no, he has to dig in.
Anyone see a pattern there?
Here is the first comment from paulw:
https://climateaudit.org/2010/12/02/odonnell-et-al-2010-refutes-steig-et-al-2009/#comment-248778
The latest dodge (“to set the record straight”) and deflection (“here’s another one for you”) from Bending Rodriguez certainly do not help dispel the impression of a feud.
Anyone see a pattern here, there, everywhere?
Why don’t you go harass Steig and Schmidt to admit their errors? That would help end the “feud”.
willard the peacemaker
Bender, the editor’s comment wrt S09 shows how Nature percieved S09 as novel (and therefore publishable). The foundation for this comment is removed by OLMC10.
Yes, LL. In the ideal world the paper would be retracted because the basis for its novelty is what’s been found to be at fault.
It wouldn’t be worth pursuing at all except that this seems to be a pattern with Mann and Schmidt – the publication of incorrect arguments, and the defense of them at any cost. How they can recruit nice people like Steig to work with them is beyond me.
“what do the colors tell us exactly?”
It’s cold and staying that way./
It’s kinda funny actually, the writing was on the wall for allmost 2 years wrt to the findings of OLMC10. There were many blog posts here, at tAV, etc which posted workings (in fine detail) and the graphs and maps of general warming patterns which were refined into OLMC10. There were many challenges from warmers in the blog comments. Examples: ‘S09 is peer reviewed and you guys are denialist bloggers’; ‘overfitting’, yada yada. But I don’t remember a single peep from RC or any other warmer saying that the blog science from OLMC was definitely on the right track…..but that it “didn’t matter” because it only affirms S09. If there is any legitimacy to this narrative then we should have seen the genesis of it long ago. Instead we have this ‘insta-narrative’ wrt a paper no one has even read yet.
The best evidence of Climate changes are (unexpected) floods, earthquakes or sudden weather changes. That has been undoubtedly caused by human´s activity. Is there anyone who can deny it? So let´s stop argue about unimportant things and start to cooperate together to find solutions how to mitigate the Climate changes.
Earthquakes???!!!
Roman,
Real estate agents are like climate scientists…they know a lot about everything, and are more expert in all those things than the specialists. 🙂
Unfortunately, your assessment is likely correct. For all I know, she may be a very good real estate agent, but the “green” portion of her blog is not impressive.
On the 7 key green events of the decade page, we have COP16 as number 2, CFL bulbs at number3 and, Katrina (as part of Extreme Weather at number 4. Not high on real science or reality. Earthquakes do not seem to have made this list.
Anyway, enough OT and back to the business at hand.
Re: RomanM (Dec 12 12:47),
The down real estate market has obviously left her with way too much free time on her hands.
Does the December 2010 issue of the Journal of Climate come out this December?
The paper is now available on the Journal of Climate online early release page.
Unfortunately of course most people can only see the abstract.
bump
Finally, finally got my copy of:
“Improved methods for PCA-based reconstructions: case study using the Steig et al. 2009 Antarctic temperature reconstruction”
Now I get it. O Donnell decided to play by the rules and phrase the refutation in the form of a methods paper. That way it is productive and novel and the Team have to say they like it. They would be exposed as pure advocates if they didn’t.
That is a whole lot of work just to point out that someone was bonehead stupid.
This makes it crystal clear why they want control of the editorial process – they think refutationists should have to jump higher than confirmationists. What would Popper & Feynman think?
Haha! I’m still trying to figure out how I can get a copy of the EOR . . . I thought that authors were supposed to get a copy of the .pdf, but I don’t see any way of getting at it for the moment.
Unless I become an AMS member.
Which is expensive.
And so it won’t happen until the Big Oil checks start pouring in . . .
Ryan, Where do want it sent?
The corresponding author email listed on the first page would work . . . thank you very much!
Sent!
Andy Revkin has kicked off a discussion of this paper in the New York Times – along with Andrew Dessler’s latest arguing for positive feedback from clouds. Revkin ends up very happy with peer review, to an extent some CA may not yet manage!
20 Trackbacks
[…] Comments Phillip Bratby on Doing it OurselvesO’Donnell et a… on Doing it Ourselvesnvw on Doing it OurselvesKate Simpson on Open Letter On […]
[…] The audit of Steig et al Nature, Jan 22, 2009 by climate bloggers Jeff Condon, Ryan O’Donnell, Nicholas Lewis, and Steve McIntyre has been widely reported on WUWT, Jeff Id and CA. […]
[…] ein Jahr später erschien nun im Journal of Climate ein Aufsatz von O’ Donnell et al der diese bequeme Geschichte wieder ins Wanken bringt. Offenbar war die damals als innovativ […]
[…] O’Donnell et al 2010 Refutes Steig et al 2009 Do some of you remember Steig et al 2009, a pre-Climategate Nature cover story? Like so many Team efforts, it applied a […] […]
[…] direct op de cover, omdat dit in hun politieke straatje paste van Global Warming Alarming. In de Journal of Climate krijgt McYntire nu met collega’s een revanche op Steig en […]
[…] the Steig et al 2009 Nature paper? As Steve McIntyre points out at Climate Audit: “Like so many Team efforts, it applied a little-known statistical method, the properties of […]
[…] Journal of Climate paper alleging that the journal had recruited reviewers whose goal was “to suppress the article.” Using an exclamation point to reinforce his pique, he described the 88 pages of reviews and […]
[…] Strike Two: The Antarctic refuses to melt, too, as seen in a recent study in the peer-reviewed Journal of Climate that corrected an erroneous report that had appeared in Nature. The Nature study claimed that large parts of the Antarctic interior were warming. Once the statistical errors in the Nature study were corrected, the Antarctic interior was shown to be cooling still, as had long been believed. An explanation by Steve McIntyre, co-author of the Journal of Climate article, showing where Nature went wrong appears on the Climate Audit website, here. […]
[…] Strike Two: The Antarctic refuses to melt, too, as seen in a recent study in the peer-reviewed Journal of Climate that corrected an erroneous report that had appeared in Nature. The Nature study claimed that large parts of the Antarctic interior were warming. Once the statistical errors in the Nature study were corrected, the Antarctic interior was shown to be cooling still, as had long been believed. An explanation by Steve McIntyre, co-author of the Journal of Climate article, showing where Nature went wrong appears on the Climate Audit website, here. […]
[…] […]
[…] Strike Two: The Antarctic refuses to melt, too, as seen in a recent study in the peer-reviewed Journal of Climate that corrected an erroneous report that had appeared in Nature. The Nature study claimed that large parts of the Antarctic interior were warming. Once the statistical errors in the Nature study were corrected, the Antarctic interior was shown to be cooling still, as had long been believed. An explanation by Steve McIntyre, co-author of the Journal of Climate article, showing where Nature went wrong appears on the Climate Audit website, here. […]
[…] errors. And later, the blog commentary dealing with the methodological problems, ended up being published as a peer reviewed paper, by Ryan O’Donnell, Nicolas Lewis, Steve McIntyre and Jeff Condon, […]
[…] titre d’exemple, voyez le traitement envers O’Donnell et al. (2010), qui ont réfuté les travaux de Steig et al. (2009), publié à l’origine en grande pompe dans […]
[…] titre d’exemple, voyez le traitement envers O’Donnell et al. (2010), qui ont réfuté les travaux de Steig et al. (2009), publié à l’origine en grande pompe dans […]
[…] In the end, the autocorrelation issue turned out to be the least of the original paper’s problems: Ryan O’Donnell, Micholas Lewis, Steve McIntyre and Jeff Condon (J. Climate April 2011, 24:2099-2115) have shown that the main results of the paper are dependent on oversmoothing that results from retaining too few principal components of the satellite covariance matrix. They find Antarctic warming to be concentrated in the Peninsula rather than spread throught West Antarctica as in the Nature paper. Furthermore, average trends are less than half what Steig et al. found for the entire continent, East Antarctica, and West Antarctica, yet were much stronger than what they found in the Peninsula itself. See O’Donnell et al 2010 Refutes Steig et al 2009. […]
[…] Of course, this seems to be straight forward questions, but it isn’t all that easy to specifically determine. Well, it wouldn’t be to entirely oblivious people not paying attention regarding temps of Antarctic. Most veterans of the climate wars will recall the Antarctic battle royale bloggers vs cli-sci team, otherwise known as O’Donnell et al 2010 Refutes Steig et al 2009 […]
[…] I doubt they looked at this paper from 2010, posted by ‘The Auditor’ and co-author Steve McIntyre: O’Donnell et al 2010 Refutes Steig et al 2009, […]
[…] O’Donnell et al 2010 Refutes Steig et al 2009 […]
[…] The Red Antarctic Steig: O’Donnell et al 2010 Refutes Steig et al 2009, ClimateAudit, Dec 2, […]
[…] From the University of California – Irvine something that pretty well makes Steig et al 2009 even more irrelevant, since in that paper they did a survey of air temperatures and then smeared them around the continent using some Mannian style math, which was later shown to be faulty by O’Donnell et al. […]