Bishop Hill Steig snippets
Feb 8, 2011 Climate: RC The reaction to Ryan O'Donnell's article about Steig has been astonishing, and the rate of deletion of comments from Real Climate hasn't been this high for...well...days.
Some interesting stories have emerged from all the noise:
"Threats"
A commenter called CAGW_99 left a comment at RealClimate noting that Steig could find himself called to testify before the House of Representatives.
Eric, It seems that the authors have both scientific and ethical points that have been very well made on the referenced blog posts. I am one of many who are eagerly awaiting your response, although we have no expectation that you will actually choose to do so. Your “Team” will be changing the subject, creating straw men, ducking and weaving, and hiding behind moderation on this site.
There is a pretty good chance that you and your Team will get invitations from the new House committees which investigate matters like this. I wonder if your moderation works in front of a CSPAN camera? I wonder if you will speak openly or take the fifth?
The response from the moderator was to delete the comment and write:
[Edit. Resorting to threats of personal intimidation against scientists eh? Was only a matter of time frankly. Thanks for including your name in your email address.--Jim]
Fessing up
We learned yesterday that Steig had in fact been one of the reviewers of the O'Donnell et al paper. Now we learn that Steig went to great lengths to pretend that he wasn't. Having had the paper for the best part of a year, Steig left a comment a Lucia's Blackboard suggesting that he hadn't even seen it:
yan, if you don’t mind sending me a preprint, and a link to your reconstructed data, I’d appreciate it.
I will presumably have more to say after I get a chance to read the paper, but it’ll be a month or more as I’m simply too busy with current projects.
I would say that RealClimate now has a serious credibility problem. Mendaciousness on this level is going to be hard to ignore.
Reader Comments (95)
After Climategate, is there any credibilty left in the Team? You would have thought they would have learnt some sort of lesson. But no, the lying and dishonesty goes on as if they have cast-iron protection. When will politicians sit up and take notice of this corruption of climate science?
As many others have said, Wow. After been up half the night trying to take it all in, this is my fist response:
Eric the half a bee
Quite extraordinary
quite at buzz at Anthony's
Jeff no longer a retiree
88 pages of duplicity
all to to keep Antarctica warm you see
but no comments permitted at RC...
(apologies to Monty Python)
lapogus: Excellent
I'm sure Josh will come up with a great cartoon - perhaps all the Team digging in a big hole.
you didnt mention that cagw_99's name and ip address were posted by moderator at rc along with the moderator comment. talk about threats, that seems ominously 10:10 to me. cagw_99 should watch out for a visit from franny n' friends.
(they were deleted quickly, apparently)
why is this not surprising?
So after 88 pages of review responses we finally have the answer why. Peer review has been corrupted by the team and people still appeal to it's validity in climate science. Interesting times.
Phillip,
the problem is that they *do* have cast iron protection. I find the arrogance of the Team breathtaking given that, even after being caught red handed in Climategate, they continue to "redefine peer review" - but that arrogance comes from knowing that the MSM will never mention your indiscretions, the BBC will continue to make expensive documentaries about "Science under attack" and that governments will continue on a path of economic self destruction to "save the planet".
I hate being ignorant. But I have no idea what's going on and why it's important. I'm not a scientist.
This is why people like me need people like Delingpole. Can someone explain to a layperson what's going on? (I get the general gist that RC is a warmist site and Eric Steig has done something wrong with regard to a paper by Ryan O'Donnell and is trying to hide it).
I need an interpreter.
Buffy: I know, but there is only so long that the big lie can continue. Slowly their protection will begin to dissipate.
Caroline
I don't have time to write this up now. I've emailed one of O'Donnell's co-authors to see if he would write a layman's version. If anyone else wants to volunteer, do let me know.
Caroline,
Thank you for that request.
I too am bathing in a pool of ignoramce about what is going on.
Peter Walsh
even ignorance (oops)
PW
Caroline
In the simplest possible terms (and this from another layman) Steig co-authored a paper on Antarctic temperatures. O'Donnell disagreed with some of the interpretations and wrote a rebuttal which he submitted to a Journal for publication.
As part of the 'peer review' process, the Journal sent the paper out to reviewers, one of whom was Steig. (If sending a paper for review to the very person whose paper you are rebutting strikes you as odd you are in good company!) Steig appears to have done his best to make sure that this was not known going to various lengths to imply he had not seen the paper.
As reviewer (anonymously) he suggested an improvement in O'Donnell's method. When O'Donnell accepted this and made changes, Steig started criticising this method.
Ryan is not, as they say, a happy budgie!
Steig (and his colleagues at realclimate) would appear to have egg on their faces.
I'm sure someone will come along with a better explanation but i think this will do for you for the moment.
BH - thank you.
Peter W - good to know I'm not the only one feeling out of my depth!
c
Sam - I follow.
Question. Is it normal/wise to send out papers for peer review to someone so closely involved. Isn't the journal somewhat culpable?
Its quite simple really. The situation is worse now than it was before it was as bad as it is now.
Hope that that is clear enough for everyone/sark
don't for get this classic Eric Steig moment..
[Response: Being listed on our blogroll does not constitute endorsement. In general, the sites we do list -- whether they are run by scientists or not -- tend to get the science right much of the time, and hence are consistent with our mission. Being not-listed could mean that a) we haven't heard of the site, b) that it is uninteresting or unimportant, or c) that we consider it dishonest or disingenuous with respect to the science. Pielke Jr, Blackboard, and ClimateAudit all fall squarely into the latter category.--eric]
http://noconsensus.wordpress.com/2010/07/01/extreme-climate/
And it was my comment at RC that prompted Eric Steig to calling Climate Audi, Lucia and Peilke dishonest.
My folowing 2 comments defending myself from Ray Ladbury, never appeared at RealClimate
on the same article, eric steig made the following comment...
eric [Response: There is, however, no evidence that 'skeptics' are being shut out of journals. There is indeed much evidence to the contrary. This is a canard."
http://www.realclimate.org/index.php/archives/2010/06/what-do-climate-scientists-think/comment-page-2/#comment-178911
Caroline - I'll try to summarise it as best I remember the events from 2009, and my understanding of the developments in the last 12 hours.
Eric Steig et al produced a paper in 2009 which got front page on Nature and widespread media coverage, which argued that virtually the whole of Antarctica (and not just the peninsula) had and was warming. (the fact that Antarctica is just as freezing as it was 30 years ago was always a problem for the Team, a bit like the MWP which they did their best to erase from the record). The apparent warming of the frozen continent was achieved by use of poor and at best dubious statistical methods, which were quickly pointed out by Jeff Id and others at WUWT etc. Basically as just about all the weather stations are located on the peninsula and coasts, they had to extrapolate and interpolate this data, into the interior. But the resolution of their statistical interpolations was poor and there are still questions about the quality of the data they had from at least some of the coastal bases - i.e. errors were very likely to have been compounded. IIRC, Steig suggested to Ryan and Jeff that rather than argue it out on the blogs, they should publish a paper, under peer review. This is where it starts to get murky. Steig should not have been asked to be a reviewer, as he had an obvious conflict of interest. In most if not all scientific disciplines the lead author of the defending paper would not have been asked, (and definitely would have declined if he had been asked), but then be given the first right of reply following publication. But instead, he became Reviewer A, who tried all he could to thwart the paper's progression to publication. (88 pages of comments and obfuscation, 10 times longer thgan the actual paper). Ryan guessed that Reviewer A was Stieg early on, but still remained patient and good natured. At one point in the review process, Steig suggested that Ryan and Jeff should use an alternative statistical technique, which they then did. But then later, Steig then criticised the paper, citing the example of the same statistical technique as an issue (the one he had suggested). So Steig has laid himself open to charges of unprofessional conduct, duplicity. And that was when Ryan decided to bring all this out in the open. Meanwhile Gavin and the other members of the Team at the Real Climate (RC) blog have gone into overdrive in moderating any commenter who ask any reasonable questions about all of this. Basically this the evidence that peer review at least in climate science is broken. A number of commenters have said that the implications are as big as climategate. Hopefully this will give impetus to future inquiries into the behavior of the team on both sides of the Atlantic. Lastly, with the stushie resulting from all this, it should not be forgotten that Steig's original paper was fundamentally flawed, and should be retracted by Nature.
Hopefully that's a fair summary, if not no problem is others want to make corrections or add details.
Caroline,
It is, indeed, not normal to allow a reviewer with this conflict of interests. An author who is the subject of such a rebuttal is normally only given right of reply in the journal. For the editor to have allowed this, and for Steig to have taken it on, is a big no-no.
The reason for this is reflected in the review process where Steig put a large number of silly requirements to Ryan et al. It amounted to 88 pages of responses from Ryan for a 14 page paper.
To the editor's credit, he eventually pulled Steig off review and gave the job to someone else - the paper was then published.
Caroline, the bottom line seems to be that a new paper, which criticises one that claims to have detected Antarctic warming, was peer-reviewed by a lead author (Eric Steig) of the subject paper. Worse still, after the new paper had changed its analysis method because ‘Reviewer A’ (=Eric Steig) had formally requested it, Steig then wrote a rebuttal of this new paper on RealClimate that openly criticised the new paper for using the very same method he pushed them to adopt in his guise as ‘Reviewer A’. Then, just to compound the insult, he suggests that they should have used a method that, in actual fact, the new paper had used in its first draft but then removed/replaced because of comments from 'Reviewer A’.
As an aside, it seems that the overwhelming majority of the obstacles to getting the new paper published were down to demands for change/clarification from ‘Reviewer A’, such that the answers were around ten times the length of the original paper. Then, to cap it all, Steig is on record as making a blog request to see the pre-print of the paper in order to analyse it when, as ‘Reviewer A’ he’d obviously already read it.
This appears to be a rather clear (and shocking) example of the kind of sneaky tactics suggested in the Climategate e-mails and raises a very big question mark over the integrity of the peer-review process in the context of climate science.
In a robust manner, with a high degree of robustness, peer review has been robustly redefined.
Time to move on.
lapogus, Huub and Dave S.
Thank you so much for taking the time to explain. I think I've got it.
As an outsider it does seem the only way to get this to have any wider meaning is to get Nature to retract and/or apologise. They appear to have acted badly.....?
This is all perfect stuff for The Science and Technology Committee inquiry into peer review process
The Committee invites all written submissions on any of these issues by Thursday 10 March 2011.
"This is why people like me need people like Delingpole."
Those who criticize JD for his polemical style should bear this comment in mind. If he wasn't polemical, he wouldn't attract readers. If he didn't attract readers, then he wouldn't attract advertisers. And if he didn't attract advertisers, then the Telegraph wouldn't pay him. And his blog would not exist. And Caroline, and thousands of others, would have to look elsewhere.
Anonemouse. I will be submitting evidence. I hope many others will do the same. I don't think it is necessary to be a scientist or involved in publishing papers to provide evidence - members of the public need to be able to trust the peer review process.
can someone reassure me that, someone capable, is gonig to to submit this as evidence, into the Sci/tech enquiry on peer review. in a form that MP's can undestand clearly, what has been going on.
lapagus
Let us hope the stushie does not degenerate into a stramash or a collieshangie.
http://www.newstatesman.com/200104160022
Sorry. lapOgus
Well I dare say that a lot of the conversations that will go on at carbonundrum will revolve around how to mitigate the disaster called SteigGate...and with the help of Harribin and Black, by jove they will do everything possible to ensure this doesnt get out in to the wild (well, at least out in to general circulation at the BBC anyways).
Mailman
Caroline, and for others (including me) I am going to try and illustrate this... should be interesting and good for my brain.
I have started with an easy one - see www.cartoonsbyjosh.com
The 'Personal Threat'-comment is gone now, and instead moderator Jim characterizes commenters as
'uninvited loons'. No new posts have been let through for at least 12 hours (and all of mine have been rejected)
Thanks everyone for the explanations. The Bishop's "Team" comes up trumps again!
Peter Walsh
To a lay person such as I, the absolutely dishonest machinations at Real Climate mirror odious tactics at the Guardian, where the CiF moderators delete anything critical and factual about Bob Ward and others who write stuff that would not be accepted by a well-edited high school student newspaper.
Sam the Sceptic, you have a gift for accurate summarising and Iagopus, you have an equal but different gift for capturing detail - thanks to both of you for confirming for me that my interpretation of that which I was reading here and elsewhere was along the correct lines at least
Jane Coles wrote
On the other hand his grasp of anything of scientific merit that doesn't attract page hits is minor.
He's the UK equivalent of Fox news. His aim is page hits by tapping an underlying demographic fear.
To them, what he actually says is not important. That he's 'sticking it to the man' is.
Although throughly boring, sites that give real information have much more merit. BH is a long way there but there are plenty of sites that give deeper insight so long as you are prepared to spend t he time to be educated.
Please can you pass on this message to the Real Climate guys - I seem to have some difficulty posting there.Thanks.
'Chaps - I have some free time in my diary around the end of February. As you know I have an excellent track record of conducting 'independent' investigations specifically tailored to the client's requirements.
You may recall my most enjoyable (but brief) time at UEA, courtesy of Chancellor Lord Acton. I;m pleased to say that he and I developed a first-class relationship - almost telepathic in our mutual understanding of what was required.
Should you feel the need to conduct an investigation into 'Stieg-gate', I can assure you of my personal attention to fulfilling your needs.
There will be no need to disrupt the normal working of your organisation, nor will I cause embarrassment by asking questions of any of the parties involved.
Should the cost of my services be an issue this can be reduced a little by re-using much of the written work from UEA. However to maintain propriety in case of any FoI requests I will need to bill you for at least four days - no matter how much time I actually spend. Again for form's sake I will need to purchase airplane tickets (first class) to USA, but to save you trouble I will spend the elapsed time studying climate change issues on the beach on Florida rather than any onsite visits.
Assuring you of my most reliable attention at all times. Yours aye, Ron.
Poor Eric will wish he had chosen accountancy before this is through.
I did wonder at the time just who Reviewer A was - but I didn't imagine that the editor would have permitted Steig to do this.
It is an absolute disgrace. Compounded by Steig's going on to push O'Donnell et al. into modifying their methodology then criticising them for adopting his suggestions.
This is actually one step beyond gatekeeping. I cannot recall ever seeing an example of someone gaming the review process like this.
I am appalled.
@jerry
'Although throughly boring, sites that give real information have much more merit. BH is a long way there but there are plenty of sites that give deeper insight so long as you are prepared to spend t he time to be educated'
Are you sure that you mean 'educated' and not 'indoctrinated?'. Would you care to name some that you would recommend, and then list the ones that allow general questions without heavy moderation.
Judith Curry's 'Climate Etc' is one - please add yours to my list.
Oops - Ron Oxburgh seems to have infiltrated my previous post replying to Jerry. Bad Ron!
Go back to the Globe International office and do some 'being independent'. Remember that your clients expect to get what they pay for.
@Geoff
'Poor Eric will wish he had chosen accountancy before this is through'
Accountants worldwide are sighing with relief that he didn't.
Lord Ox
Personally I read Science of Doom, Pielke Jr, Lucia, even Watts (but with a high degree of suspicion). Then less often people like Matt Briggs and a whole host of lesser but honourable bloggers. I've tended to stop reading right--wing US Republican sites that have a veneer of science - although they are still revered on Watts.
Forgive my ignorance, Dreadnought, (Feb 8, 2011 at 10:31 AM), but what on earth is a 'collieshangie'? Can I use this expression to educate my collie???
I can’t see anything wrong with an editor receiving a critical paper and then sending it for review to the person criticised. In fact I would expect it, and in the area I work in it is commonplace. Of course, there are other referees and the editor must be strong enough to protect the subject. It’s not that hard to distinguish between serious objections and idle point-scoring. But if a reviewer’s comment is original and serious, while not being fatal to the paper in question, the correct procedure is to invite a “response” paper from the referee (and even invite a response to that, if one is available). For this process to work, referees must identify themselves – sheltering behind anonymity in such circumstances is completely unaccacptable. Equally, referees should not gang up on authors behind the scenes. That is what has gone very seriously wrong in this case.
Now that I spent the better part of the morning reading up on this, here, on CA, WUWT and tAV, I'd like to point out the reason why this is, imho, pretty much in the same league as ClimateGate.
While we all know about pal review by now, and how it is used to keep papers from being published which are not to the liking of The Team, this particular pal review is on a different level altogether:
* one of the reviewers was the author of the paper being rebutted, Eric Steig;
* the usual team tactic of delaying and b***sh***ing didn't work (88 pages of 'review'???);
* next step - propose a different method, whose application would be time-consuming - pity the authors accepted this, and put in the work;
* paper is out - and the reviewer now has the gall to criticise the authors for using the method he himself proposed;
* he does that not in a reply in the journal, he does it on The Team's blog, where the authors' reply have no chance of being published in full, if at all.
This is such a clear case of dishonesty that one has to ask if The Team are actually aware of the consequences of what they are doing.
I hope this will indeed find it's way to the HoCSTC!
On what the fuss means, I'd also add
given 'reviewer A' had previously had 10+ months to review it, presumably before and after the changes he requested and is now criticising were made. I followed the instructions, downloaded and ran it and it worked for me, but there seems to be a few new faces at WUWT saying they can't get it to run
Caroline,
Steig et al used algorithms to model the temperature of Antarctica. In doing so, he claimed to have identified statistically significant warming across the whole continent. Early on, Jeff Id at the Air Vent along with other readers including Ryan O started trying to analyse the technique used. They found that in configuring the algorithm, Steig had chosen a parameter, the number of eigenvectors, to maximise the warming across the whole of Antarctica. This was despite the fact that most of the observed warming appears in the Antarctic peninsula. The O'Donnell paper does a similar model, using 6 or 7 eigenvectors.
If you read RyanO's comment, you will notice that he talks about a test where he increases the temperature trend at one weather station by 0.5deg/decade. Using Steig's method, this causes an increase of the modeled temperature trend on the other tip of Antarctica of >1/10deg/decade. In contrast, using the method in the O'Donnell paper, the increase is 0.02deg/decade. From a modelling perspective, this clearly shows that the O'Donnell model works better.
Steig seems to be drilling down into particular areas of Antarctica and comparing the output of the models with other data where the comparison favours his model. However, the claim of the O'Donnell paper is not that they have an accurate model, but simply that Steig's model is one of several possible parameter choices, that his choice led to the greatest warming, and that other parameter choices change the status of the warming across the continent from statistically significant to not statistically significant.
In past careers I have been involved in reviewing papers (and having my own reviewed) in fields related to advanced engineering analysis. Several points come to mind:
- Having a paper reviewed (I had not come across the term "peer reviewed" prior to climategate) provides assurance to the editor that it is worthy of publication (presents original work, does not contain obvious errors, etc). Pre-climategate, I had not come across the notion that it provides some sort of confirmation of validity of the results. I have the impression that the phrase "not peer-reviewed" was concocted specifically to rubbish Steve McIntyre's work.
- A paper received by an engineering research journal (IEEE Transactions, for example) that comments on another author's paper commonly *is* sent to that author for their comments - but this is done openly, with no pretence of anonimity.
- Such reviewing is, in the fields I am familiar with, invariably done with civility and in a straightforward fashion. The author of the original paper may thank the new author for their corrections or perhaps may take the opportunity to clarify points which have been misunderstood.
- People and organisations behave consistently, in my experience. A firm that is nasty to deal with remains nasty to deal with. What we have seen here is entirely consistent with what we saw in the climategate emails - it was only a matter of time until something like this surfaced. I am sure there will be further instances from "The Team" and the RC crew. Perhaps they are already in the pipeline.
"[Edit. Resorting to threats of personal intimidation against scientists eh? Was only a matter of time frankly. Thanks for including your name in your email address.--Jim]"
Not much help if they don't include the OP, although clearly difficult, as it didn't contain any such threat!
When in a hole...
Viv Evans
Chambers’s Dictionary has:
Collieshangie, kol-i-shang’i, n. (Scot.) a noisy wrangling: an uproar: a disturbance. [Origin unknown]
It sounds as though it has something to do with dog fights, but that would be guessing.
It looks to me that Steig et al pulled a statistical trick similar to the Bristle Cone Pine trick used by Mann to generate a desired outcome. Ryan's paper points this out.
Can anyone quickly refresh my memory as to why Ryan et al's paper did not appear in Nature as at least a comment?