Wednesday
Jul282010
by 
Bishop Hill
Bishop Hill Snippets
Jul 28, 2010 Climate: CRU Climate: Jones Climate: Oxburgh I have guests at the moment, so no time to blog. Here's some interesting links though:
Jul 28, 2010 Climate: CRU Climate: Jones Climate: Oxburgh
Reader Comments (23)
From the Jones interview in NS, a comment from his minder Trevor Davies
"All at UEA now recognise that we should treat all requests for information equally, and not to try to distinguish between well-intentioned requests and those which appear spurious or mischievous."
Better yet, publish more data online by default, then there would be fewer FOI or EIR requests to deal with. Free the data, free up resources.
Jones sounded like a little boy just back from Mother Superior's office. Apparently she didn't have her ruler handy (in-joke for those of us who went to parochial grammar school)
Maybe next time she will have it handy. He is in serious need of it still.
The Phil Jones one is an exercise in not asking question which the interviewee may find ‘difficult’ , and shows all the journalistic skills of Hamster. Guess, it looks like in practice Jones controlled the questions that could be asked and made it clear that follow on questions would not be answered.
Now that sounds familiar and beings right back to the complete lack of any journalistic skills of enquire. The title really should have been ‘Phil Jones press release’ but I suppose that would make it too clear what this article really was.
But, atomic, methinks it's even worse than that!
Jones's first shot:
Hmmm ... how then to explain the inclusion of Wahl and Amman (2007) in AR4? Oh, wait, I get it ... it's only "alternative views" to which this "rule" applies! Didn't Jones, TD or Brahic bother to read Richard Horton's "Appendix 5: Peer Review" which constituted part of Muir Russell's creative writing exercise:
And elsewhere in this "let's clear a path for poor old Phil" campaign disguised as "journalism" ...
If Brahic, the interviewer, had been more on the ball, she would surely have recognized that "reviews for a journal" was not the issue vis a vis FOI. The issue is so-called "peer-reviewed papers published in a journal".
But regardless of whether it was a "review for a journal" or a "peer reviewed paper" ... why on earth would an author "need time to develop a dataset" after this "review" (which presumably was based on some data) was published?
Or has my understanding of the English language been corrupted by reading too much "climate science"?!
Phil Jones: " There should be an acceptance that the climate has warmed since measurements began. OK, there's then debate about what caused that warming. But I do find it difficult engaging with people who deny the evidence and say the world has not warmed."
But most of the Climategate emails did not concern "deniers" refusing to accept that the world has warmed. They dealt with very specific scientific peer-reviewed debates, such as McIntyte/McKittrick papers, about very technical issues such as the use of statistical tests, the size of tree ring samples for some periods, the smoothng techniques used, the number of factors retained in Principal Component Analyses of dendrochronologies, and suchlike.
On the other hand, many defences of the CRU actions end with the idea that ultimately what "matters" is whether warming is occurring, and whether humans have anything to do with it. But that is not in fact essential. Even if AGW is ocurring, the question is how much. Even if GW is occurring, it is important to know whether it is "unprecedented" or just cyclical. Most climate policy responses would hinge on these questions, not on the general question of AGW being taking place.
Again Phil Jones: "I think there are too many journals - and more are coming in climate. So if our critics use the excuse that they can't get papers in scientific journals... well, there are enough journals around. Getting work published is not a problem".
Of course, but publication in core journals such as Climatic Change or Climate Change have been very proactively discouraged and obstacled, even with Jones personally "going to town" (his words) to keep some article from being published, or getting another one pre-published in time to meet IPCC deadlines. Scientists in established climate science have also repeatedly argued that the peer-reviewed journals where dissenting articles are published (such as Energy and Environment, or Geophysical Research Letters) are somewhat tainted, precisely by letting skeptical papers be published, and should not be used by "real" climate scientists to publish their work. The published papers have also failed to get a considered response in peer reviewed literature (there is indeed a concerted effort to avoid citing them or their authors at all). All this does not diminish the worth of any paper, on either side of the debate, but is evidence of non-sanctum activities designed to influence peer reviewed literature one way or another. As Mao Ze Dong said once, let a hundred flowers blossom. The best "flowers" will ultimately endure the test of time and scientific criticism.
Jones again: "I haven't deleted any emails that were the subject of FOI requests". Important statement. He advised M. Mann to delete emails concerning informal exchanges relative to the IPCC report, and asked for others (like Briffa or Ammann) to delete them too. This happened two days after a FOI request was presented concerning those specific exchanges. So he says he requested everybody to delete emails but abstained from deleting any himself. Even if that is true (which one is entitled to doubt) he should explain why he asked others to delete such emails themselves. And those other scientists should be asked whether they actually complied with Jones' request.
Some time ago I learned that even if you delete mails in your email client, those emails remain all the same in the system, either in the entrails of your computer or in the mail server, and could if needed be retrieved by those clever IT young guys. It is actually quite difficult to completely obliterate an email trail, I'm told. I wonder whether British police or others have looked into this angle.
Paper by Emanuel contains several key admissions, and it would be interesting to analyze them (not attempted in this comment). One paragraph by Emanuel, however, merits mention: "Dividing the entire field of climate research into “believers,” “skeptics,” “deniers,” and so on is a particularly egregious tactic deployed by those who wish to discredit climate research. Science is not about belief, it is about evidence".
I could only applaud this contention, except that the conflation of "skeptics" and "deniers" is a usual tactic of the core team of scientists involved in Climategate, and is very frequent in realclimate posts (in fact it is one of its "egregious tactics"). Specific statistical criticisms by Steve McIntyre, for instance, and never discussed in detail, but instead we get a sweeping condemnation of "deniers" supported by "big oil". McIntyre is neither a denier nor supported by anyone but himself. He only has the annoying habit of posing precise, technical questions about details of past climate reconstructions, most of which have not been refuted (nay, not even considered or discussed) in "orthodox" climate science literature, where one only finds general dismissals based on ad hominen arguments or vague statements that his points have been already "refuted" elsewhere. One never gets the details of such refutations, and a mere reading of Climateaudit is enough proof that the few actual discussions of alleged McIntyre theses are based on misrepresentations of his contentions (one recent example is Steve McIntyre's discussion of a recent attempt by "Tamino" at realclimate, available at http://climateaudit.org/2010/07/27/taminos-trick-mann-bites-bulldog).
One hopes, perhaps too optimistically, that things will start to change anyway.
Hector M. "...those emails remain all the same in the system, either in the entrails of your computer or in the mail server, and could if needed be retrieved by those clever IT young guys. It is actually quite difficult to completely obliterate an email trail, I'm told. I wonder whether British police or others have looked into this angle."
If they have not, after how long - eight months?- on the case, it would indicate a staggering level of incompetence on the part of Plod.
While I agree with Emanuel in
Any one not familiar with the issue should read Sokal's Intellectual Impostures and be prepared to be shocked.
However, I think his comment:
is removed from reality. We should not forget that those "very few" had the power to make true their conspiracies. They were not being harassed they were the perpetrators of harassment by blocking publications, pressing reviewers, by the occultation of data and by spreading an attitude that unfortunately was not that uncommon in academic circles. One should not forget that this pressure has effectively barred many critical views from the scientific literature and has stopped the funding of critical research.
Have a look at these links if you are in doubt:
Paltridge's paper rejections
Pielke's NSF funding rejections
McIntyre has reported quite a few obstructions and I could give a few more from colleagues working in that field but they prefer anonymity to reprisals
Never underestimate the incompetence of Inspector Knacker and the boys in blue.
" PJ I think some information should be exempt from FOI requests. If I do a review for a journal I don't think I should have to release my review."
Why would it worry Jones if has given an honest , well argued review ?
I am aware of one Journal , at least , that now puts on its online version of the Journal the "pre publication history ". This is put at the bottom of the paper online and includes the reviewers name , the comments , the editors responses with corresponding dates etc. This is obviously easier to do with online versions but I believe it is a great development .
Maybe "Nature" and "Science" etc should be challenged to do the same.
hro001, you may be reading too much into that Jones quote.
I read it as saying that if he acts as a reviewer for someone else's submission to a journal, he'd like the review to remain anonymous (as many of these reviews are). If his review is subject to FOI/EIR requests, then it will likely not remain anonymous. I'm sympathetic to the justification that an anonymous reviewer can be more open and blunt in his/her criticism; this is a valid claim to the extent that the reviewer (who *does* know the identity of the article's author) is unprejudiced.
As to datasets, I think he's asking for a priority claim to first publication of analysis of a dataset, before releasing it to others. While that would be the norm for data in which there's a substantial contribution in analysis, I wouldn't think it would extend to, say, HadCRU temperature series, which are published in near real-time. But it would make sense to apply to, say, an analysis of station temperature series to establish UHI influence.
wtf?
-I think it is in the interest of science to know 100% of the reviews. All of it, every single word.
-is 150y enough , before releasing a dataset? or does he need more time.
The moron rattles on about him "doing science" (the new form of concilliums): Have to see the first bit of his doing science. It certainly is not his UHI China report.
I guess there aren't a whole lot of (ex)-Catholics on this blog.
For those lucky enough to be CoE or CoI or other, I will explain. Mother Superior slapped your hand (open and in her hand) with a 18" ruler. You could hear the damn thing break the sound barrier as it come down. Something you never forgot. Outlawed, sadly before AGW got going. It should be brought back for misbehaving "Climate Scientists". But that is just my thought.
I do remember one lad at my grammar school who had (past tense) the balls to pull his hand out of hers as the ruler came down. We all learned a lot of new words not in the catechism that day. You might say it was a "cataclysmic" education for us all, as it happened in the weekly "coming to Jesus" assembly. I have no idea where she learned those words. She must have been in the navy before going into the convent.
That is what master Jones needs, he does.
You could well be right, Harold. And perhaps it's not Jones's fault that the interviewer tossed him a question that gave him such a wide berth in which to ramble so ambiguously.
Nonetheless, since the question pertained to FOI and Jones's ramble did include:
I'm not aware of any instance(s) in which FOI was used:
a) before publication of a peer-reviewed paper (how would anyone know unless the paper had been published - or was at least "in press"?!); and (more to the point):
b) before the requestor had asked the author for the "dataset" (used in the paper)
AFAIK, no one has used FOI to go on a fishing expedition for previously unreferenced "datasets" that may or may not be used in some future paper. Ergo, why would anyone need "time to prepare a dataset" in response to an FOI request?
However, his response above could well lead an uninformed reader (perhaps new to the art and artifice of "climate science") to wonder what all the fuss was about. And/or to conclude that those darn harassing skeptics should leave these poor beleaguered "climate scientists" alone, so that they will have all the time they need to talk amongst themselves and tinker with their models.
Patagon,
In light of what climategate and the coverups reveal, the more interesting question is why would anyone mistake the people in the e-mails for scientists, and their work product as science?
And I think we should be given some time to develop a dataset before releasing it.
As Jones has said previously that for any of his papers the reviewers had never asked for the datasets I believe they have got into the habit of publishing before the dataset is fully documented and then never finishing it as the priority becomes the next paper. Hence Jones admissions about lost data and other colleages comments about the piles of unsorted data in his office.
And of course poor Harry will never be able to finish his data cleansing.
Don, I can see the scene clearly, tho I think you were lucky to be schooled before metrification, they had 1M rulers in my day, BOHICA!
Patagon
Physicist Alan Sokal's 1995 evisceration of 'Social Text' is a fine example of the discipline that hard science employs and social science (and climatology) doesn't! I particularly liked Richard Dawkins's phrase "postmodern metatwaddle" - I'm sure we'll find a use for that...
As hro001 and JohnH have, I think, correctly divined (Sorry your Grace), Our Phil does his work without organizing his data and moves on. That he should state this in so many words and so ingenuously suggests that that he still doesn't "get it."
Is this the approach one would want in a person responsible for maintaining important datasets?
I cannot understand why our governments, which seem to think this stuff is very serious, do not think it sufficiently serious to warrant a process audit by outsiders.
I suspect that we are not being well-served by our public servants on either side of the pond.
JohnH -
I think you're correct about the priorities, and moving on without leaving behind a "clean" set of data. At least poor "Harry" seemed to think so. And on a personal level, I've certainly seen such things in industry, where someone was given a task to analyze some fault or effect, and the only artifact which one can find afterwards is a report (if that!); the data can no longer be located.
I hope that journals become more strict with enforcement of data archiving requirements. First of all, it means that the data become part of the product, rather than just a supporting element to be assembled and then discarded. There will therefore be considerably more attention paid to compiling it with an eye to making it useful to others, e.g. the addition of documentation concerning its format and provenance. And second, if done properly, it would eliminate the need for FOI/EIR requests in at least some cases. But it takes a wilful effort, as it's much easier to skip these "non-productive" steps.
We can perhaps draw a parallel with software, and proper documentation (e.g. comments) -- it takes some time to produce useful documentation, and in "the old days" people were often not willing to spend the effort. But it became painfully obvious that good practices saved much more time during later maintenance of the software, especially as personnel turnover increased and some other "Harry" had to take over responsibility for some ill-described software. It's similar here: if a dataset is only intended to be re-used by its author, then documentation seems an unnecessary effort; but if others need to access the dataset (whether for auditing or for extending) then documentation becomes very worthwhile.
Regarding Emanuel's editorial --
"Climategate is merely the latest in a series of coordinated, politically motivated attacks that represent an aggravated assault on scholarship that should be of concern to every member of NAS who, if they are like me, joined this organization because we were tired of seeing scholarship enslaved to ideology, particularly in academia."
Wow! What a mouthful!
1) Sounds like Emanuel was a completely unbiased participant on the Oxburgh Inquiry Panel, doesn't it? Does anyone still have any doubts about the objectiveness of Oxburgh and Co.? A real vindication of Jones and CRU, eh?
2) The double speak about "tired of seeing scholarship enslaved to ideology" is absolutely disgusting -- Climatology has degenerated into nothing but ideology. As Patrick Michaels recently remarked in the WSJ:
"Climate Research and several other journals have stopped accepting anything that substantially challenges the received wisdom on global warming perpetuated by the CRU. I have had four perfectly good manuscripts rejected out of hand since the CRU shenanigans, and I'm hardly the only one. Roy Spencer of the University of Alabama, Huntsville, has noted that it's becoming nearly impossible to publish anything on global warming that's nonalarmist in peer-reviewed journals."
http://online.wsj.com/article/SB10001424052748704075604575356611173414140.html?KEYWORDS=patrick+michaels
One of the most illuminating remarks by Emanuel in his piece is: "Thus into the arena steps a whole host of charlatans posing as climate scientists." We therefore know where he was coming from when he sat on the Oxburgh panel. By implication he makes the suggestion that Steve McIntyre is one of a whole host of charlatans.
One thing Steve McIntyre does not do is pose as a climate scientist, despite frequently being invited to do so. Steve McIntyre is a statistician whose professional expertise was the analysis and, where appropriate, the demolition (or verification) of optimistic mining hockey sticks. He has shown that the Climategate climate scientists have been dreadful at statistics. The whole catastrophic warming scenario depends on competent statistical analysis. Remembering that the British are masters of understatement (which provides numerous cultural misunderstandings with our friends across the pond – see BP’s Tony Hayward’s experience), the killer, second excoriating conclusion of the Oxburgh enquiry is:
“We cannot help remarking that it is very surprising that research in an area that
depends so heavily on statistical methods has not been carried out in close
collaboration with professional statisticians. Indeed there would be mutual
benefit if there were closer collaboration and interaction between CRU and a
much wider scientific group outside the relatively small international circle of
temperature specialists.”
Re: hro001 and Jone's first shovel full:
Yet subsequent literature searches on the IPCC showed AR4 was quite happy to rely on magazines, NGO and activist web sites, tourist guides and a vast array of grey and murky literature. But this is all part of the shell game. Challenge your opponent to publish, knowing a small clique of right-minded climate scientists will likely be asked to review it, and they can then reject it.
See Hockey Stick Illusion for more details :)
They've published numerous 'peer reviewed' papers without providing supporting data to validate their results. People like Steve McIntyre have reverse-engineered it, found errors, so datasets have had to be developed to try and support the original paper(s).
Or, the data were just so sloppy and badly organised, they'd need time to clean it up. But then that's been the big problem with CRU. They seem very disorganised and sloppy from the Harry readme file and more concerned about jetting off to conferences than practicing quality data management. To me it's a bit strange CRU's now been invited to participate in the JISC data project, unless they're there as a case study of what not to do.
As for Catholics, current or lapsed, I wasn't lucky enough to have nuns, just Jesuits :p