May 29, 2011 
I try to be accurate in my commentary, If you think that my commentary has been inaccurate, I’d appreciate it if you would direct me to specific errors so that I can make appropriate corrections.
My criticisms of the proxy reconstructions here have been detailed and technical. My opinion on the lack of worth of the canonical reconstructions is based on my assessment of their technical and statistical defects, not on “innuendo and smear”.
Here’s a small start at responding to his post. I invite others to add to it.
Steve, just looking at your post on the Nursery, I see lots of evidence. Here, for example:
.
I haven’t heard of any incidents in which anyone requested “drafts of scientific papers prior to their publication in journals, with annotations, explaining why changes were made between successive version”, let alone “lots of requests” of this nature to multiple scientists.
Are any readers aware of any such requests? Or is this more fantasizing by climate scientists? Like the time reversal mechanism assumed by Nature when they blamed data obstruction by climate scientists back in 2005 on FOI requests in summer 2009.
This is an example of what I call the “If I didn’t see it, or my clique hasn’t mentioned it, it didn’t happen” fallacy falsely equating the ego with the whole world.
For someone who claims to keep their commentary accurate and technical, this seems to fail that standard by a long shot. It’s ridiculous on its face — as if the claim made by Nurse is proven/disproven by whether you or your readers are aware of any requests…
Second — technical? Accurate? The suggestion that “climate scientists” as a group engage in “fantasizing” on a regular basis hardly seems “technical” but rather to be opinion and a general attack on the whole group with only one reference to any evidence.
You follow that up with another example when speaking of climate scientists obstructing access to data. Again, you provide no evidence, present opinion as fact, and are less than precise in your terminology. Not very technical.
A technical commentary would be detailed and specify which climate scientists you mean, would provide evidence rather than referring to the entire discipline in nebulous terms and without data to back it up. I realize you have repeatedly discussed this before, but at least you could attach a link to where you do provide technical commentary…
That’s just the “Nursery” post which was short. I found more on the Wegman thread as well, but I won’t bother detailing them here. But it’s a great project.
~~~
So everyone — please, have at it. Post your examples of where CA posts have failed the accurate, technical and detailed test.
I’d have posted my response at CA, but I thought it would start a flame war and so decided to do it here instead.
The Policy Lass 
I would like to point out that his latest FOI request was for ongoing and unpublished research. Perhaps our dear Steve should look in the mirror.
Forgot to add: the Nurse comment appears to be the result of a game of telephone (or perhaps as others may know it “Chinese Whispers”. However, the ATI FOIA request to UVA certainly appears to fall under this category.
http://www.whatdotheyknow.com/request/all_information_held_on_ipcc2001#outgoing-42721
I’m willing to say Nurse may be referring to this:
Even more interesting, junkscience.com published at least some of the drafts (in contravention of the agreements necessary to get access) and the comments and responses are available online from the IPCC. Holland may be right, but he is WRONG!
Yeah, that would be another one. Since Steve and Dave are good buddies I would be surprised if Steve was not aware of this one…
Basically, he is a liar. But I’ve been aware of that for a few years. It is time others (the press and those who take him seriously) did the same.
TPL: Rattus, I feel uncomfortable with you claiming Steve is a liar without including direct evidence of a falsehood with the allegation. Please couch this in terms of your “opinion” or provide evidence to show where a lie occurred. Thx!
Sorry, perhaps I should say that in my opinion Steve is a master of dog whistle politics by using innuendo, cherry picking and selective disclosure of facts to create an implied atmosphere of fraud and deceit which surrounds the work of leading climate scientists.
And right in the next thread….
steven mosher
Posted May 29, 2011 at 10:24 AM | Permalink | Reply
Re: Nick Stokes (May 28 20:02), What you miss Nick is that UVa can start to produce documents as quickly as GMU did. You seem to think that they have to do a complete check of everything before delivering anything.
If you cannot see the difference between GMU and UVA then I’ll suggest this.
Write a very tight FOIA to UVA and see how they respond. That’s an idea for everybody.
All the readers here could take a specific aspect of the ATI FOIA and ask UVA to deliver on that specific aspect.
Wanna bet they fight it
Isn’t that how the CRU FOI bomb got going? And wasn’t it Mosher who started it?
TPL: Rattus — I appreciate your anger, but please mind your use of derogatory labels.
Susan, Mosher described it himself as an FOI spamming, at Keith Kloor’s, which he freely volunteered that he coordinated including his own request being designed to fail.
I get it – Steve asks for data/results from something done in 2006 in reference to an email which appears to directly contradict public claims earlier made by Briffa (that they hadnt considered including the Schweingruber data), and because “staff at the CRU have returned to this data recently as part of a project funded by NERC, which commenced in May 2010” this suddenly becomes ongoing research.
You guys are only fooling yourselves.
If I have 3 sets of data and only include 2 in my results (because the third set screws up the nice picture I want to paint), then claim I didn’t consider using the third set when asked, have an email made public shown that I did actually try them, then when faced with a FOI request for the code/data, refuse it because i’m working on another reconstruction 6 years later with some of the same data: would you let me get away with that ?
TPL: This is a direct example of the innuendo and smear I wrote about in the post. You suggest without providing proof that ‘the third set screws up the nice picture’. No evidence. Just allegation. Thanks for proving my point.
Btw, something noone is talking about but check the last line from the email:
“Does that give you enough information to be going on with? I’d recommend using CRUTEM3 rather than HadCRUT3, because the correlations seem to deteriorate with the inclusion of SST data in some cases — though of course you can look into this yourself.”
They pick the temperature series to use not based on the quality of the data but on how good a correlation they get to their tree rings. Climate science well and truly in action.
TPL: I’m no expert but I see two scientists discussing the value of various data, not some conspiracy.
“TPL: This is a direct example of the innuendo and smear I wrote about in the post. You suggest without providing proof that ‘the third set screws up the nice picture’. No evidence. Just allegation. Thanks for proving my point.”
No that was my hypothetical reason for not including the data in my own hyphothetical study, where I was asking if you’d let me get away with that behaviour.
Anyway, I automatically assume everyone familiar with the “story” knows the results of rerunning the study with additional tree rings from the region. The study was rerun using all tree data within the 10 degree box around Yamal:
So adding more tree rings from the area really does destroy the results of an unusual 20th C rise.
The problem is that he reran the study by getting rid of what called “the CRU 12” and substituting those series with other series from a single site which show an anomalous growth trend. See http://www.cru.uea.ac.uk/cru/people/briffa/yamal2009/
Briffa’s rerun core count for his rerun was approximately doubled for the late 20th C from 10 to 20, from 6 to 10 for the last part (still VERY low, and statistically shaky). The additional trees he added had a late spurt 20% lower than the originals (and there was no corresponding jump in local temperatures measured by thermometers so the whole thing is somewhat academic anyway). McIntyres rerun on the other hand added ALL known tree data within approx 800 miles, with over 300 cores throughout the 20th C, and the result that there is no reconstruction temperature spike.
A dendrochronologist had something to say about McIntyre’s blind use of data:
There’s more on DO’s blog on the topic.
Is that the link you meant to paste ?
It seems to primarily discuss the “divergence problem”, which many climate scientists hope is due to some as yet undetermined anthropological effect (hence feeling free to hide that particular decline), rather than being a systematic problem with dendro reconstructions. Since however the point of these reconstructions is to model PAST temperature, you can’t chop out a proxy just because it exhibits the supposedly AGW caused recent divergence. It would still be valid for historical records, being unaffected by AGW climate disturbances.
“TPL: I’m no expert but I see two scientists discussing the value of various data, not some conspiracy.”
You think it’s ok to pick one temperature series to compare against because it gives better results ? I’m not saying a conspiracy, just that this kind of comment is typical for confirmation bias – you know the answer you want, and tend to stick to using data which reinforces that.
None,
as with McIntyre youjust complain that you think what is being done is wrong. To be honest, who cares if you think it’s wrong. Give it some more thought and demonstrate what is wrong, give it a quantity, show that the reconstructions fail or ANYTHING AT ALL. Fluffing around complaining that one particular data set doesn’t meet some standard you think it should is fairly useless. You need to be more scientific.
During his last round of hide the decline on paleo, not even a few months ago, McI gives a great example of an
here
I’m assuming our definitions of ‘technical’ and ‘innuendo’ differ.
The whole project is one big long dog whistle.
We’re maybe reading the same thing and you’re not understanding it and I am, but I think it’s pretty technical. He’s pointing out changes in Briffa’s methodology in the course of 2 years which went from showing a 14th C with an increase in temperature greater than the 20th C (how inconvenient) and a decline in 20 C temperatures (which was not shown in the original version – or should I say hidden) to having a much flatter 14th C fitting in with Mann’s discredited hockey stick.
And what is one of the known problems with that new principal component methodology which produced a better fit with Mann ? Well “The effect of using principal components on regional averages is to change the weights for individual sites, including the possibility of negative weights i.e. flipping the regional MXD series.” ie, tree rings getting smaller could mean warmer… makes a lot of sense.
I note you chop Steve’s quote btw, totally misreprenting what he said, which was:
“My opinion on the lack of worth of the canonical reconstructions is based on my assessment of their technical and statistical defects, not on “innuendo and smear”.”
Note quite sure where you meant to go with that, but my opinion of the lack of worth of these reconstructions is also based on my assessment of their many defects. This does not mean I never make a little bit of innuendo about other things – for example like Steve I may make some innuendo on why Briffa would change his methodology so much or why you would chop Steve’s quote.
TPL: He claims his opinion is based on assessment of technical and statistical defects, but he goes way beyond mere claims about the validity of a reconstruction to the character of the scientists involved and thus I think he crosses the line.
Deniers are involved in an attempt to discredit climate science because they know they can’t win the policy debate as long as the consensus is with AGW. So, attack AGW and the scientists who work in the field. Hence, the witch hunt against Mann and Jones.
If you want a more detailed reason why this quote chop is irrelevant to the fact that McI uses innuendo instead of technical detail, well, here it is, and he does it more than once. His first quote “The only reason that I can deduce is that the Briffa 2001 reconstruction had a rhetorical similarity to the Mann and Jones reconstructions in the 1400-1550 period – and therefore was shown, while the Briffa and Osborn 1999 version showed a major discrepancy – and was therefore not shown” is speculative, and by saying his is using innuendo. It is obvious that you heard the dog-whistle, as did many of the others in the comment section in the thread, as well as the many trackback posts from other “skeptiks”. If Steve wanted to remain technical, he would have done what Nick Stokes did. As McI himself noted, the site samples get smaller as time goes further back toward 1450. Nick Stokes, instead of assuming scientific malfeasance, did a thread on it on his blog * showing “what is bad about drawing plots when the data runs low. You produce spurious deviations which are not reproducible in artificial data, and are misleading. That’s why Briffa didn’t do it, as most scientists wouldn’t.”
* http://moyhu.blogspot.com/2011/03/whats-it-like-to-run-out-of-data.html
Secondly, McI says, “he effect of using principal components on regional averages is to change the weights for individual sites, including the possibility of negative weights i.e. flipping the regional MXD series. In particular, the closing uptick in the Briffa 2001 reconstruction may well depend on the flipping of data – a point that I’ll try to examine in the future.” Just more innuendo. This one is bit more sly. He uses the word “flipping” (2x) to describe the use of age-banding, which was a method to allow for the use of the older, smaller sample sizes from 1450-1550. As far as I know, he never go to the thread where he shows us why it is “sketchy”, yet another buzz word, lacking any technical use but full of innuendo. Maybe he couldn’t find a good narrative to do a post on it, or maybe others, like deep climate * already described the process as to head off any more innuendo, I’m not sure. But I what I just did is show that that innuendo is very easy without having to do any technical meanderings, and, as far as I see at CA, is merely a replacement for being ignorant of what is going on within the science.
* http://deepclimate.org/2011/04/06/open-thread-9/#comment-8539
“If Steve wanted to remain technical, he would have done what Nick Stokes did. As McI himself noted, the site samples get smaller as time goes further back toward 1450. Nick Stokes, instead of assuming scientific malfeasance, did a thread on it on his blog * showing “what is bad about drawing plots when the data runs low. You produce spurious deviations which are not reproducible in artificial data, and are misleading. That’s why Briffa didn’t do it, as most scientists wouldn’t.””
Hm… you do understand that McI is pointing out that one Briffa study using data X is chopped pre 1550 and post 1960, and another Briffa study which ALSO uses data X is NOT chopped pre 1550 (but still chopped post 1960 – hiding the decline) ? The study which agrees with Mann’s view is the one which does not have the results chopped pre-1550 despite using more or less the same data.
Remind me about consistency again.
Btw, since you are so concerned over consistency and are sure real scientists like Briffa would only chop results when caused by lack of data, you must have strong words for… Briffa…. who produced his Yamal series with an exponential late 20th C rise, refused to release the data, then eventually when forced to it transpired there were only 10 tree samples for Yamal in 1990 and 5 in 1996!! Breathtaking.
“He uses the word “flipping” (2x) to describe the use of age-banding, which was a method to allow for the use of the older, smaller sample sizes from 1450-1550.”
Er no you do not understand what he is saying. The particular principle components based methodology used has the property that if the growth rate of the rings decline during the calibration period while the temperature rises, that is picked up by the algorithm. Then outside of the calibration period, further reductions in growth are pulled out as temperature increases – ie the whole meaning of the tree ring (more sun = higher temperature = more growth) is flipped.
“as far as I see at CA, is merely a replacement for being ignorant of what is going on within the science.”
I can only assume you do not have any science/maths background or have not actually bothered to read what is written there. It’s so far off the mark it’s not funny.
snip
TPL: Keep insulting people here and keep trying to out people who wish to remain anonymous and you’ll be banned.
A nice Moshpit impersonation.
Hilarious.
Where are Steve McI’s climate reconstructions? Oh you mean he’s had all that data and never attempted to make his own re-construction?
He’s a joke.
TPL: Nathan, I appreciate snark as much as the next person but Steve McIntyre is not a joke, nor is what he does a joke. He has had a serious impact on climate policy in the US – far beyond his own expertise. He is a force to reckon with, and not a joke at all, sadly. I wish he was a joke, then we could all just go on about our business.
> Steve McIntyre is not a joke,
He should be, though. To the general public just like he already is scientists. Yes, that’s a bit of a challenge.
To me, what rubbed it in was his role in the silly r^2 kerfuffle — statistics expert my foot. He’s a sick joke.
I think perhaps you are missing the point of this thread. And you are proving it’s hypothesis, so thanks, I think. McI supposes that data is hidden to match other reconstructions. THAT is the innuendo. He does not offer any evidence of this supposition. Just a dog whistle. And you heard it. If he thinks this is inconsistent with Briffa 2001, he does not offer a detailed technical explanation, but handwaves with slyly chosen words that appeal to his narrative. A narrative you have taken as true, as shown by your Mannian gish gallop in the above comment.
Well, this isn’t what McI said, and what McI said is actually a bit closer to what Briffa describes (without the innuendo), but once again, you are missing the point. It is not about whether he is correct, it is whether he is offering any technical explanation as to what is correct and why. Instead, buzz words such as “sketchy” and “flipping”, which I’ll add is quite reminiscent of the Tiljander narrative of “upside-down proxies”. Thanks for reminding me.
I don’t need a doctorate in anything to discuss the current subject. I just need to know dog-whistles and politics, two subjects which interest me greatly.
More dog-whistling over at CA where Steve insists that the copy-past of boilerplate is not that serious. Really. Since Wegman didn’t know about it, he can’t be blamed. Wahh wahh wahh the mean climate scientists are so unfair!
/snark
Besides, “professionals” are saints, pure and virginal, while “academics” are evil, rotten pond scum.
/snark (no, I promise!)
Seriously, if this wasn’t such an important issue — the science inquisition — it would be laughable because it’s so transparent.
HTF does McIntyre know that Wegman was clueless?
Ah yes, the “refuse to release data”-claim. The one where McIntyre had to admit he had the data for many years already, when the UEA e-mails showed Briffa had contacted the *rightful data owners*, who had *sent McIntyre the data*. Even funnier is that these rightful owners completely confirmed the original data analysis with a more extensive dataset.
None, you’ve fallen hook, line, and sinker for the dog whistle.
Sigh. McIntyre didnt “admit he had the data” – he had been asking for the data sources of Briffa’s study. Briffa refused to release it, but when forced to 10 years after initial publication, it transpired that the data was a subset of some that was already in the public.
Most rational people would then ask why Briffa had not just said this in the first place, but instead the realclimates, deepclimates, stoats and taminos of climate science spin it as McIntyre “admitting” he had the data already.
Don’t know what email you are talking about, I think you have the story mixed up – please post a link to the relevant email.
Nope. I’m afraid you (and McI) are the one spinning this story.
Marco’s right – you have fallen hook, line and sinker for the dog whistle.
Lets clarify things first by stating the chronology of events:
2000: Briffa publishes reconstruction using data obtained under confidentiality agreement, with another party (because the other party were also in process of publishing based on this data)
2004: The third party having published some time ago, the data (as part of a larger set) is pretty much freely available and in fact Steve already receives the larger set (not knowing it contains Briffa’s subset) while looking into other reconstructions.
2005/6: Darrigo et al attempt to resolve differences between their reconstruction and Briffa’s but Briffa does not give them data (or tell them who to ask and what to ask for to get it)
2006: Steve asks Briffa for the data. Briffa tells Steve he’ll pass on his request to the owners (again without saying who it is, or what version of the data it is).
Note at this point that the data is already pretty much freely available – all that would have been required from Briffa is a statement “I used data X version Y” and everyone could have got it for themselves. Yet Briffa refuses, hiding behind his 6 year old, and no longer relevent confidentiality agreement.
2009: Due to data archiving policies in a journal, Briffa is effectively forced to reveal where his data has come from.
Maybe you think that Steve being told that Briffa will pass on the request to the third party absolves Briffa of any responsibility on the matter – despite the fact the data was clearly openly available at that point.
Again, all he had to do was say what data he used (not provide it, just say what series it was). Instead he said he’d pass the request on to the owners. Did he ? Well something went wrong somewhere because Steve never found out what the data source was. Given the field day Steve had when the Yamal data was outed (due to the low core counts, poorer correlation to the regional temperature recordings than other chronologies, and outrageously abnormal growth spurts on the cores in late 20th C) it is obvious if he knew what the dataset was he’d have blogged about it immediately. And since he had two years previously already been given a superset of the data (presumably by the authors) why would the authors not tell him that he already had it if Briffa asked them to ?
Anyway, here’s Steve’s specific reply to someone pointing out these accusations which you are repeating:
I think it’s you who’s falling for spin, not me. All Briffa had to say was I used dataset X version Y and it would have been the end of the story. Given the problems with Yamal, it’s not surprising he was unwilling to say that.
Funny you should cite the McI quote, as willard’s already done that below. 🙂
I am also amused by your speculations in both your chronology and in your presumptive assumption of what I think.
BTW, you appear to like using many parentheses and the phrase “I think…” a lot in your other replies – an obvious red flag there in case you’re unaware. You sure you’re not spinning the facts?
Anyway, this is the only relevant bit of your chronology, after removal of your speculations:
2004: The third party having published some time ago, the data is pretty much freely available and in fact Steve already receives the larger set while looking into other reconstructions.
Thanks for conceding that he already had the data in question. End of story.
As I noted over @ Eli’s, McI’s defense in the post is remarkably jejune considering his self-appointed role as an “auditor”.
This, and Nathan’s point below, damns McI. Oh – and as Rattus pointed out over at Eli’s, McI admitted to analyzing the Hantemirov data and found similar results as Briffa’s. Gee whiz.
——————-
In any case, this affair – and McI’s bleatings in general – is old news, and another sad indictment on the general irrelevance of McI. Unlike Susan, I opine that McI is but a bit player in the US climate policy stalemate. Michaels, Christy and the Pielkes are far more influential.
Frankly, I’m more interested in your motivations here. Why do you still defend him, despite all these valid criticisms?
Former skeptic, (I’m a former AGW believer 🙂
Well spotted about the link – I just think there’s more to it than Willard implies. I’m pretty sure it supports “my” version of things rather than his.
“BTW, you appear to like using many parentheses and the phrase “I think…” a lot in your other replies – an obvious red flag there in case you’re unaware. You sure you’re not spinning the facts?”
Yes pretty sure, not deliberately anyway, i’m after the truth. I also don’t like to state things absolutely because there is always the possibility i’m wrong, i pride myself on changing my mind when the facts change – it’s also unpleasant to state something as absolute fact then find out it’s not true, or there’s more to it. So better a cautious approach. The reason for my long delay in responding is that I actually contacted the source of the data to ask him if my view of what happened was accurate.
His response was rather interesting, and I have to correct some of what I assumed to be true.
This point:
“2004: The third party having published some time ago, the data (as part of a larger set) is pretty much freely available and in fact Steve already receives the larger set (not knowing it contains Briffa’s subset) while looking into other reconstructions.”
is incorrect. According to the source, the communication of the data to McIntyre was within the context of Briffa: McIntyre had been querying the source about differences in data between two papers one of which was Briffa’s. As a result of this the source sent his Yamal data to McIntyre “to verify Briffa’s calculations”.
The source also said something very interesting which i’m dying to write (I asked if it was ok with him that I posted this particular point in a blog comment, but he has not replied) but out of courtesy won’t (unless he ok’s it).
So overall the communication with him has changed my mind slightly on the matter. I now think McIntyre probably thought the Yamal data he had was what Briffa had, but had no way of knowing for sure, particularly since the data he had did not match Briffa’s claims of it being “more highly replicated” than the updated Polar Urals (which also lack a 20th C spike).
I still think the Yamal data is extremely unreliable in the 20th C portion, I still think by not having earlier stated its extremely low core count for that portion Briffa was being borderline fraudulent, and I still think that Briffa refusing to pass on the data when asked, relying on 6 year old confidentiality agreement which was no longer relevent (the source of the data at that point distributing it fairly freely), was behaving in a childishly petulant manner.
> I just think there’s more to it than Willard implies.
I’d like to know what I imply and what my version of things is. Pray tell.
In any case, the version of None is not very different from the one explained in the second sentence of Steve’s reply to Michael Ashley:
> What I didn’t know until a couple of weeks ago was that this was the actual version that Briffa had used.
Six months of research can save an hour in a library.
None, there’s nothing childishly petulant to have the original data owners give the data. It is also clear from Briffa’s response that he did not even know Hantemirov already sent McIntyre the data (because, oh surprise, McIntyre didn’t tell Briffa that).
Did you already read the link to Delayed Oscillator I have you? You might want to do that, to see how McIntyre makes plenty of questionable claims, and how you guys are cheering him on, likely simply because you like the ‘conclusion’.
You might want to read this: http://www.cru.uea.ac.uk/cru/people/briffa/yamal2009/
Note the portion at the end where he indicates what Hantemeriov said about supplying the data to McIntyre, some which McIntyre said was true in a comment on his blog.
I’m aware of that link and think it just spins furiously, implying Steve already knew he had the data. Given the field day he had when he found out what the data was it’s inconceivable he would not have blogged about it immediately.
That the authors had already sent McI the data has to cast doubt on the truth of Briffas email that he had forwarded the request to the relevant data owners (2 years after they had already given McI the data).
So why doesn’t he create his own reconstruction?
Steve’s method is to look at the data, raise a question/issue, then walk away. He leaves it unresolved. That’s because he isn’t interested in actually finding out what is going on, just in raising issues.
It’s the same tactics used by tobacco companies. Raise some pointless issues that make it look like there’s some doubt, then walk away.
Have you ever seen any reconstructions that exclude Yamal? Nothing changes.
If I claim to have proven the Collatz conjecture and you point out an error in my proof, it’s ridiculous for me and my supporters to retort with “well let’s see you prove it then!”.
Temperatures may be rising, and they may be historically unprecedented (within several millenia) but reconstructions such as Briffa’s Yamal are simply shoddy science and can not be relied upon as accurate representations. If you really want the science to be accurate and reliable, exposing the problems with things like Briffas Yamal is important.
None, this isn’t just math, this is empirical science. The idea with McI doing his own recon is demonstrating that he actually understands this stuff. You know, for the sake of his credibility with those who do.
…and you know, every time he has pointed out an “error”, it has invariably turned out (as established by those who do know what they are doing) that it was either not true, or without practical consequence. McI would have found that out for himself, had he gone through the whole process. It’s educational.
Why would it cast doubt on Briffas e-mail? How could he have known McIntyre already asked Hantemirov for the data? It was quite very clearly identified where the data Briffa used came from.
No, there are frequently multiple versions of data series around. McIntyre himself got mocked by climate scientists after his initial publication on the hockeystick because he had used the wrong data – data which was given to him from Mann in the fist place. Added to this was certain claims made by Briffa about high replication of the data which McIntyre could not get from the data he had, leading to potentially believe he had a different version (or at the very least could not be sure he had the same data.
Dude, all he had to do was to ask Briffa whether X was the same dataset as he used. Instead he kept on bugging Briffa to give him the data, even after Briffa made it clear it was not his to give.
No he didn’t keep on bugging Briffa; as far as I’m aware he asked once, then when Briffa published in a journal which REQUIRED publication of the supporting data he pressed the journal, who pressed Briffa.
Nice attempt at shifting the discussion. The facts are simple, McIntyre never asked Briffa whether the data he had was the same as the data Briffa used.
Moreover, McIntyre has also tried to get the data by pushing Science to get Briffa to release the data (and failed).
Most interesting remains the fact that McIntyre repeatedly discussed Yamal, but NEVER EVER told his audience he already had the Yamal data. Until he had no other choice…
Marco,
McIntyre revealed that he had the data in a comment in one of the Yamal threads (sorry, I don’t have a link handy and don’t want to wade through CA to find it).
This confirms what Briffa said in the link to his response to McIntyre which I provided above.
No Climategate emails were involved AFAIK.
Let’s not forget this one too:
[As previously commented on http://rabett.blogspot.com/2011/02/staked-goat-stew.html%5D
> Remember Yamal.
A link always makes it easier.
In his reply to Michael Ashley by Steve McIntyre, dated Oct 7, 2009 at 9:22 AM, we read:
> [I] already had a version of the data from the Russians, one that I’d had since 2004.
Here’s a permanent resource to remember Yamal:
http://neverendingaudit.tumblr.com/post/6078252343
I might even click through, but don’t want to give CA the click. Do you have a solution to that willard?
Hmmm. A hit is not much. I believe that if you click and come back, it’s even less.
I guess I could link to Archive.org’s CA, but it disappeared from the archives last November.
Perhaps wget is your friend.
Let me think about it. If that takes too long, my email is on my tumblog.
Rattus,
Only today’s NA post, referring to this thread, reminded me of your question.
I have found an answer to it: Webcitation. So here you go:
http://www.webcitation.org/61MB68LxL
will help everyone remember Yamal.
Thanks! Good idea.
(Link currently doesn’t work though)
Actually yes… tree rings in one place getting smaller and temperatures in another place going up, may well go together. Welcome to large-scale patterns.
“Actually yes… tree rings in one place getting smaller and temperatures in another place going up, may well go together. Welcome to large-scale patterns.”
You are not making any sense. The tree rings are supposedly responding to the temperature in that location. Not somewhere else. When compared to the temperature in THAT area (using the above mentioned principle component based methodology) they are inverted – not some other area (which is poor science anyway unless reasoned on a priori grounds – you cant just say “the tree rings indicate it got colder, but that must mean it really got warmer if we assume an inverse relationship”).
None,
I think you seriously misunderstand GP’s point.
The point that is being made is that it is entirely possible for temperatures (tree ring widths) to go down (temperatures lowering) in one area while temperatures on a larger (global) scale. Once you are able to understand that, you are on your way to understanding the problems and findings of millennial, global scale, climate reconstructions.
> while temperatures on a larger (global) scale
…go up. Precisely, Rattus, thanks.
Thinking faster than I type… Thanks for finishing the sentence.
No I perfectly understand the point, you do not seem to be reading what I am writing though (or I’m not writing it clearly enough) because you are not understanding mine.
Let’s say I use tree rings to calculate a history of the temperature around London. The idea is that if temperatures go up, the trees grow more giving larger rings. So, I start by taking tree ring samples around London. Next I put them into an algorithm which matches them against the recent known temperatures, and lets say recent temperatures have risen.
For some of the trees, it just so happens that the initial theory does not match, and in fact tree ring widths got slightly smaller over the period the temperature rose (remember we are only talking around London). The algorithm then picks this out as an inverse match and so these particular tree rings get “flipped” and in effect, are thrown into the calculation as if they had gone up, with their history being correspondingly inverted. Clearly this is bananas because these are London trees.
The only difference is that instead of talking about individual trees should be talking about groups of trees.
Even if we were talking about larger global areas, the concept is still bananas because then you effectively say everywhere where it got colder (over the callibration period in a warming climate) must have an inverse relationship to global temperature. Instead of having some places warm and some places cool in a somewhat random fashion, everything has a meaning – as I understand it it’s exactly this mechanism that caused MBH to mine hockey sticks from random data: the callibration period always went up (because negatives proved warmth, as did positives) whereas the further back you went in time outside the callibration period the less cohesive they were.
You also get bizarre behaviour like in Mann 2008 (IIRC) where depending on callibration period, the same proxy could be used inverted or non inverted (I’m not saying it affected the output, I don’t think it did, but it’s clearly not good science – if you don’t know which way up a proxy sits, you shouldnt be using it – you cannot sensibly use it both ways)
I think you need to look at the SI for that paper.
Mann does a reconstruction without Yamal, I believe.
That comment had nothing to do with Yamal in specific, I was pointing out the danger of having algorithms which opportunistically flip data against the “causality” of the data, as in 2008 (in response to the comments blindly supporting this type of algorithm behaviour as reasonable).
None,
as ususal you speak in motherhood statements. This is why you can’t be taken seriously as you complaint is specific. They did the reconstruction without Yamal. Were there any other data thatyou claim are ‘opportunistically’ flipped?
There are many factors which go into tree growth. Ideally you want trees which are not limited by anything except the length of the growing season which is why trees come predominantly from the cold limits of their range (treeline). But not all sites are created equally here so some sites which should show a response to climate do not because other factors are limiting growth (competition, nutrient or water availability, etc. etc.). Chronologies are developed based on the following heirarchy:
tree -> site (collection of trees) -> regional (collection of sites). Once a regional chronology is developed it is compared to the regional temperature history (if temp is what you are looking at) and at that point it is decided if the regional chronology has a correlation with temperature.
For multiproxy recons, some proxies have values which go down with increasing temps, some go up. For Mann 2008/2009 he used a a priori test dependent on the class of proxy being used, although for some (the Tiljander series in particular) there was no a priori expectation, so they used a two-sided test (BTW, the use of the Tiljander proxies is a weakness in the paper). For the most recent Mann papers he tested the effects of early calibration, late calibration and calibration over the full instrumental record. IIRC, no real differences were found.
I don’t recall any weird effects based on changing calibration period in either of those papers. A ton of this stuff is described in the SI for both those papers.
The reason that MBH was accused of “mining for hockey sticks” is because of the incorrect use of his centering method which promoted records with greater variance (in the PCA processed NOMER series) to more significant PCs in resulting network. This is a valid criticism. However, when corrected (see Wahl and Ammann, Climatic Change, 2007 — the “Jesus paper”) the results were substantially the same as in the original papers. For a first cut, it wasn’t too bad. Interannual variability in the past was underestimated, something which Mann himself found in subsequent studies of differing methods using synthetic proxies, but other than that a pretty decent picture of what happened, as far as can be discerned.
“(the Tiljander series in particular) there was no a priori expectation”
There very much was an apriori expectation – the scientists who took the measurements had given a reasoned argument for how the thicknesses of the silt should be aligned wrt temperature (I can’t remember if it was thicker = warmer or the other way around), and also pointed out that the previous few 100 years or something like that were badly affected by human farming residue and could not be used as an indicator. Mann ignored all that and threw them into the mix (noting it though in his paper). One might ask why he did that knowing the problems (and even though it didn’t actually affect “the reconstruction” significantly anyway) – IIRC (I had a quick check and can’t find a link though) McIntyre showed that without the incorrectly aligned Tiljander series it failed verification statistics.
“This is a valid criticism. However, when corrected (see Wahl and Ammann, Climatic Change, 2007 — the “Jesus paper”) the results were substantially the same as in the original papers.”
“substantially the same” but with very badly failing verification statistics:-
which Wahl and Ammann fought to keep out of their paper.
Regarding how MBH mined for hockeysticks, I look at it like this: The PCs are orientation independent, so if you take purely random data (L = low, H = high) you can have 4 combinations of (early, late):
HH
HL
LH
LL
Since the PCs are oriented to match H (high/rising) temperature in the callibration (ie late) period, you then get:
HH
LH
LH
HH
which when averaged together gives H for later period and midpoint for early period) – or a shaft plus a rise. A simplification I know but I don’t think it had anything to do with the PC centering – that was another issue (…I think)
Yeah, I think I estimated your “understanding” of these issues, and where it comes from, about right 😉
Try the literature sometimes. Or a textbook.
Do you have a good understanding of PC analysis then or are you just bluffing ?
Can I ask what your background is ? I took physics at uni, obviously involving a fair amount of maths, and PC analysis was not covered (or even mentioned AFAICR).
If my example does not have merit, i’m wondering why the red noise sequences (which would typically have more or less matching numbers and size of rising and declining end points) would not just cancel out since they have similar variance ?
Anyway, i’ll look into it and get back to you with a suitably groveling apology if my existing view of it is invalid (may take me a couple of weeks…)
As an aside, I love that “thagomizer” is now an accepted term for that part of a stegosaur.
I suppose it is better than “tenderizer”!
Well, I called out Mosher at a different forum:
Oh my. The inability of some there to understand plain facts is astounding.
A collection from the Yamal times here:
“Accurate, Detailed and Technical” or “innuendo and smear”?
appalled at the procedure
It’s very embarrassing for the field.
they’ve made biased picks.
pick the version (Yamal) that suits their bias
its enormous HS blade was like crack cocaine for paleoclimatologists
One never knows – it’s climate science …
And as a bonus from further past:
Is Gavin Schmidt honest?
Try not to puke.
And the full-throated defence of Wegman continues on unabated at CA! And more to come! How delicious.
In the meantime, plagiarism keeps popping up centered around Wegman, and Andrew Gelman, not quite an unimportant statistician, is quite clear on what he thinks about the plagiarism by Wegman and its implications.
> My opinion on the lack of worth of the canonical reconstructions is based on my assessment of their technical and statistical defects, not on “innuendo and smear”.
Even if this is a correct statement, it has nothing to do with Susan’s comment. It is quite possible to point out technical and statistical defects in one paragraph, and use smears with innuendos in another one. It might be possible to mix these rhetorical devices in the same paragraph. Actually, it might even be possible to do that in the very same sentence.
The underlying line of defense is that as long as one promotes correct statements, anything else can go. This is not an uncommon trick.
Trying to bring this out to a wider place in the discussion, where we can actually read each other (SW: It would be nice if you limit the depth of replies so that it is easier to follow…)
Good enough here…
Steve asked for the data Hantemeriov and received it. I think that it is safe to assume that Steve was smart enough to realize that he should ask the owners of the data for a copy. They supplied it, not as part of a larger subset which was freely available, since they were still working a longer term project.
Huh? What paper did D’Arrigo publish which was trying to reconcile differences with Briffa, et. al. QSR 2000? Link to paper please…
Was Briffa aware that Steve already had the data? Proof please…
Not true. The data was not freely available at this time. Briffa did not publish on this chronology, AFAIK, until 2009.
This much is sort of true. The data was not freely available until the publishing of the Phil. Trans. B article in 2008. I think that Steve has made the abundantly clear in his posts on this matter.
As for the rest of your rather incomprehensible comment…
Steve obviously knew enough in 2004 to request the data from the owners. He got it.
The version issues are pretty clear, it was the version used in the QSR 2000 paper. He asked for it and he got it.
Core counts were low in the most recent period (post 1983, apparently) because of when the cores were collected. This is problematic. Not having read the QSR article I don’t know if any caveats were given there. Do you?
Rattus, the paper can be found here:
Click to access ..%5Cdocs%5Cannual_climate_variability_in_the_holocene.pdf
Caveats are not given, but this is likely because Briffa focused on long-timescale variability.
Aaargh, I should have said caveats on the post-1983 core counts are not given. Plenty of caveats discussed in the paper otherwise!
“Steve asked for the data Hantemeriov and received it. I think that it is safe to assume that Steve was smart enough to realize that he should ask the owners of the data for a copy.”
Briffa told McIntyre he’d contact the authors. I doubt that he did (since the authors had already given him a copy of the data they were using for their paper – though McIntyre had no way of knowing it was the EXACT same version – why would there be no reply back to Briffa or Steve directly since they’d already actually given it ?).
“Huh? What paper did D’Arrigo publish which was trying to reconcile differences with Briffa, et. al. QSR 2000? Link to paper please…”
I was talking about DArrigo et al 2006 – confusion on my part sorry. There was some weirdness in their paper where they plotted Yamal, and called it Polar Urals, and gave it the Polar Urals core counts. Apparently they never received the Yamal data from Briffa though and so that’s why they used POL.
“Was Briffa aware that Steve already had the data? Proof please…”
I don’t know, I wasn’t making any claim just attempting to state the facts. Do you disagree McIntyre contacted Briffa and asked for the Yamal data, and Briffa responded saying he’d pass the request on to the data owners ? I thought that was generally accepted as true.
“Not true. The data was not freely available at this time. Briffa did not publish on this chronology, AFAIK, until 2009.”
I think we are crossing our wires somewhere, or I am confused as to what data was being requested. Wasn’t McIntyre asking for the Yamal data from Briffa 2000 ? He used the same data again in 2008, which was what eventually forced him to make the data public.
“As for the rest of your rather incomprehensible comment…”
Thanks.
“Steve obviously knew enough in 2004 to request the data from the owners. He got it.”
He got the Yamal data that went into Hantemirov, Shiyatov, 2002 and contacted them because he was trying to reconcile differences between that paper and Briffa’s Yamal of 2000. Can you jump to the conclusion that the data is exactly the same ?
Clearly the low late 20th C core count for Yamal was an embarrassment – especially since it was not revealed by Briffa and one would think this kind of caveat on the Yamal data was important. If Steve was sure he had the same data as Briffa he’d have posted about the low core count immediately. Instead he did not do it until Briffa actually released the data through the 2008 journal requirement.
“This is problematic. Not having read the QSR article I don’t know if any caveats were given there. Do you?”
Well I can’t find any:
Click to access briffa.pdf
OK, I looked at the D’Arrigo JGR 2006 article and it appears that in one panel the incorrect core counts were used for the POL series which is stated in the table of data sources to come from Briffa QSR 2000. However, in another panel in the same figure (replication) a rapid dropoff in the core count is shown for POL, so we can only guess that there as a screw up (as Steve states here. The data for POL clearly came from the chronology developed in Briffa, as stated in the paper. It is probably safe to assume that Briffa sent the the data for the Yamal chronology or that they obtained it from Hantemirov (exact source is unclear). Another possiblity for this problem is that Briffa did publish a chronology called “Polar Urals” located at 66′ 30″N 65′ 50″ E in both 1995 and 1996 as seen here. An error in one subpanel of one panel of one figure in a paper.
Steve asked Briffa for the data 2 years after he had already obtained it. He could have asked Briffa if the data was the same, something like “I have this data for a regional chronology that I obtained from Hantemirov in 2004 which matches the site roster for your 2000 QSR article. Is this the same data?”
Instead Steve made a data request which Briffa could not ethically fulfill and he did the correct thing by forwarding the request to Hantemirov (I am making the assumption here that Briffa did indeed do this). Of course I accept this sequence of events as true, Briffa states that it is true. My guess is that Hantemirov got the request and realized that he had already sent the data to McIntyre and the no response was required.
Yes the low core counts post 1990 are a (small) problem. But this is only for the last 10 years of a 2000 year reconstruction! The drop in core count is only visible in the lower panel here which covers that last 200 years (10%) of the recon.
I don’t know what Steve was looking at in Feb. 2004 since that is nearly a year before Climateaudit got it’s start. The first posts I could find on the Hantemirov data were in Feb 2006. He might indeed been trying to reconcile the differences in the variations shown between the authoring groups, but did not seem to understand that the corridor standardization method used by Hantemeriov was inferior to RCS when trying to capture long term variations. His complaint at this time seems to be “it’s different” w/o understanding the difference in standardization methods.
I read the QSR article at the link provided by Marco several days ago. No caveats about core counts in the Yamal series. I remain unconvinced that this is a problem for the series as a whole since it only affects the last 10 (7?) years of a 2000 year series. Cut off the last 10 years and it still looks the same…
“Instead Steve made a data request which Briffa could not ethically fulfill and he did the correct thing by forwarding the request to Hantemirov (I am making the assumption here that Briffa did indeed do this). Of course I accept this sequence of events as true, Briffa states that it is true. My guess is that Hantemirov got the request and realized that he had already sent the data to McIntyre and the no response was required.”
Well Hantemirov said it was ok if I post his response as long as it’s complete and includes my original questions, so i’ll just post them below (and note that Hantemirov does not recall getting a request from Briffa concerning sending data to McIntyre)
“Instead Steve made a data request which Briffa could not ethically fulfill”
Well apparently he did try to fulfill it ethically by forwarding the request to Hantemirov (or someone else who never contacted him ?).
I am very interested in the following quote at http://rabett.blogspot.com/2009/10/some-people-not-me-one-michael-ashley.html
“Steve these data were produced by Swedish and Russian colleagues – will pass on your message to them]
cheers, Keith”
Anyone got any idea where to get the full context ? I had quick search on the net and could not find it, only loads of post citing that snippet.
==== My Email ====
Dear Mr Hantemirov
I’m sorry to bother you out of the blue, I’m just trying to get to the
bottom and truth of an issue concerning Yamal data, McIntyre and
Briffa.
As far as I have been able to piece together, you sent the Yamal data
that Briffa had used to McIntyre 2004 (although it was not within the
context of a request for the Briffa data ?).
According to fragments of an email on the net, in 2006 Briffa replied
to McIntyre’s request for the Yamal data he had used by informing
McIntyre that he had passed the request on to the data owners, of
which I assume you are one of the principles. If this indeed did
happen (that Briffa passed on McIntyres 2006 request to you), why
would you have refused the request if you had already sent it to
McIntyre two years earlier – or did you actually respond to McIntyre
with details of the data which you supplied to Briffa ?
I hope you can help me find out what happened.
Yours sincerely
==== End of My Email
==== Hantemirovs Response ====
Dear XYZ,
in January-February 2004 I was in correspondence with Steve McIntyre
regarding discrepancy of two Yamal temperature reconstructions
published by Briffa, 2000 and Hantemirov, Shiyatov, 2002. It seemed to
me McIntyre was satisfied by my explanation concerning diverse methods
of standardisation used to process the same Yamal data. Nevertheless
he had asked me to send him these data “to verify Briffa’s
calculations”. I sent these data, McIntyre appreciated this.
Henceforth I didn’t receive any queries neither from Steve McIntyre
(except question of the April 2004 concerning Polar Ural dataset that
I forwarded to Stepan Shiyatov because I was not related to this) nor
from Keith Briffa concerning McIntyre.
> (although it was not within the context of a request for the Briffa
> data ?)
It was just within this context
> According to fragments of an email on the net, in 2006 Briffa replied
> to McIntyre’s request for the Yamal data he had used by informing
> McIntyre that he had passed the request on to the data owners
As far as I know today, Briffa had recommend to McIntyre apply to me
himself.
Would you enlighten me on yourself and reason for query? I believed
the scandal has abated.
Best regards
==== End of Response ====
So… in 2004, McIntyre asks Hantemirov for the data used in Briffa 2000 and Hantemirov 2002 to answer his questions about discrepancies between the two papers. Hantemirov provides McIntyre with the data and an explanation for why there was a discrepancy. This wasn’t good enough. McI requested the data from Briffa himself. Briffa responded that the rights to distribute the data resided in Hantemirov and that he should ask Hantemirov for the data. Briffa offered to pass along the request.
So McI had the data and an explanation of the difference in methods. He could duplicate both papers’ methods and had the data.
Seems to me he was more interested in spinning the whole issue of data access rather than doing science.
Explain to me where I am wrong.
As I understand it:
McIntyre emailed Hantemirov asking about the Yamal differences between Briffa’s paper and Hantemirov and Shiyatov. Hantemirov responded with explanation, and McIntyre asked if he could get Hantemirov’s data (presumably as of the 2002 study) which he could run through Briffa’s 2000 study.
“Briffa responded that the rights to distribute the data resided in Hantemirov and that he should ask Hantemirov for the data. Briffa offered to pass along the request.”
According to:
http://rabett.blogspot.com/2009/10/some-people-not-me-one-michael-ashley.html
and countless other sites, Briffa said he WOULD pass along the request.
“Seems to me he was more interested in spinning the whole issue of data access rather than doing science.”
You seem more than happy to speculate over McIntyres motives whilst being incredibly hostile towards others speculating over climate scientists motives.
As McIntyre has repeatedly stated, there are frequently multiple versions of data floating around. For Yamal there were possibly 4 or more (Shyatov 96, Hantemirov 98, Briffa 00, Hantemirov/Shyatov 02) – had McIntyre jumped to the conclusion that the version he got from Hantemirov (in relation to the 02 paper) was the same as Briffa 2000, and that had not been the case, would you be mocking him for that ? Instead, he asked Briffa once, then pushed for access via the journal when Briffa published another study using the data in a journal which required data archiving.
I think Steve had and has a legitimate gripe regarding climate science data archiving proceedures. Clearly they were not used to having their work rigorously checked, or archiving their data and code as a matter of course.
http://www.guardian.co.uk/environment/cif-green/2010/mar/01/phil-jones-commons-emails-inquiry
What kind of review is that ?
Fortunately that has all changed – outside parties are now looking at these climate studies trying to find problems with them. This can only improve the science: to a large degree you have to thank McIntyre for that.
You seem more than happy to speculate over McIntyres motives whilst being incredibly hostile towards others speculating over climate scientists motives.
This is an opinion blog. I don’t claim that it is anything else. I am not a scientist so I can’t claim this blog is about the science. Not that I’m uninterested in the science, but I don’t think the real problem is with the science. The real problem is the politics and economics.
I also clearly indicate that I am interested in people’s motives and interests — stated and unstated — because I approach this from the perspective of a policy analyst. A policy analyst looks at the players in a policy dispute and tries to ferret out people’s interests and goals, values and biases in order to understand their positions on policy issues and what resources they can draw on to influence the outcome of the policy dispute.
If you don’t understand this aspect of a policy dispute, you will never be able to understand what is taking place.
None said: “Fortunately that has all changed – outside parties are now looking at these climate studies trying to find problems with them”.
… and being unable to do so instead default to projection mode and concoct nonsense conspiracies. Remember what the surfacestations project set out to demonstrate? Remember D.O’s laying bare why Steve was hopelessly wrong? “Fortunately” (in your parlance) a whole raft of corporately sponsored echo chambers are there to keep the spirit in which such nonsenses were forged in alive, while conveniently forgetting the results.
None also said:”Clearly they were not used to having their work rigorously checked,”
Certainly not by the likes of number crunching Steve, who approaches fields outside his expertise with the same degree of appreciation and finesse the playing of a delicate piano concerto might be approached by an arthritic, tone-deaf walrus.
None concluded his occluded bout of hero-worship by proposing: “This can only improve the science: to a large degree you have to thank McIntyre for that”.
I expect that sound you hear is the sound of the non-blogscience community chortling. Or barfing.
A couple of comments here.
In 2004 McIntyre discussed the differences (and got the raw data, which he seems to crave) between Briffa 2000 and Hantemeriov’s own chronologies. Hantemeriov explained to him that it had to do with different standardization techniques (corridor standardization vs. RCS). The corridor method has some severe limitations in capturing long term variability since it is limited by segment length (the length of individual cores or, in the case of the Hantemeirov data, chunks of subfossil wood). Yet in 2006 when McIntyre first posted on this he showed no understanding of the source of the differences. It wasn’t the raw data which changed here, it was the analysis method Yet Steve didn’t tell his readers this.
It does appear as though Briffa did not forward the request, although he did tell McIntyre who to get it from (which McIntyre already knew). Touche.
Reviewing is unpaid and giving a good (critical) review is a hard enough job w/o trying to rerun all of the analysis. Steve seems to be complaining, especially in this case that there are multiple versions of the analysis floating around. This is not the same as multiple versions of the raw measurement data, as we have seen here.
Finally, I don’t see how people who don’t know what they are doing criticizing the work of people who do know what they are doing do advances the science at all. There really is a place for expertise in this world .
And BTW Susan, if you could convince wordpress to put and overflow-y:scroll on the new comment container, it would help a ton. It is almost impossible to tell what the heck you are typing when you get to the bottom of the box in FF4.
With regards to the Hockey Team, McIntyre doesn’t do science, he audits. With regards to the NIPCC Team, Christy, et al, he doesn’t even do that. So much for “improving the science”. It’s not as if Jones hasnt asked McIntyre to do the deed, and it’s not as if others haven’t done the deed, but McIntyre just backs away from walking the walk.
Something “sceptics” forget – even before the Climategate manufactroversy, Phil Jones had already been making efforts to have all of the data publicly released (interviewed August 2009). Does he get any thanks from the CA crowd? Of course not.
It’s almost as if the denialist faction is intent on ignoring actual facts in order to construct a mythology, or – dare I say it – a religion.
I would also like to point out that in one of Briffa’s papers (something that Steve made a big todo about, but which I can’t find right now, the post was titled something like “how low can you go” a quote pulled from the paper and used to misrepresent it, IIRC) he analyzed just how few cores were necessary to generate a reasonable RCS chronology. The number was surprisingly low.
“he analyzed just how few cores were necessary to generate a reasonable RCS chronology. The number was surprisingly low.”
Well i’d like to see the link, i’m pretty skeptical though. For a start, how on earth do you know if divergence has set in or not ? Maybe set in on some of them and not others, maybe growth limited by moisture, or cold (but not too cold and plenty of water), or ravaged by an elk ? Nope, professional statisticians (the ones not associated with climate study) seem to find much larger groups of proxies having very little statistical significance in temperature prediction, so I find it very hard to imagine a decent signal can be had from a small number of cores.
Or were you saying you can generate a decent RCS chronology but it won’t have any particular association to temperature ? THAT’s possible (and likely). I assumed we were talking about it in the context of temperature hindcasting.
This is what I was saying. The chronologies presented were just that, chronologies.
Oops,
Should read
More “accurate, detailed and technical commentary” from Steve McIntyre over at the Klimazwiebel:
http://klimazwiebel.blogspot.com/2011/05/i-read-news-today-oh-boy.html?showComment=1307020138499#c8549626564097375284
A recommended read, as the dog whistle is quite audible. Let’s see who picks it up, McIntyre himself can claim plausible deniability again (just as with his “I did not call it fraud”-claims).
Oh, that’s a good one. Watch for this meme to spread…
And GenBank (the gene sequence repository) was started at the Los Alamos National Laboratory. Wait – genetics? Nuclear weapons? New Mexico?
Is it irresponsible to make the connections, or is it irresponsible not to?
So he wants to discuss tangential industry connections that may effect scientific findings? Oh boy!
Remember Stevie has become a part of the fossil fuel industry again…(dog whistle)
Probably more accurate to say “I still think the late 20th C Yamal reconstruction based on the late 20th C Yamal tree rings, is extremely unreliable”.
My understanding is that in his SI Mann said something similar. Along the lines that the data wasn’t as good as other data he had, and he made reconstructions without the Yamal data. These were essentially the same as the reconstructions including Yamal.
If Steve McIntyre was serious about his problems with Yamal, he would quantify them. Steve doesn’t do that, however,as he is just interested in making broad motherhood statements about data quality to cast doubt on the results. Until he actually quantifies his problem he’s just hot air.
My vote for most illuminating example of Steve in action would be the Sudbury airport story that was discussed on a Deep Climate comment thread some time ago. It’s easy to understand. No esoterica like tree rings or red noise, just the kind of Internet blocking software that most people have experienced. Plus it’s much harder to dismiss the explanations from officials of the airport and software provider as transparent excuses from “the Team”. These experts have no stake in climate science.
DC mused about writing a full post on the incident. I’d love to see one.
Steve does, from time to time, show some evidence of the paranoid style…
That is one criticism I may agree on, but he has at least some grounds for it. Plenty of stuff in the “climategate” emails talking about ignoring his requests completely or just fobbing him off.
Yep… by people with a history of trying to deal with such “requests” reasonably, and failing. That’s in the “climategate” emails too, if you look around a bit.
http://assassinationscience.com/climategate/1/FOIA/mail/1226451442.txt
This is the most clear exposure of the McIntyrian tactic. Santer’s a smart man, and he shows exactly what McIntyre is all about. The Douglass et al paper was awful, but McIntyre declined on multiple occasions to discuss it. He did the same with Loehle’s reconstruction, declining to discuss it. He did the same with McShane and Wyner, where Eduardo Zorita found many basic mistakes even after a cursory reading. It is clear that Steve McIntyre does not want to audit climate science. He wants to find even the smallest of smallest potential issue, which he can then blow up.
For background:
Please note the timestamps.
Here is Steve showing some pre-Climategate behavior:
> [Steve] Phil Jones copied me on the following reply to Santer’s email: […]
Again, note the timestamps.
There are no words…
That said, I did like Santer’s email!!!
The email is spot on.
However, the assassination science site triggered my bullshit meter almost immediately. How come anti-scientists always list their qualifications as early as possible, no matter how irrelevant they are to the topic. Then the use of ‘whistle-blower’ made it clear that he was taking sides, and not that of scientists.
Getting data to replicate studies should under no circumstances depend on being friends or friendly with the scientists. It should have been accessible without even having to contact the scientist. In particular any studies used to push public policy should be fundamentally open, downloadable, and as far as possible “click to run”. The idea that you have to grovel/tip-toe around to get data to replicate these papers disgusts me.
Even if McIntyre was wrong, annoying, impolite, and stupid – in terms of getting the data it should be irrelevant.
PS I disagree that in general McIntyre has been unreasonable. The hysterical behaviour by climate scientists started IMMEDIATELY upon his publishing of a response to MBH (which did raise several valid points). He could hardly be at that point described as a major pain to climate scientists since he’d only “responded” to one paper.
Dude, he should have access to the raw data. Let him do his own homework.
Oh, he’s too lazy to do his own homework? Let him rot. I liked Santer’s (and Jones) email. It was right on.
The UK Met Office is a Trading Fund (required to generate revenue to pay for itself, which it does), is an agency for the Ministry of Defence, is owned by the Ministry of Defence who receive dividends from UKMO profits made from the licensing and selling of UKMO product.
The entire concept of the Met Office was born from the Royal Navy’s meteorological efforts from centuries ago. The Met Office even has a unit that operates in combat zones.
I’ve said it at CA and I’ll say it here: If you have a problem with the data being proprietary, thank Tony Blair and lobby your Member of Parliament.
Harassing Jones et al was probably the most useless and pointless exercise ever conceived in the history of the climate debacle. Y’all spammed the wrong departments.
“Harassing Jones et al was probably the most useless and pointless exercise ever conceived in the history of the climate debacle. Y’all spammed the wrong departments.”
Well I wasn’t involved in that (if Y’all is supposed to include me) – though I followed it with interest at the time.
Not sure why you think it was a useless and pointless exercise though – quite a lot was eventually released. Now there is also action on completely open temperature series data and calculations. Yet another improvement largely thanks to McIntyre.
It also wasn’t an effort to harass Jones. CRU had temperature data, some of which was protected by (a very small number of) confidentiality agreements and the vast majority of which was not. With CRU refusing to provide any data at all, based on the small number of confidentiality agreements, the suggestion was made to break the countries down into small groups and submit FOI requests for them individually. The effort was for the data not covered by confidentiality agreements, not to harass Jones.
Actually a totally open access policy, such as the US has, is an exception. Most countries put some data into the GHCN, but withhold other data to sell to commercial services. The effort was not for data not covered by confidentiality agreements (which these days are often online) but for the agreements themselves. The data not covered by agreements is available in GHCN, which BTW, covers about 95% of the data.
95% of the data used by CRU. 100% of the data used by GISS.
And the results are for all intents and purposes identical — even if you randomly throw 95% of the data away. It takes a real effort (euphemism for “cock-up”) to get results that look significantly different.
Yes, the preliminary BEST results based on a random sample of 2% of the records in their database provides results, much to the chagrin of Mr. Watts, which substantially agree with other land only analyses.
None, see the link I provide below in my response to GP. You’ll see a VERY good reason why McIntyre’s requests are being ignored/refused.
Jones also had a good reason, confidentiality agreements, but that’s been widely ignored by ‘the auditors’ also.
My vote would go to this other one underlined by Rob Broberg:
> While I missed it the first time I read the Met’s subset webpage, I don’t think it is fair to characterize the UK Met Office as hiding the provenance of their network. The following quote is listed just above their applet for downloading station data.
The whole inline comment is a fine example of an accurate, detailed and technical commentary:
> Steve: I “missed” that statement the first couple of times that I read the webpage as well. As did other CA readers to date. It’s possible that they added a clarification after the matter was raised here. We’ve seen changes in other official webpages in response to comments here – with the changes being inserted on the fly without notice. I didn’t save the webpage as a I first inspected it. It’s also possible that we both missed the statement. The Google cache at the time of this comment (Dec 28 8 am Eastern) is a Dec 28 cache and sheds no light on whether the page might have been amended. Further Update: Arggh. As Ron points out, there is convincing evidence that the webpage is unchanged – I quoted it myself!!! I haven’t previously handled the RBCN data set and missed the reference. Shame on me.
Not to overshadow Ted’s choice, here is the link to his story:
Small test post to test formatting:
> this is a quoted line
> so’s this
then back to normal
The main reason McI does so well, IMHO, is because he’s the master (or victim) of the narrative fallacy. As EM Forster (via Nassim Taleb) illustrates it, compare:
“The king died, and then the queen died.”
to…
“The king died, and then the queen died of grief.”
The latter is both easier to remember and sell, as practiced daily by the press.
The other reason he does so well is his team of nameless rodeo clowns/picadors who rush in waving very old long dead red herrings any time they notice a discussion his failures.
Too true Hank, this thread is a case in point. Keep the examples coming guys and gals.
There are some more wonderful examples of McIntyre’s shenanigans from when he was nominated as one of the most influential people.
Now we need a good journalist to investigate McIntyre and his team, he will no doubt not appreciate the same scrutiny that he has subjected others to, and he will not emerge looking good. Look what happened to Wegman when people started looking closely.
McIntyre is clearly scrambling to deflect attention away from the hopeless Wegman report and his potential involvement in that. Well, people are not going to fall for it. Actually, the only one “falling” seems to be McIntyre….well at least he has some loyal friends like “none”.
Dispensing with dog-whistes, McIntyre is now just putting it out there:
“In contrast, if someone dares to criticize the Team, adverse reviewers (whose adverse interest is not disclosed to the authors) are permitted to flagellate the submission, as we experienced with Steig’s reviewing of O’Donnell et al 2010. In that case, Steig’s review distorted the process. His primary interest was not the improvement of the O’Donnell submission but in preserving his own product.”
This Tu quoque argument he makes in that thread does not help reestablish the “team” narrative he’s been pushing for years now. Yes, SNA is in tatters, and has no evidence to support it. Interesting how he’s trying really hard to retell the story all over again tho.
Yeah, that one was pretty bad. And, if you look in the comments he still doesn’t like Wahl and Ammann because they showed his complaints to be immaterial. I read WA and it seemed a pretty decent do over of the original analysis, but Steve has an aversion to proxies which show an increase in the mid to late 20th century.
I need to say, that it is pretty hard to engage with a narrow minded autodidact. He is pretty smart but refuses to see the larger picture.
Dare we criticize the Team that we might call it by its name:
http://web.archive.org/web/20050222224714/http://www.climate2003.com/blog/hockey_team.htm
Source: http://climateaudit.org/2005/02/25/a-red-noise-spaghetti-diagram/#comment-32255
The Kyoto Flames might be more accurate, easy to detail, and technically correct.
McIntyre claims that he produces “Accurate, Detailed and Technical Commentary”. Well that statement is an oxymoron.
Precise maybe, but not always accurate.
Technical and detailed are not substitutes for ethics, integrity or being correct.
And he claims not to engage in innuendo or smear?! What did he say about Hansen and his acolytes again? Then there was the ‘crack cocaine’ whopper. On and on it goes.
Briffa told McIntyre he’d contact the authors. I doubt that he did (since the authors had already given him a copy of the data they were using for their paper – though McIntyre had no way of knowing it was the EXACT same version – why would there be no reply back to Briffa or Steve directly since they’d already actually given it ?).
Dumb. Briffa contacts the Russians. The Russians say, hey we already did that and cross it off the list.
And OF COURSE McIntyre had a way of knowing if it was the same set, just run the analysis.
In short grab a cup of coffee and a clue.
From RealClimate.
Assuming DC and others have their facts right, then I’m positive that McIntyre will issue a groveling and prominent correction and apology, because he assured Susan, “If you think that my commentary has been inaccurate, I’d appreciate it if you would direct me to specific errors so that I can make appropriate corrections.”
So either he does that or he is a liar. Ball’s in your court Stevie.
“Deep Climate says:
24 Jun 2011 at 12:04 AM
#31
That’s a mixup between U Penn and PSU no doubt, combined with McIntyre’s rather skewed world view. As it happens, I’ve just posted the following comment at my Open Thread # 10.
=============================
Says Steve McIntyre, commenting on the new sea level reconstruction paper by Kemp et al at PNAS, “Climate Related sea-level variations over the past two millennia”:
A few days ago, PNAS published Kemp et al 2011, a submission by one of Mann’s graduate students. …
It was certainly generous of PNAS to give a “prearranged editor” to a submission by a graduate student at Penn State.
Only problem is, Andrew Kemp received his PhD at the University of Pennsylvania, not Penn State, and his PhD advisors included Ben Horton and three others, but not Michael Mann.
It makes one wonder what else McIntyre got wrong.
http://deepclimate.org/2011/05/20/open-thread-10/#comment-9340“
So far, as of this comment, McIntyre must be aware of the several errors he has made in his attempt to plaster the PNAS as giving preferential treatment to the “team”, and several others as hypocrites, and impugning the work of the scientists who dared work with Micheal Mann. He has crossed out the ‘one of Mann’s’ part and fixed the school, but has yet to acknowledge the mistake in writing, or any other of the mistakes. He has also allowed the “graduate student” to stand. This hasn’t been pointed out yet, so I’ll just suggest it is more poor reporting. Kemp is post-doc.
Doctor of Philosophy (Earth and Environmental Science), September 2004 – May 2009
University of Pennsylvania, Philadelphia, Pennsylvania, USA
Thesis: High resolution studies of late Holocene relative sea-level change (North Carolina, USA)
In other words, better educated than McIntyre.
Thanks gryposaurus, but you are being too generous–McI is a liar (e.g., having certain dendro data and claiming otherwise) and a cheat (e.g., see Nick’s recent post at Moyhu and DC’s posts on McI’s selection of HS shapes). I’m not cutting him any slack on this. If he is going to make such slanderous allegations about Mann and Kemp, he needs to be up front, clear and unequivocal about his errors, and correct *all his errors*. But I know that him being up front and honest is expecting far, far too much from someone like him. But I bet his acolytes are lapping it up….and the latest myth started by McIntyre is probably doing the rounds in the denialosphere.
If McIntyre elects to opine about Kemp, then he had bloody well use “Google” (he can do that much right?) and fact check *first*. Correcting him is not your job, or DC’s or Susan’s….the buck stops with him. He wants to bask in the “glory” that is CimateFraudit, then he owns all the errors and lies on his beloved blog. The day is coming when he is going to get slapped with a libel suit, just like Tim Ball has been.
There was a time when he would have (perhaps) actually looked at the data and done some half-arsed analysis, but no more. Now it is just lies, slander, innuendo and politics.
Holly, yes of course…and that fact bugs the heck out of people like McI.
And a quick correction. In his laziness, and careless attitude toward his fact-less opinion piece, he forgot to change the school name in the last paragraph.
Hi SheWonk,
Another error that I’m sure McIntyre will dutifully acknowledge and correct now that it has been brought to his attention 😉
http://www.realclimate.org/index.php/archives/2011/07/revisiting-historical-ocean-surface-temperatures/
Hello, TPL.
I am interested in your title post and would like to focus on it.
First, I wonder if you would consider adding a link to the original discussion at CA ?
I googled the McIntyre quote and went to the thread there, and found the setting offered a different impression than the post here offers due to the topic and discussion covered.
Following the comment you made …
“The attempt on the part of self-proclaimed skeptics to discredit paleoclimate studies via innuendo and smear and the efforts to defend against that attack is a distraction from what really matters.”
…shows that the response he gave related to “paleo-recons and TPL assessment of the related skeptic efforts. He replied with what he tries to do, what he has done, and what his opinion is based on and what it’s not based on.
However, in your opening post, you seem to go astray, immediately:
You give an example from “the Nursery”, and dissect it.
Could you now give an example that relates to the topic that I would suppose was being discussed
I try to be accurate in my commentary, If you think that my commentary has been inaccurate, I’d appreciate it if you would direct me to specific errors so that I can make appropriate corrections.”
“My criticisms of the proxy reconstructions here have been detailed and technical. My opinion on the lack of worth of the canonical reconstructions is based on my assessment of their technical and statistical defects, not on “innuendo and smear”.
You didn’t do it there, and you have not done it here.
Thanks !
I see the link to the CA discussion now, thank you TPL. .(Permalink button)