01 October 2009

Hockey Stick Gets Personal: Lies from Real Climate

Steve McIntyre must be on to something, judging by the nasty and vituperative comments coming from Real Climate, where Gavin Schmidt levels a serious allegation:
So along comes Steve McIntyre, self-styled slayer of hockey sticks, who declares without any evidence whatsoever that [Keith] Briffa didn’t just reprocess the data from the Russians, but instead supposedly picked through it to give him the signal he wanted. These allegations have been made without any evidence whatsoever.
I have followed this issue closely, and it is clear that Steve McIntyre "declared" no such thing. In fact he declared exactly the opposite:
I don't wish to unintentionally feed views that I don't hold. It is not my belief that Briffa crudely cherry picked. My guess is that the Russians selected a limited number of 200-400 year trees - that's what they say - a number that might well have been appropriate for their purpose and that Briffa inherited their selection - a selection which proved to be far from random and which, as you and I agree, falls vastly short of standards in the field for RCS chronology (as opposed to corridor or spline chronologies).
Gavin's outright lie about McIntyre is an obvious attempt to distract attention from the possibility that Steve may have scored another scalp in the Hockey Stick wars. Rather than distract attention from McIntyre, Gavin's most recent lie simply adds to the list of climate scientists behaving badly. When will these guys learn?

Unlike Gavin, Keith Briffa (who authored the primary work at the center of attention by McIntyre) responds in a much more dignified fashion, suggesting that indeed there may indeed be issues to examine here:

We have not yet had a chance to explore the details of McIntyre's analysis or its implication for temperature reconstruction at Yamal but we have done considerably more analyses exploring chronology production and temperature calibration that have relevance to this issue but they are not yet published. I do not believe that McIntyre's preliminary post provides sufficient evidence to doubt the reality of unusually high summer temperatures in the last decades of the 20th century. We will expand on this initial comment on the McIntyre posting when we have had a chance to review the details of his work.

However the substance of the issue turns out, by lying about what McIntyre said in order to cast aspersions on him, Gavin Schmidt has given his field another self-imposed black eye.

93 comments:

  1. It's quite amazing that RealClimate can't see what they have done to their site. While out-right lies might be rare, the in-line responses by almost every member there very often are less than completely accurate and even more frequently are designed to distract from the real issues raised by a comment. Commenters are censored if they attempt to address the strawmen and falsehoods in the in-line responses.

    RealClimate is in a race to the bottom with tamino, bigcitylib, Rabett and the more disgusting labelers and users of foul language who are denizens of these self-made echo chambers.

    Climate Science cannot be proud of what it has created.

    ReplyDelete
  2. Is it possible you're being suckered in here Roger? Is it possible that Mr mcIntyre is actually the most disingenuous person in the history of disingenuousness? Saying one thing outright, implying the opposite in other comments. For example, what could this possibly mean?

    If you can get a single dendrochronologist to support Briffa's use of 10 trees in 1990, I'll be flabbergasted. They will be astonished and appalled at the procedure. The young dendros will be wincing and some of them will probably be bit shell-shocked at this news. It's very embarrassing for the field. I don't expect any of them to announce their disappointment (we've encountered the silence of the lambs phenomenon before), but make no mistake: no young dendro will stand up for what Briffa did here.

    Hmmmm. Rather than make the accusation of cherry-picking explicit, mcIntyre puts words into the mouths of un-named 'younger' colleagues, simultaneously implying malpractice and deploying a fairly disgustingly ageist sideswipe.

    You endorse this, Roger?

    ReplyDelete
  3. Real Climate was built around the hockey stick. Michael Mann's career, in particular, was built off of the hockey stick.

    I find Mann's methods outrageous. It is possible, a priori, to guess that a paper with his name on it will contain dubious methods and be somewhat confident of being correct (witness Steig et al).

    But there is a long tradition in science of proponents of scientifically discredited method clinging to them long after the larger community has moved on.

    It would be very surprising to see RC do otherwise. As they satirically write in today's piece "apparently everything we've done in our entire careers is a MASSIVE lie".

    They may joke about it, but their careers, and the very meaning of their lives, are on the line here. It would be unreasonable to expect them to treat the hockey stick as some abstract scientific point which can be abandoned in favor of new research.

    If McIntyre prevails, not only will the RC contributors be personally discredited, but they will have dealt their own cause, immediate action on climate change, a serious setback.

    Its hard for me to imagine something worse than dedicating your life to a cause, and then irreparably damaging that cause through your own actions.

    ReplyDelete
  4. In Australia we call this dog-whistle politics. Steven doesn't have to say Briffa cherry-picked the data. It can be left to his audience to infer, and a google blogsearch makes it very clear that this is what his audience did infer:

    "The implication is clear: the dozen were cherry-picked."

    "Briffa cherry-picked about 10 tree-ring cores"

    "Briffa cherry-picked tree rings"

    " Keith Briffa's cherry picked bristlecone pines tree ring data"

    "the dozen were cherry-picked"

    "Briffa's data is probably a cherry picked set of trees to match temp or something"

    "What McIntyre discovered was that Professor Briffa had cherry picked"

    "Briffa, had cherry picked 10 tree data sets"

    "data was cherry-picked and manipulated"

    "based on a "cherry picked" set of data "

    "Briffa's cherry picked Yamal data"

    "All from cherry picked trees undoubtedly."

    "this same set of cherry picked data"

    ReplyDelete
  5. A good time to make a point: you can look at this, and all other such behavior, two ways. First, it is possible that the English-speaking paleoclimate world is full of people doing good work who have serious - if not pathalogical - problems dealing with scientific debate. On the other hand, you can see their behavior when dealing with criticism as saying something about their science. In a somewhat related manner, Samuel Clemens said that when you tell the truth, you don't have to remember anything. In a similar manner, when you trust your science, you don't have to fear critical examination of your work. The fact that the paleo crowd clearly can't deal with a normal - if public - examination of their work pushes me to assume the worst.

    ReplyDelete
  6. -2-Phil Clarke

    I don't see anything in your excerpt other than a complaint about methods

    -4-Andrewt

    I don't hold Steve McIntyre responsible for what others have said, when he has clearly said the opposite.

    ReplyDelete
  7. Phil Clarke,

    Steve isn't suckering anybody in your quoted passage. He is being perfectly clear. He had said (multiple times) that 10 cores is not a sufficient basis on which to draw any conclusions.

    He has repeatedly and explicitly dared anybody to find a dendrochronologist who is willing to say otherwise.

    There is no disingenuousness here. There is no subtlety. He says that Briffa drew an unsupportable conclusion, and then hid the evidence for the better part of a decade until the Royal Society forced him to disclose it. He has now dared Briffa (or anyone else) to produce a single dendro who will support his procedures.

    If Briffa has a leg to stand on, this will hardly be difficult. There are a great many people in the dendro community who are unhappy with Steve and the light that he has shone on their profession. As documented on his website, they have followed past CA discussions, and ranted about him on the dendro listservs.

    If is very safe to bet that if any significant portion of the dendro community believes that Briffa's procedures are scientifically supportable, Briffa will receive their support.

    Perhaps on other issues (cherry picking) Steve is being careful with his words. He should be. He has no way of knowing the process behind Briffa's proxy selection because it was never documented.

    But with respect to the use of 10 trees, he has very appropriately called Briffa out. It is up to Briffa to answer.

    ReplyDelete
  8. Roget you are missing the point of dog-whistle politics. You say one thing, but your choice of words means your target audience infers something quite different. Steven McIntyre says:

    "and, to an outside observer, seem to have done no more than pick the version (Yamal) that suits their bias"

    "In my opinion, the uniformly high age of the CRU12 relative to the Schweingruber population is suggestive of selection"

    "It is highly possible and even probable that the CRU selection is derived from a prior selection of old trees"

    "I do not believe that they constitute a complete population of recent cores. As a result, I believe that the archive is suspect.",

    but the message his audience gets is something like this:

    "What McIntyre discovered was that Professor Briffa had cherry picked “tree data sets” in order to reach the conclusion he wanted to reach."

    ReplyDelete
  9. Andrewt,

    Imagine somebody who wants to audit a bank for racial favoritism. For the better part of a decade he attempts to get data from the bank. Finally, a federal regulator discloses a limited slice of information about that bank.

    In an area which is 50% white, 10 out of 100 applicants in a particular range of incomes received loans. All were white. The bank continues to refuse to disclose many critical pieces of information. For example, how many of the 100 applicants were white? Were there any other factors other than race which also correlated strongly with the distribution of approved applicants? He doesn't know.

    So, the auditor publishes his report, stating what he knows and what he doesn't. Immediately there are cries of racism. The auditor knows that racism is a serious offense, so he explicitly refuses to make such a claim, and even offers an alternative explanation.

    Who is wrong here?

    In my opinion the auditor is entitled (especially after waiting so many years for data) to make his results public. He is being very responsible by very carefully claiming only those things that he can personally vouch for.

    The public is also entitled to claim racism. All things being equal, it is incredibly improbable that all the successful applicants would be white.

    If all things aren't equal, it is the bank's repeated refusal to disclose, that has prevented this information from becoming public.

    It is perfectly reasonable to conclude that if the bank weren't doing something wrong, it wouldn't have tried to avoid disclosure for such a long period of time. Even if the auditor refuses to endorse such a claim. (After all, how could the auditor know?)

    In my opinion it is the bank and its refusal to disclose the data that should be held at fault. Absent evidence to the contrary, I would assume that the bank is racist.

    Until Briffa and his Russian colleagues archive most or all of the cores that they have collected near Yamal, and that data is shown to be free of cherry picking, I will assume that cherry picking occurred.

    I do not accuse Briffa of lining up 300 cores, and choosing 10 or 12 which had the greatest hockey shaped stick. I doubt very much that he did this.

    But he did do the following:

    1. He presided over a process in which he received access to many different cores from this area grouped in many different ways.

    2. He chose a very small set of cores, that shows a far stronger hockey stick shape than any other presently know group from the same area.

    3. He deliberately withheld this data, despite publishing multiple papers based on it in journals requiring disclosure.

    It is now incumbent on Briffa to show that no cherry picking occured. Accepting a set of 12 cores from the Russians and not altering it after receipt is not longer (thanks to Briffa's long cover up) sufficient to absolve him of the charge of cherry picking.

    The fault lies not with Steve who has been very careful with his words, nor with the CA faithful who honestly call them as they see them. The fault lies with Briffa who has created at least the appearance of a coverup.

    ReplyDelete
  10. Jason S confirms my point - he hears Steven McIntyre's dog whistle ("the charge of cherry picking").

    ReplyDelete
  11. Andrewt,

    My post makes another point:

    Many of those who are excoriating Briffa for covering up his dishonest practices ARE CLEARLY AWARE THAT THAT MCINTYRE HAS MADE NO SUCH ACCUSATION.

    ReplyDelete
  12. Andrewt,

    Do you have evidence that Briffa didn't cherry pick? Because as I read it, he still hasn't released the entire data set so that anyone else can confirm or deny.

    Also, the 10 (or 5, or 12, depending on the time frame) trees were picked because they correlate with another Russian study between 1800 and 1900- yet that other Russian study doesn't show the same warming pattern after 1900 as Briffa? Explain that- if you can.

    Or are you just here to cast aspersions instead of solve anything?

    ReplyDelete
  13. What Schmidt and by extention RealClimate have done is reinforce an impression of a small ideologically motivated clique who get too caught up in their "cause" to stop and check data thoroughly and objectively.

    Like Briffa and his tree rings, or Mann and his pinciple components, Schmidt should have had no prior about what was or was not true and objectively sift ALL available data. In this case. Steve's actual words.

    Worse. This is a straw man.

    I mentioned in an earlier comment that the technical points are really a sideshow to the most important discussion here:

    Why is this news nearly a decade after the publication of this paper? (because we only just got to see the data)

    Why have we only just got to see the data? (the authors refused to archive it and esteemed Journals refused to enforce their own data policies)

    ReplyDelete
  14. Re the accusations to Steve McIntyre about the cherry picking of Briffa, this is what he writes on his blog:

    "Steve: if someone read my head posts and comments, I don't see how there could be any misunderstanding as to what is meant here. First of all, this does not in any way imply that Briffa made the picks, which is what people are accusing me of alleging. In other comments, I had clearly said that it was my surmise that the Russians had made the picks and that Briffa had applied the subset for a different standardization method (one requiring more replication) than that used by the Russians. I noted that the Russians said that they had picked trees that were older in the 20th century than the corresponding subfossil trees. They also said that they selected cores for "sensitivity", which might also have introduced a bias. Because there was such a difference to the Schweingruber population, it was my surmise that this selection procedure, however it was done, prima facie introduced a "bias" - which is a term of art in statistics and not derogatory. I observed that this surmise was rebuttable and was and remain interested in alternative explanations. I don't see that this language in any way supports an allegation that I accused Briffa of cherrypicking cores to match the instrumental record."

    Again Steve's opponents use straw men arguments and red herrings in order to avoid adressing the REAL problems, and this is not the first time.

    ReplyDelete
  15. Wow. Just wow (in reading these comments).

    ReplyDelete
  16. Briffa's response certainly was gracious given the innuendoes Mr. McIntyre has been content to spread. However, in an email from T. Osborne alerting me to Briffa's statement, it was noted that Briffa's 1st response would "deal with the allegation of deliberate data selection/exclusion".

    It would be astoundingly naive to think that the natural reading of McIntyre is anything other than suggesting that Briffa rigged the data.

    ReplyDelete
  17. I actually believe that Briffa did not purposely cherry pick these ten trees for the results he wished to see. What I believe this displays is an astoundingly bad data quality review by Briffa et al to identify and eliminate/minimize alarmingly biased data.

    What he/they are attempting to cover here is not willing deception and manipulation of the data, but a alarming incompetent treatment of there own data and an overwhelming study bias due to their own preconceived notions.

    Everything they post displays their position bias going into each one of their studys, Briffa's own words ...

    "I do not believe that McIntyre's preliminary post provides sufficient evidence to doubt the "reality" of unusually high summer temperatures in the last decades of the 20th century. "

    Appropriately, this man should NEVER use the word reality when discussing the object of his study. The word he should be using here is evidence.

    In my view they aren't attempting to cover up their lying, I don't think they ever were willfully lying. They are trying to cover up their overwhelming incompetence which when it comes to light will destroy their careers. I kinda feel bad for them.

    ReplyDelete
  18. -16-bigcitylib

    You are welcome to post here, but you'll surely understand why I don't see any point in engaging with you:

    http://rogerpielkejr.blogspot.com/2009/08/pathologies-in-climate-science-there.html?showComment=1250817290852#c6259208145354060840

    ReplyDelete
  19. "Briffa's response certainly was gracious given the innuendoes Mr. McIntyre has been content to spread."

    Did you actually read the comments of McIntyre in my above post or just did looked at it? Can you please specify the inuendo you mention? Or is it just punching holes in the air, as usual?

    ReplyDelete
  20. I don’t understand RC’s reliance on straw men and ad hominem arguments. After reading three straight similar false arguments I’m not open to any logical argument that may (or may not) follow.

    The good old appeal to authority argument has been worn down to a nub. They need to mix it up some with an occasional red herring or missing middle argument. Perhaps someone at RC can check out http://home.att.net/~tangents/issue/think/fallacy.htm for fresh ideas.

    Of course there are always the options of facts and logic.

    ReplyDelete
  21. Perhaps Briffa misspoke in his response to McIntyre, where he starts a paragraph with this suspect little sentence:

    "My colleagues and I are working to develop methods that are capable of expressing robust evidence of climate changes using tree-ring data."

    Maybe he should watch for wording like that in the future -- though he has been respectful in his response. You can find it below:

    http://wattsupwiththat.com/2009/10/01/response-from-briffa-on-the-yamal-tree-ring-affair-plus-rebuttal/

    ReplyDelete
  22. Who cares who said what? McI did a perfectly valid sensitivity test which clearly shows that the reconstruction is not robust. The thrust of Gavin's post is that Yamal doesn't matter even if the data is wrong and in support he plots up a series of other hockey sticks to prove his point:

    One of those is the CO2 hockey-stick which is irrelevant because the CO2 sensitivity to temperature is still very much a guess.

    Another is the MBH stick which he said Amman had verified (he hadn't) which is also irrelevant because it has not only been busted by McIntyre but also superceded by Mann himself.

    Two plots extended back only to 1600 so are irrelevant. The extended borehole plot actually shows a healthy MWP.

    Another plot is from Briffa himself - hardly convincing when Briffa's poor methodology is the issue.

    Then Gavin shows the instrumental record which we can't properly review because that data is also withheld by Briffa's colleague Jones. Of course it's also irrelevant because it tells us nothing about the past.

    Finally he gets to Mann's newest plot - the only relevant plot - which McIntyre has also ripped apart in several in-depth analyses, the jist of which is that if you leave out tree rings and use the Finnish lake data the correct way up rather than upside-down then there is no hockey-stick to be seen there either. Manns answer to McI's published comment amounted to a mere high-handed dismissal.

    As for Briffa, it's quite obvious that his data with-holding tactics and data pre-selection methodology are both utterly appalling. He defends himself by saying that the reality of latter day warming is unaffected. But the issue is whether it is now unusually warm. Briffa actually has no answer to that despite 10 years of pretending he did. He should darn well apologize for wasting everyones time and money.

    The only argument supporting Briffa seems to be that we should trust him because - well just because he's a scientist and we should trust scientists. On the contrary this episode demonstrates exactly why we shouldn't trust anyone in this climate debate because confirmation bias is utterly rife!

    All we really need to do is to analyse if these many gross assumptions are backed by the actual data - no more, no less. For that we need the data, preferably with and without adjustment and before and after any pre-selections and preferably properly checked by a 3rd party this time. Such 3rd party checking is actually mandatory elsewhere for important policy decisions and rightly so. If it turns out that there is no hockey-stick then it is actually good news. Is it the job of scientists to suppress good news?

    ReplyDelete
  23. Joel Jarrard,

    Actually, Briffa appears to have secured funding specifically to develop more robust methods for using tree ring data to analyze climate.

    He meant what he said.

    The timing of this incident is most unfortunate for him.

    ReplyDelete
  24. Jason S

    I'm actually pulling for him to develop more effective methods, but he doesn't really say that. He says his team's intentions are to develop methods that can "produce more robust evidence that describes climate changes..." Doesn't sound too scientific to me. I guess he doesn't even have to attempt to sound like he's remotely objective anymore. Shouldn't he be attempting to develop methods that accurately describe the state of the earth's climate?

    It sounds like his aims are to produce scary looking graphs for RC's PR agency to send out. Maybe that's not what he intended to say, and maybe it's just semantics, but it just doesn't sound good.

    ReplyDelete
  25. How can you develop a "robust" tree ring methodology when you've got a sample size of 10, where the signal is noisey, not related to local temperatures, and theres one of the ten showing a response four times larger than the rest ?

    However he got the "sample", it was indeed cherry picked at one point (whether before he got his hands on it is irrelevent), is extremely non-robust and was hidden for 9 years. What a poster child for climate science. Briffa is one of the big boys, in charge of IPCC chapter editing - absolutely shocking.

    ReplyDelete
  26. I don't know if it's been mentioned before on Roger's blog but realclimate.org is owned by Environmental Media Services, a Washington DC public relations firm which is owned by Fenton Communications. Fenton is the largest "progressive" lobbying firm in DC with deep ties to Gore. To put it bluntly, RealClimate.org is a propaganda outlet wrapped with a science veneer. They are not interested in a science debate which is why they censor inconvenient comments. I wonder how much EMS pays Gavin and company for their propaganda work?
    Anyone doubting this can do a Who is? domain search to confirm.

    ReplyDelete
  27. As Steve McIntyre showed in plotting the actual results for the 10 trees, the variance in this sample is so large given YAD06 that one must increase the sample size in order to deal with this potential outlier. This is elementary data analysis.

    ReplyDelete
  28. xen that kind of stuff is totally irrelevent - I think you have the cart before the horse.

    The scientists genuinely believe their positions are supported by the evidence - an extreme case of confirmation bias if you will. The site is "supported" by that organisation merely because that kind of organisation seeks out "problems for people requiring governmental intervention" (at least in their opinion) kind of issues and, if it was true, AGW really would.

    RealClimate is merely guilty of group think and confirmation bias. The message of the scientists is from the heart and genuinely held (although completely wrong, imo).

    ReplyDelete
  29. Phil Clarke-perhaps you don't understand that CA has a long history in which a number of in-jokes have arisen. In fact, young dendros refers to something rather specific and not what you mean to imply:

    http://www.climateaudit.org/?p=2506

    ReplyDelete
  30. Andrewt- this quote:

    ""In my opinion, the uniformly high age of the CRU12 relative to the Schweingruber population is suggestive of selection""

    It is shameful that you would repeat the out of context use of these words by RC. Do you even know what Steve actually said? "“In my opinion, the uniformly high age of the CRU12 relative to the Schweingruber population is suggestive of selection – in this respect, perhaps and even probably by the Russians”" Emphasis added.

    Now who is disingenuous?

    ReplyDelete
  31. "They may joke about it, but their careers, and the very meaning of their lives, are on the line here."

    This week's entry into Crazy Ranting Internet Guy's Massive Hyperbolic Outgushing.

    ReplyDelete
  32. Tree rings, tree rings, lots and lots of tree rings,
    So many tree rings, all across the globe...

    Briffa, Briffa, playing with your data,
    that Briffa's such a 'player' in his statistician's robe...

    Hey Briffa, Briffa, share your data, could ya?
    Share your data NOW, not 10 years down the road...

    Oh briefly, could ya, share your reasons, won't ya?
    For tossing out the trees that didn't match your goal...

    Oh Briffa, twister, tweakin' all the data,
    Making all errata this Global Warming Ploy,
    As your castles all fall down around you,
    Your premise looks absolutely UNtrue,
    Unprecedented warming used to sound so very charming,
    But your shattered hockey stick could really use some glue...

    Oh well, Briffa, as you exit,
    just remember these words from a skeptic
    Trying hard to forgive you
    While murmuring a groan,
    Don't let the back door hit ya
    Where the good lord split ya,
    And next time please do SCIENCE -
    Leave politics alone.

    ReplyDelete
  33. Andrew the extended quote makes my point even better. Steven McIntyre words are such that Roger can try to claim SM didn't actually accuse Briffa of cherry-picking, but when SM says:

    "In my opinion, the uniformly high age of the CRU12 relative to the Schweingruber population is suggestive of selection – in this respect, perhaps and even probably by the Russians”

    Google's blogsearch tell us what Steven McIntyre's audience actually hears:

    "Yamal tree fraud by Briffa et al."

    "Briffa and his team could select the data that matched their beliefs, rejecting the data that did not produce a Hockey Stick shape"

    "a paper that manipulated the data in several places and actually fabricated the hockey stick upswing"

    "Briffa, had cherry picked 10 tree data sets"

    "Briffa massaged his data to get a desired result"

    "Briffa only selected 12 cores that were “cherry picked” to show what he wanted"

    "Briffa cherry-picked tree rings that supported a hockey-stick shaped temperature curve"

    "12 cherry picked tree ring measurements were used to paint the fraudulent picture with"

    "a second cardinal sin is uncovered. That of cherry-picking data. In the cross-hairs is Keith Briffa"

    "Briffa, had cherry picked 10 trees data sets out of a much larger set of trees sampled in Yamal"

    "And it is a scandal. Briffa cherry-picked about 10 tree-ring cores"

    "Briffa has gone off sick. Wouldn't you be fealing sick if you had to reveal your pseudo-scientific cherry-picking methods for all to see"

    ReplyDelete
  34. Andrewt, perhaps you might comment about the worth of a scientific paper that does not discuss issues of data selection (rejection and inclusion) and of sample size/sub sample size) when these are critical to the outcomes of the study. Perhaps you might want to also comment about the peer review process that allows such a paper to be published.

    Shouldn't this, and the lack of data diclosure and archiving be the real topics of discussion?

    They are substantive matters of fact, rather than your guess at the hidden motivations of the words people are using in this discussion.

    ReplyDelete
  35. andrewt,

    You complain about these comments as if nothing had gone wrong here.

    Something has, in fact, gone terribly wrong.

    Irrespective of how the data was chosen there should be no doubt about the following:

    1. Briffa DID use just 10 cores to establish the modern portion of his hockey stick.

    2. Those 10 cores DO behave very differently from other several other groups of cores collected in or near the same location by Briffa's collaborators.

    3. Those 10 cores do NOT behave in a manner that can be validly examined using the methods Briffa used. Specifically, he assumed a roughly normal distribution in the absence temperature change, and a roughly linear temperature response (post age adjustment).

    4. Briffa DID avoid disclosing his data in the face of numerous request, ultimately disclosing it after a year of requests by the Royal Society.


    Even if it turns out that all of these defects were due to some sort of comedy of errors, and no malicious intent was involved, a very serious series of errors was committed.

    Briffa had a responsibility to get this right, and for whatever reason, he did not. Some hyperbole is in order.

    ReplyDelete
  36. Jason S - I think there are 12 cores not 10.

    But of those 12 cores, just one core accounted for most of the current warming, ("The Blade of Briffa")

    When the graph was drawn without the 12 cores using the rest of the Schweingruber cores, the "blade" turns downwards instead of upwards.

    When all the cores are used the current warm period is displayed but with the medieval warm period now showing up as warmer that the current. (This in a nutshell)

    This Briffa Yamal chronology was used in constructing 10 of the 12 graphs in the IPCC's AR4 spaghetti temperature graph on the basis of which the IPCC declared that the warming of the last 50 years has been unprecedented in at least the past 1,300 years.

    Apparently this is still important as apparently Briffa has been awarded another study worth £231,441, because among other things, some have challenged this "current scientific consensus, represented in the 2007 report of the IPCC"

    http://gotw.nerc.ac.uk/list_full.asp?pcode=NE%2FG018863%2F1&classtype=

    ReplyDelete
  37. Cartoonasaurus!

    I was singing along!
    You rock!
    http://www.youtube.com/watch?v=JKDtUzRIG6I

    ReplyDelete
  38. Incidentally here is a very simple explanation for the tree that caused the most warming "The most influential tree in the world". It is by a humble woodcutter who spent 50 years cutting trees, living among them and studying them.

    It has nothing to do with Global warming, anthropogenic or not.

    Just a tree that has spent its childhood in the shade of its elders followed by its “day in the sun.”

    The story is here: http://wattsupwiththat.com/2009/10/02/a-hands-on-view-of-tree-growth-and-tree-rings-one-explanation-for-briffas-yad061-lone-tree-core/

    ReplyDelete
  39. "Gavin's outright lie"?
    Climate Audit featured Andrew Orlowski's Register report in a special post. And Orlowski says:

    "The implication is clear - the dozen were cherry-picked"


    If this is an outright lie, why is that report featured on CA?

    ReplyDelete
  40. An interesting comparison. Anthony Watt takes detrended data and make a great deal of its lack of trend ("flat as roadkill") and how different it is to Briffa's results.

    Steve McIntyre corrects him but using language such that none of the 100+ other commentors even seem to understand Anthony Watt has made an error, let alone such a basic & misleading error and
    no one says anything uncomplimentary about Anthony Watts, certainly nothing like:

    "potentially criminal scam"

    "criminal charges for fraud"

    "he should be sacked summarily - and then prosecuted for fraud"

    "thrown in jail for an attempt to take over the world"

    "Let's hope we can look forward to the guilty parties being charged over this matter before very long."

    "Do any of the lawyers here know the grounds for bringing on criminal negligence charges?"

    ReplyDelete
  41. Well, there's been a lot of emotive comment and exaggeration of fact with 'the message' taking precedence. If people chose to author rants because a small but substantive issue has been identified, most of the people reading those articles will read them for what they are, and the science machine will filter out the chaff.
    It seems to me that the only way to resolve this, rather than saying that it does not matter, is to accept that today's data is inconclusive. It does not need more analysis, it needs agreement on how much new data is required, what the collection criteria should be, and sample exclusion rules defined in advance. The data can then be opened to anyone who wishes to create a reconstruction, without prevarication.
    I don't know how much it costs to take a helicopter out to Yamal for a week, but I'd be happy to be be part of the team and fund my own place.
    Who can define the experiment?

    ReplyDelete
  42. Andrew (8) already assembled quite a list showing some of McIntyre’s insinuations, and an explanation of his tactics (dog-whistle politics): Making the insinuation subtle enough to be able to defend himself against accusations later, and clear enough so that the message gets through to his supporters. Which it does: For miles over the internet, people refer to McI as the source of having uncovered this ‘scandal’.

    Some additions, from RC:

    “Because the selection yields such different results from a nearby population sample, there is a compelling prima facie argument that they've made biased picks.” (http://www.climateaudit.org/?p=7168#comment-357407)

    [Response: Fair enough, so here goes (a couple of allied quotes as well): 1) "In my opinion, the uniformly high age of the CRU12 relative to the Schweingruber population is suggestive of selection", 2) "It is highly possible and even probable that the CRU selection is derived from a prior selection of old trees", 3) "I do not believe that they constitute a complete population of recent cores. As a result, I believe that the archive is suspect.",4) (Ross McKitrick) "But it appears that they weren't randomly selected.", 5) (Anthony Watts) "appears to have been the result of hand selected trees", - gavin]

    “If you can get a single dendrochronologist to support Briffa’s use of 10 trees in 1990, I’ll be flabbergasted. They will be astonished and appalled at the procedure. The young dendros will be wincing and some of them will probably be bit shell-shocked at this news. It’s very embarrassing for the field. I don’t expect any of them to announce their disappointment, but make no mistake: no young dendro will stand up for what Briffa did here”

    I think Roger’s allegation to the address of Gavin is lot farther removed from reality than Gavin’s allegation to the address of McIntyre.

    Bart

    ReplyDelete
  43. The most revealing fact in this entire fiasco is the fact that neither Gavin nor Briffa nor the whole chorus of Apologists Extraordinaire for All Things Global Warming has made any substantive response to the sensitivity issue raised by Steve McIntyre's analysis.

    Instead, the response has been the obtuse and evasive reply of Briffa -- who said nothing to defend his study's reliance on what really amounts to 1 (!) tree -- dripping sophomoric sarcasm from Gavin (an utterly pathetic substitute for actual argument) --and the "straw man" tactic of quoting Steve out of context to make it appear he said something he did not.

    If there is an actual, scientific defense of what Briffa did, I would sure like to see it. Otherwise all this howling and quoting of headlines that Steve did NOT write or say is completely irrelevant to those interested in hearing the truth.

    ReplyDelete
  44. Richard, its a set of 12 cores of which only 10 cover the last decade.

    ReplyDelete
  45. Nick Stokes,

    I think we've already established that:

    1. Real Climate contains outright lies and

    2. Real Climate has previously been featured on CA

    I would therefore propose that the Register article was featured for the same reason that CA and many other websites link to Real Climate:

    Entertainment Value


    I'm happy I could solve this mystery for you.

    ReplyDelete
  46. Sean Houlihane,

    There have been numerous repetitions of the Tree Ring studies contained in the hockey sticks.

    In all that I am aware of, the studies failed to replicate the original results.

    Of particular note, the Bristlecone studied which dominates the early hockey sticks produced by the Team (without that single study, there would be no hockey stick) was repeated by Ababneh, and found to have no hockey stick.

    More recently Steve actually went out there and took some cores himself, successfully identifying some of the strip bark trees used in the original study. He found that these trees are so mishapen, that the results you get depend on the angle from which you take the core, and often vary spectacularly within the same tree.

    Unfortunately, neither of these results were sufficient to convince the Team to repudiate its earlier work, and they have simply "moved on"

    ReplyDelete
  47. andrewt,

    "Steve McIntyre corrects him but using language such that none of the 100+ other commentors even seem to understand Anthony Watt has made an error, let alone such a basic & misleading error and
    no one says anything uncomplimentary about Anthony Watts"

    The reason for this should be obvious. Anthony Watts tries (and does a good job of) publishing details about his methods and procedures. He deliberately leaves the details out there so that others can identify errors in his work. A person who was deliberately committing scientific misconduct would attempt to hide their data and prevent anybody from discovering their errors, so Anthony Watts is assumed to be honest.

    Briffa did everything possible in his power to hide his data other than not publish.

    People who hide things appear guilty. People who are open appear innocent.

    ReplyDelete
  48. So let me get this straight, because people who read what Steve writes may misconstrue it, he shouldn't be allowed to say anything.

    It's rather remarkable that Steve is considered guilty of saying anything readers do...so, he's supposed to, what, police the words of all readers? The shear volume of quotes you've uncovered actually proves my point. You just don't want Steve doing anything but having to deal with silencing any accusations by others that arise from his work.

    ReplyDelete
  49. Its all just holier than thou nonsense from the climate alarmists.

    Briffa did cherry pick the sample. Probably not by himself hand picking the 12 cores but rather by taking the small sample and deliberate not padding it out with more data from the area (as he had done with previous small sample sizes). On top of that, he did not allow anyone to know that the study was so dominated by such a small sample, nor state any caveat that his results should not be relied on as being statistically significant.

    So whether or not people are incorrectly claiming he actually picked the cores himself to prove the point is a side issue to the core (hah) issue that his results are not statistically significant, are not robust to the inclusion of other data, are critically dependent on the core of one tree, and all these issues have been hidden for the past 7 years. Yes Briffa has a credibility issue now, whether or not he actually hand picked those cores or took them on a plate from the Russians.

    For all we know the Russians just took the group of living cores as a subsample, just to see what it looked like, and passed it along with all the rest to Briffa, who tested with it and liked the results. Still casts Briffa in a bad light.

    ReplyDelete
  50. In my view, the current dispute is not about science but about public relations. It is interesting to note that RC recently had a thread on the need for better communications (i.e., PR ) of the scientific basis for catastrophic AGW – which ties neatly into someone’s earlier comment about the funding of RC by Fenton Communications. Many proponents of AGW are highly sensitive to any negative stories - and frequently accuse others of astro-turfing - which I take to mean the deliberate spreading of misinformation - hence the slightly paranoid belief that Steve McIntyre has launched a deliberate campaign to discredit Briffa, AGW and its associated scientists.
    There is little that actually can be said to dissuade pro-catastrophic AGWers from this viewpoint – they are way too invested in their position and, besides, nobody likes auditors. Somewhat paradoxically, the most likely source of dissuasion is Briffa – since he knows what he did, why he did it and the scientific implications of the issues raised by Steve McIntyre. Gavin and others know exactly the importance of the Yamal proxy to paleoclimatology and dendrochronologies in particular. It goes beyond the actual empirical importance of this proxy to other proxy-based studies. It links back to the Mannian statistical errors, the ongoing non-disclosure of data for other proxies and over exposure of the flawed hockey-stick icon. They also know that Steve McIntyre is likely to be correct in his basic facts and statistics – even when they disagree with his conclusions from those facts. McIntyre after all does not offer an alternative scientific theory; his audits, like all audits, examine the essential competence and basic scientific integrity of the scientists involved.
    The paleoclimatological field is sufficiently novel and fraught with inherent technical and data complexities that believers see the patterns that support their beliefs, while non-believers can readily find holes and inconsistencies. The unfortunate reality is that current theories of catastrophic AGW depend upon the current warming being unprecedented, hence the need to go back beyond the instrumental record. The potential value of paleoclimatology to this effort is clear. Unfortunately the state of the science and the policy implications demand maximum transparency - which brings with it the risks of negative results (e.g., the divergence problem) and missteps (e.g., Briffa's insufficient sample size).
    Which goes back to Steve McIntyre’s unabashed core mission: ensuring the transparency and replicability of climate science.

    ReplyDelete
  51. Rp - I don't see anything in your excerpt other than a complaint about methods

    Well that's very interesting. Others might well see no examination of methods at all in the quote, might even see McIntyre using the rhetorical device of putting words into others' mouths to get his point across, and of stating how an entire discipline will react to his 'revelations'.

    It is hard to interpret his phrase 'what Briffa has done' as meaning anything other than using a too-small unrepresentative sample that produced a HS, which is another way of saying cherry-picking. You might want to count the number of times the phrase 'cherry-picking' come up in the thread versus the number of times the blog owner objects to the phrase being deployed.

    On the question of whether McIntyre made accusations of selective data mining, here is what he said ...

    it is my opinion that there is considerable evidence that the 12 cores are not a complete population i.e. that they have been picked form a larger population.

    I do not believe that they constitute a complete population of recent cores. As a result, I believe that the archive is suspect.

    And, not directed specifically at Briffa, but unspecified 'people in the field' (but given McI's constant refrain that climate science is dominated by a small clique this must include 'Team' member Briffa)...

    Unfortunately, to date, people in the field have not honored this responsibility and, to an outside observer, seem to have done no more than pick the version (Yamal) that suits their bias.

    It is stated elliptically ('an outside observer might believe') but the accusation of data mining IS stated. To pretend otherwise is to set a new record for naivite.

    ReplyDelete
  52. After having seen the sensitivity tests by Tom P on climate audit, I withdraw my incorrect claim that the late 20thC 10 core sample was critically dependent on the one freak tree - it was not (merely added what looks like 10% to the signal).

    This assumes that Tom P's sensitivity test was in accordance with Briffa's procedure...

    ReplyDelete
  53. Xinghua,
    RealClimate may have "featured" on CA, but quite different to the way they featured the Register report. The title of that post was "The Register picks up the Yamal story", and the post "savoured" the conclusion. It quoted the Register's title "Treemometers: a new scientific scandal". True, a mild demurral was (much) later inserted by Steve. But the explicit inference drawn by Orlowski was exactly the same one that, when drawn by Gavin at RC, becomes a "serious allegation" and an "outright lie", according to RP.

    That's the point of andrewt's list above. All these SM supporters are drawing the same inference that Gavin did. But no-one was then calling those inferences "outright lies". In the case of the Register, CA was savouring it.

    ReplyDelete
  54. Phil Clarke,

    Using too small a sample is NOT the same as cherry picking. It simply means that the number of samples used by Briffa is too small for his result (or the numerous other results that depended upon his) to be valid in its present form.

    Accusing Briffa of cherry picking specifically means that Briffa had access to multiple sets of data, and he chose the set of data that best suited his goals.

    There is a link between the two: Why did Briffa choose an obviously defective sample? (Not only are there too few cores, but the chosen cores do not conform to the assumptions of his chosen analysis).

    One possible answer is that Briffa was deliberately cherry picking; that he was aware of the existence of multiple data sets, and chose the one that suited him best (ignoring questions of suitability).

    Another possibility is accidental cherry picking. Briffa wanted to show a particular result, and accidentally used methods that allowed him to express this bias in his selection of cores.

    But there are numerous possibilities that have nothing to do with Cherry picking. For instance,
    Briffa may not have understand his own methods of analysis or never checked the data sets he was using to verify that his analysis was valid. Since Briffa has recently received money specifically to address the non-robustness of some methods of analysis, he may even have subsequently become aware of his errors and be looking to remedy them.

    I am sure that there are other possible explanations as well.

    For at least the moment (and quite possibly forever) we don't know whether Briffa cherry picked (consciously or otherwise) or how he came to select his data set.

    All we know with reasonable certainty is that he chose a data set which is not scientifically valid for the analysis which was actually performed.

    Specifically:

    1. There were too few cores

    2. The cores do not have anything remotely close to Gaussian noise.

    3. The cores do not respond linearly to local temperatures.

    There may also be other defects that are not yet understood (certainly if Briffa did cherry pick then the data set also suffers from selection bias), but for now we don't know.

    ReplyDelete
  55. db:
    The charts that Steve McIntyre presented for 10 trees that produce the blade of the hockey stick clearly show the same upward trend. YAD06 is clearly an outlier within this set. At the same time, my understanding of Steve McIntyre's point is that this group of 10 trees bears little resemblance to 17 or 18 other trees that are in close proximity. Briffa needs to explain why he used his trees given the small sample size.

    As to TP's sensitivity analysis - I would like to track which trees stay and which ones go as the age of tree filter is applied. I feel pretty certain that Steve McIntyre will provide a substantive response (and possibly even a confirmation) of TomPs conjecture. It still does not remove the issue of Briffa's lack of transparency.

    ReplyDelete
  56. Nick and Phil:
    All Keith Briffa has to do is to "show his work" and data. He can put an end to this discussion pretty quickly. In his response, he intimated that he would. Perhaps we should simply give him some time and in the meantime avoid pouring gasoline on the fire.

    ReplyDelete
  57. Eli is still going around asking WHAT data everyone is talking about. If it is the cores, then that belongs to the Russians, and McIntyre should have asked them. Briffa cannot ethically share what was loaned to him but is property of others. McIntyre tried that nonsense with Mann early on. If you loan a car from Roger I cannot demand you loan it to me.

    ReplyDelete
  58. EliRabett,

    That's a neat trick. "Sorry, love to show you how I got my results but the data belongs to someone else."

    By that logic it's Briffa that behaved unethically. By archiving the data for anybody to see, he loaned a car that didn't belong to him.

    ReplyDelete
  59. Roger, just noticed the headline on the newspaper article you linked to approvingly:

    "Only by playing with data can scientists come up with the infamous ‘hockey stick’ graph of global warming"

    They've inferred the same message from Steven McIntyre as the "nasty and vituperative" people at RealClimate.

    ReplyDelete
  60. EliRabett,

    Nobody actually needs Briffa to disclose anything else.

    As things stand, he has been caught using data that is obviously inadequate for its intended purpose, and he is now the subject of rumors (whispered and otherwise) that the reason for this inadequacy is a deliberate attempt on his part to create a hockey stick like result.

    Briffa may be able to put a stop to these rummors by disclosing additional data, and he may not.

    But the only people who actually need additional data disclosure right now are Briffa, and the people who have depended on his results.

    If he releases no additional data, you are free to assume that it is because those nasty Russians refused to give Briffa a chance to clear his name.

    I'm sure there will be other interpretations.

    ReplyDelete
  61. Jason, do you know whether he got permission from the Russians to do so?? If not you owe Briffa another apology.

    Moreover, you need to keep up kid, as more people get their fingers on what McIntyre said, his story is falling apart. For one thing it is clear that the Yamal reconstruction is robust to subtraction of many of the tree cores including the one that has been most at issue.

    For another it is not clear that the Schweingrubber cores that McIntyre added add any value, being mostly short and thus contribute mostly noise in the most recent years of the reconstruction.

    The simplest test is that pretty much all proxy reconstructions show the same structure, with a steep rise at the end (see the Real Climate article) even those that do not contain any tree ring cores.

    Mc is dialing it back as we write.

    ReplyDelete
  62. Eli, your re-direct is back to RC and a post by Tom P. - a euro-lurker. His math has continued to work on solutions that are biased by the author. He is looking for a signal - nay - looking for a hockey stick. Have you experienced one? Honestly, if you hadn't been told what was going to happen would you have gone out looking for any signal? we were told 20 years ago that there would be exponential changes in temperature, CO2, sea level... a hockey stick. In the fervor of identifying the blade, the best Q.C. is the shaft... I trust you will get it.

    ReplyDelete
  63. EliRabett,

    Actually you need to keep up. McIntyre had a quick look at Tom P's code, and couldn't even get it to reproduce Tom P's graphics.

    Oops.



    BTW, I have no idea what you are saying here:

    "Jason, do you know whether he got permission from the Russians to do so??"

    Did who get permission from the Russians to do what?

    ReplyDelete
  64. I'm rather confused about proxies. From what I can gather proxies (and in particular tree cores) are selected/discarded according to whether they "fit" a preconceived notion of the current rise in temperatures. In Briffa's case they were selection by persons unknown because they showed the latter 20th century hockey stick. Some showing it with an alarming 8 sigma variation in growth.

    Hence does it really surprise anyone that these selected samples each show the hockey stick? As EliRabett stated "robust to substraction". That is exactly how they were selected - each and every one.

    Selecting trees purely because they match your preconceived latter twentieth century hockey stick, doesn't mean they then can be used to confirm the latter twentieth century hockey stick. This is very twisted thinking.

    I thought the whole point is that you use these selected trees because you think they can be a reliable thermometer because they show some matches to data you have some confidence (the 20th century temperature recordings of firestations, airports and a bunch of places that professional climate scientists have managed to lose the original data for).

    At this point you hope (since Briffa presents no cogent argument to support) that these trees represent an accurate temperature proxy for their entire life. What's more you hope their response is linear to temperature and no other environmental stimuli.

    Of course if most trees behaved this way, you'd have a little confidence that they might act as proxies. A sensible person would probably lose some confidence in their accuracy over the longer timescale, since you've selected for 20th century matches, the odds are they won't all be perfect matches for their entire life.

    To fix this unease, you'd tend to prefer more trees in the hope that the signal is stronger than the noise.

    Unfortunately Briffa only used 10, and since so many more were discarded than used, you'd have to conclude that my assumption above "that most trees are good temperature proxies" doesn't seem to be supportable.
    So perhaps the trees you have selected are nothing more than a bunch of trees with a sudden growth spurt, perhaps due to other factors, and the odds of them all acting as proxies throughout their lives must be even lower.

    Of course, as a competent scientist, assuming you proceed down this tenuous path, you'd document your assumptions, why you have made your choices, how many samples make up your analysis, the confidence bounds, and archive the data so that it can be reproduced or extended by any scientists who wish to rely on your results.

    Or if you are a climate scientist, you don't.

    ReplyDelete
  65. "Eli is still going around asking WHAT data everyone is talking about. If it is the cores, then that belongs to the Russians, and McIntyre should have asked them. Briffa cannot ethically share what was loaned to him but is property of others. McIntyre tried that nonsense with Mann early on. If you loan a car from Roger I cannot demand you loan it to me."

    Boy this is tough one. How about this as a novel idea: If you borrow data from someone else and are unable to disclose it because you don't own it, then explain that you don't own the data and identify who does.

    ReplyDelete
  66. Eli
    In reverse order
    "The simplest test is that pretty much all proxy reconstructions show the same structure, with a steep rise at the end (see the Real Climate article) even those that do not contain any tree ring cores."
    Noone disputes the steep rise at the end, people looking into these proxy reconstructions dispute the methods used (and badly used proxies, eg upside down Tiljander, persistent use of non-proxies like bristlecones) which have the very obvious effect of flattening other steep rises and declines over the past 1000 years.

    "For another it is not clear that the Schweingrubber cores that McIntyre added add any value, being mostly short and thus contribute mostly noise in the most recent years of the reconstruction."

    Mostly "short" and contribute mostly "noise". That's nonsense - the trees selected do not correlate at all to the regional temperatures anyway - how are you managing to figure out what is noise and what isn't ? Oh yes, you knew what signal you wanted before you did the analysis.

    "Moreover, you need to keep up kid, as more people get their fingers on what McIntyre said, his story is falling apart. For one thing it is clear that the Yamal reconstruction is robust to subtraction of many of the tree cores including the one that has been most at issue."

    Interesting definition of robust. Removal of ONE tree reduces the late 20thC spike by 15-20%.

    In the other analysis of all 10 cores, approximately half of them are 200-250 years old, with the rest being older. So until you get to the 250 year cut off, it won't really change anything because very few (I think it is only 1) of the blessed Briffa 10 get removed in any of the selections prior to that. Once you remove half of the blessed Briffa 10 at the 250 year cutoff, you immediately get a warming period around 800AD which is greater than the 20th C:

    http://img202.imageshack.us/i/cru250.pdf/

    Robust ? Hardly. Embarassing ? Definately.

    ReplyDelete
  67. "Moreover, you need to keep up kid, as more people get their fingers on what McIntyre said, his story is falling apart. For one thing it is clear that the Yamal reconstruction is robust to subtraction of many of the tree cores including the one that has been most at issue."

    This may be a well-done analysis, but I am not going to put too much stock on comment 349 on a website that has banned SM from posting.

    ReplyDelete
  68. Btw, Eli I have to say I consider it hilarious that the main charge levelled on the Briffa reconstruction is that its late 20th C hockeystick is only dependent on 10 tree cores, and you come out and tell us it's robust to the REMOVAL of tree cores ?

    Just imagine your reaction if you criticised a homeopathic study due to it's small sample size, and they came back with the claim it was robust to REMOVAL of data.

    ReplyDelete
  69. Courtesy Deep Climate at Deltoid, a Steven Mcintyre quote I missed:

    "I'd be inclined to remove the data affected by CRU cherrypicking but will leave it in for now."

    I assume you missed this too Roger, and will now be retracting the claim that Gavin lied and apologizing.

    ReplyDelete
  70. -69-andrewt

    You guys are hilarious. There is no need to pluck out-of-context quotes from deep in comment threads to divine what McIntyre _really_ thinks. He spoke directly to this point as follows:

    "I don't wish to unintentionally feed views that I don't hold. It is not my belief that Briffa crudely cherry picked. "

    How clear is that?

    ReplyDelete
  71. Roger,
    The vexing thing about blogdom is that one sentence never does the trick.
    "... Crudely...?"

    No-one who has thought abut it would think the cherry-picking was done "crudely." But my suspicion, and I think Steve's is that the apparent effect is a result of a proxy selection process which Briffa did not describe, and may not have been consciously biased, but biased it was, nonetheless.

    And on that note, shouldn't it continue to be worthwhile to discover and qualify families of proxies with temperature signals which could carry us back thousands of years.

    It looks like that is what Briffa said he was doing.

    ReplyDelete
  72. Roger plucks a quote from deep in the comments to demonstrate that one doesn't need to pluck a quote from deep in the comments to divine what McIntyre thinks. Hilarious indeed.
    But what "McIntyre really thinks" is irrelevant to this discussion - it's what he's said that matters. And the fact that he's contradicted himself certainly isn't in his favour.
    By the way, what context is missing from Andrew's quote?

    ReplyDelete
  73. -72-lenny

    Actually I had no need to pluck anything for the comments since Steve McIntyre did a headline post on this exact subject:

    http://www.climateaudit.org/?p=7257

    All this what-he-said-meant-or-implied-or-did-not-say is quite fun, but tell me, why does it matter, exactly?

    ReplyDelete
  74. JohnF,

    Including just 10 cores covering the most recent period, none of which display the statistical properties which Briffa's analysis assumed, is most certainly crude (whether or not Cherry Picking occurred).

    A sufficiently intelligent person with access to hundreds of cores from a region COULD cherry pick a set which supported his conclusions, was scientifically adequate (or better), and was not identifiable as being cherry picked through any method other than the lack of replicability.

    But that's not what were are looking at here. If any sort of cherry picking occurred, it was most definitely crude.

    ReplyDelete
  75. Roger,

    You wrote: "All this what-he-said-meant-or-implied-or-did-not-say is quite fun, but tell me, why does it matter, exactly?"

    Your post was about a supposed "outright lie" stemming from RC, so that makes me think that it presumably matters to you. Your McIntyre quote together with the numerous quotes Andrewt, myself and others provided reinforces the tactics of McIntyre as Andrewt described in his comment nr 8: He speaks with two tongues. One is meant for his supporters, one to defend himself against critical voices.

    Bart

    ReplyDelete
  76. Jason, S
    Maybe flagrant is a better word. Although the flagrance went undetected until Sept 9, 2009.

    ReplyDelete
  77. Roger,
    All this what-he-said-meant-or-implied-or-did-not-say is quite fun, but tell me, why does it matter, exactly?
    This is rather a climb-down. Your post is all about accusing Gavin at RC of lying about what Steve McI implied. That's your headline.

    ReplyDelete
  78. -75-Bart and -77-Nick

    You guys are confused.

    This post is about Gavin Schmidt, a US government scientist and self-proclaimed leader of the scientific community, making an outright lie to cast aspersions upon a blogger who sticks his nose into the scientific community in inconvenient ways. Does this matter? Yes, I think it does, because it makes the entire climate science community look bad. Do I have high standards for climate scientists? Yeah, they are professionals and should act like it. You may have lower standards.

    If you guys want to talk about something McIntyre said-meant-or-implied-or-did-not-say, good for you. In the main body of the post I have provided my answer to this question. If you are trying to do this to excuse Gavin's behavior, well that is s stretch. If you are doing it to somehow discredit McIntyre, I doubt you'll achieve that goal, but good luck.

    ReplyDelete
  79. Roger, the confusion is all yours. You're saying that it's important that Gavin "lied" about what SM was implying, but when people like andrewt make an overwhelming case that SM not only was implying cherrypicking (despite occasional later denials) but explicitly used that phrase, you say this doesn't matter. It does - Gavin wasn't lying. He was right.

    When Steve begins a post with "The second image below is, in my opinion, one of the most disquieting images ever presented at Climate Audit." and soon after his associate, Anthony Watts, is running a post calling for Briffa to resign, really, what do you think is going on?

    ReplyDelete
  80. Gavin Schmidt is a U.S. government scientist? That is insane. I want my money back. That guy is anything but professional. He uses sarcasm, insults, personal attacks. Rarely does he actually address scientific issues. I am really appalled to understand that someone with his unprofessional attitude is on a public payroll!

    ReplyDelete
  81. You guys are hilarious. There is no need to pluck out-of-context quotes from deep in comment threads to divine what McIntyre _really_ thinks

    Here is the surrounding context for the last Steven McIntyre quote I provided:

    "The logical data set, as I said above, is the combination of Urals and Yamal and russ035 in a NW RCS, which I was working on earlier today and will finish tomorrow. I'd be inclined to remove the data affected by CRU cherrypicking but will leave it in for now"

    Yes it is deep in a thread (comment 184) but the quote you start your post with is comment 254 in a thread.

    Steven McIntyre unambiguously says what Gavin claimed he said. Gavin did not lie.

    Steven McIntyre has now added a retraction after this comment.

    Will you also add a retraction to this post and an apology to Gavin?

    ReplyDelete
  82. Thanks once again Roger for creating an interesting discussion - as usual for you, it ends up being about something that the main post is not, and also as usual it is more telling of peoples a priori biases than about the facts.

    ReplyDelete
  83. Roger,

    I’m not aware of Gavin having proclaimed (or even insinuated) to be the leader of the scientific community; perhaps you could point me to such?

    How I read this whole sorry story is that a lot of scientists are getting understandably frustrated with self-proclaimed auditors of science who cast doubt about a whole scientific field by blowing minor flaws out of proportion and insinuate accusations of scientific misconduct (“the archive is suspect”, etc.).

    The only way you could support your claim of an outright lie is to close your eyes for McIntyre’s (cleverly packaged) insinuations.

    This science-bashing is quite popular in some quarters, but I hold you as a political scientist to a higher standard.

    Bart

    ReplyDelete
  84. Climate audit responds to Tom P. and Gavin http://www.climateaudit.org/?p=7278

    ReplyDelete
  85. -83-

    Bart-

    Thanks. Just a few replies on this thread that is now circling back over well-covered ground.

    1. Anytime you publicly claim to be representing truth on behalf of an entire scientific field to the media, public and policy makers you are taking a leadership position. This seems fairly obvious.

    2. Scientists getting frustrated does not excuse departing from standards of professional conduct. Climate science is now of big time political importance, and guess what, it'll be treated like knowledge in any other big time political issue. Climate scientists need to get used to it and grow some thicker skin, as well as a deeper appreciation of how the political world works.

    3. This thread has been interesting, not least because it shows how much of a Rorschach test events like this are. You (and a few others) see this as "science bashing" while I (and a few others) see it as trying to uphold what it most important about science, i.e., the opposite of bashing. If nothing else it shows how the world looks through the eyes of another.

    I've looked at the events, and concluded what I've written in the post. Others may choose to disagree. And after our exchanges, if we don't come to a shared agreement, then we agree to disagree.

    However, one thing you'll find here and not at Real Climate and not in my experience at Climate Audit -- on this topic and any other, all perspectives are welcomed as part of the discussion.

    ReplyDelete
  86. Dr. Pielke:

    Your recent statements here with regard to your original post are disingenuous. It is quite clear that Mr. McIntyre accused Dr. Briffa of professional corruption, no matter how coy the language he used. It was certainly clear to the various media that immediately began shrieking "fraud," none of which have seen fit to modify their stories in the light of McIntyre's tepid post hoc disclaimer.

    It is most disappointing to see you defending McIntyre's dishonest behavior by attacking a colleague. That you should claim you are doing this in the name of "high standards for climate science" is disgusting.

    ReplyDelete
  87. Adam, that's hilarious. Steve McIntyre did nothing of the sort. Cherry Picking may have been done unintentionally, and SM even hinted at the possibility that it were the russians who did it in the first place, also unintentionally.

    The big problem here is that there are a lot of people confusing "cherry picking" with "professional corruption". Cherry picking is done extensively and often in many statistical analysis, without corruption of any kind, but rather unintentionally (invisible bias), or intentionally, but with a fair reason behind it.

    Keith Briffa acknowledged that there were things to be "resolved" in his own letter, so my guess is that SM was right on his criticism, for it is quite rare to see a scientist making such a retracted claim wrt a lifework's being smashed like SM is doing.

    So Pielke is right by calling this a Roscharch test, and I find quite hilarious that you try hard and harded to impugn invisible motives unto SM, who has shown professionalism while being amateur, and while Gavin, who aknowledges his own lack of knowledge about "tree stuff" in his own blog, but does not stop himself at accusing SM of saying things he did not say, and of making erroneous criticisms. Perhaps "it doesn't matter".

    Well, I do not think so. Briffa does matter, and so does MBH99. Most climate models are modelled with the assumptions created by these climate reconstructions, as the IPCC states, somewhere around page 660~680.

    What is disgusting is your disgust. Shed your tears and enlighten yourselves over these issues. When you are done erasing your own hubris and indignated snobbiism, perhaps you are ready to see the mess that this leaves climatology.

    Or perhaps you already did this and are only on denial mode....

    ReplyDelete
  88. Adam you really have drunk the RC Kool Aid as they say.

    The blade of a team climate staple hockeystick like reconstruction has been shown (after almost a decade of being hidden) to be due to a picked subselection amounting only to 10 tree cores which do not even correspond to their local temperatures.

    Scandalous.

    However, RC has seemingly successfully convinced their followers that the important issue here is that Steve accused Briffa of cherry picking the proxy data.

    Even this claim is not true though. In his very first post on the topic, as Pielke said, McIntyre stated:

    "In my opinion, the uniformly high age of the CRU12 relative to the Schweingruber population is suggestive of selection – in this respect, perhaps and even probably by the Russians"

    McIntyre stated clearly that he thought it was probably the Russians who created the "cherry picked" chronology.

    However. There is ANOTHER cherry picking going on (the kind which is very common in climate science), where the chronologies themselves are "cherry picked". That is to say Briffa CHOSE to use THAT chronology from the Russians, and CHOSE not to beef it up with other chronologies from the same area.

    The reason nor method for the Russian cherry picked chronology are UNKNOWN.

    The reason for Briffa's preference for the cherry picking of this particular chronology is UNKNOWN.

    Steve comment which supposedly proves his guilt:
    "I'd be inclined to remove the data affected by CRU cherrypicking but will leave it in for now."
    was posted late at night, and at least to me it's not clear whether Steve is talking about the cherry picking of the proxy data, or the cherry picking of the chronology, or even both, and was just a quick shorthand late at night OVER A DAY AFTER it was originally posted.

    It obviously means CRU's cherry picking of the chronology, or CRU's use of the cherry picked chronology. Both of these statements can be made without accusing Briffa of personally handpicking cores, and the term "CRU's cherry pick" for both those scenarios when that is what is being discussed is perfectly clear shorthand. It's the RC people reading into it what they WANT it to mean.

    Anyway, as I said, you've been fooled into righteous anger over a complete side issue. Are you not slightly hot under the collar that the hockey stick shape of a well used reconstruction is so statistically non-robust, depending on just 10 particular cores ?

    ReplyDelete
  89. Something more about dr Gavin. He posted this on RC in response to an unsatisfied client, asking for quotes where SM accused Briffa. This is what happened.

    The half he quoted:
    “In my opinion, the uniformly high age of the CRU12 relative to the Schweingruber population is suggestive of selection” (McIntyre)

    The full sentence:
    “In my opinion, the uniformly high age of the CRU12 relative to the Schweingruber population is suggestive of selection – in this respect, perhaps and even probably by the Russians”

    Talk about CHERRY PICKING! Dr Gavin is showing childish behavior on and on. I so wished he cut the infantile act and behaved like the gentleman he so aspires to be!

    ReplyDelete
  90. Words mean something and in clear terms, Dr. Pielke is right and Gavin has misrepresented Steve McIntyre. It's unambiguous. Implications are irrelevant. What commenters say is irrelevant.

    What Steve and Gavin have said, EXACTLY is what matters.

    Was the data cherry picked? It certainly seems it may have been by someone. Did Steve accuse anyone in particular of doing so? No, not that I saw.

    Done. Gavin is wrong. Steve is right. Gavin owes an apology.

    M

    ReplyDelete
  91. Briffa is a fraudster pure and simple. He delivered the data when the Royal Society demanded it so he obviously could have done so years ago when McIntyre requested the data. Science and Nature have been exposed as corrupt by not adhering to their own policies on data availability. Money corrupts and the tens of billions being pumped into climate science for political reasons has corrupted absolutely. Hansen received a $1 million award from the Tides Foundation for his great work "saving the planet". If you audited the personal finances of RC contributors finances I suspect you would find similar but smaller awards (payoffs) for propaganda activities.

    In a few years when the global cooling cycle is well advanced and the political hysteria dies down there will be a true accounting and it will be the equivalent of Watergate for organized science. The RC crowd, Briffa, Mann, the IPCC, etc. will be the 21st centuries Piltdown Man.

    ReplyDelete
  92. We can give Briffa many benefits of the doubt, and assume that the 10 tree-ring selection was made not in a deliberate attempt to skew data, but through some other process.

    However, the motivations are not relevant to the science. What IS relevent is whether a proper treatment of the data supports the conclusions presented. THEY DO NOT. There is no way, after you look at what is actually in the data, that you can fairly conclude the 'hockey stick' conclusion is justified and those 10 tree rings are somehow a robust representation. They are not.

    The practice, in general, of lining up any such record with last 20th century global temp trends and throwing out those that dont match as "noise', while keeping those that do, is a simple case of lying with statistics that in general, will ALWAYS lead to a hockey stick. Here is why: Lets assume the tree ring consists of 2 independent variables - a random variable and a 'temperature signal' variable, each about 50% weight. Any given data set will have a 50% correlation level on average and thus the temperature signal

    Anytime you weight data or select data based on recent temperature trends you are automatically creating a bias (again, statistical term, not personal pejorative) that overemphasizes the recent temperature trends for the simple reason that correlations overall are not 100% to temperature. the logical flaw is the assumption that correlation over a short range implies better correlation over the full range. That's not a correct assumption in a large population. The smaller the temperature 'signal' to non-temperature 'noise' in the tree-ring data, the greater that error.

    Some of McIntyre critics are attacking him not on the main substance, but on a (false imho) claim of 'who said what'.

    Meanwhile, the elephant in the room, the fundamental logical error that underlies the statistical selection method is left un-discussed.

    ReplyDelete
  93. "no YOU said"

    Really, this refusal to accept statements which are quite obvious to the lay person is laughable. Dr. Pielke, your "cherry picking" of which comments to evade, avoid or answer, and your manner in those you dein to reply to, does you and your argument no favours.

    It goes without saying that you should always separate what a person says and the personality of that person, but you make it hard to give what you say any credit whatsoever.

    But i'm sure if you ever read this you'll ignore it with some hugely misguided air of superiority - because people who demonstrate themselves as you have done never accept they're wrong or have anything to learn in any sphere.

    ReplyDelete