M&M: PNAS Comment

Ross and I submitted a comment on Mann et al 2008 on Dec 8, 2008 within the 3 month time period for comments permitted by PNAS. The comment, pursuant to PNAS rules, was less than 250 words and had 5 (or less) references (one of which was Mann et al 2008 itself). The 250-word limit doesn’t enable one to say a whole lot. However, one advantage of this is that PNAS has a more active comment section than some journals.

While PNAS has pretty liberal rules on pre-prints being available, I’m not going to press the point and am not going to post it up right now.

Thanks to UC, Jean S and Hu McC, all of whom commented on drafts of the Comment.

As a little game, readers might be interested in trying to guess the other 4 references. One should be very easy; two more should be guessable by alert readers; it will be hard to get all four.


  1. Sylvain
    Posted Dec 13, 2008 at 12:27 AM | Permalink

    250 words!!!

    Do they want to make sure that no one can make a substantiated case against a paper?

  2. Posted Dec 13, 2008 at 1:28 AM | Permalink

    The following three would have high potential:
    1. MM, 1998
    2. McIntyre/McKitrick 2003
    3. AR4, 2007??
    4. ???

  3. skepto
    Posted Dec 13, 2008 at 3:25 AM | Permalink

    This site has now become the exclusivity of those who understand the physics and maths behind the science of climatology. I wish I had the time to study and understand all that’s discussed here. Sadly, I don’t. It might as well be written in Greek. Sorry, you all lost me (and I have a PhD in Physics).

    • Gerald Machnee
      Posted Dec 13, 2008 at 8:27 AM | Permalink

      Re: skepto (#3),
      Now CA has one of the largest group of competent contributors. This means progress and that is great even if I cannot follow all the details. But i am enjoying it. We look forward to the results of this comment by Steve and Ross.

    • Jeff Alberts
      Posted Dec 14, 2008 at 4:55 PM | Permalink

      Re: skepto (#3),

      I feel your pain, Skepto. I don’t have a degree in anything, and almost failed Algebra I more than 30 years ago. So you can imagine how my eyes glaze over. But I can usually understand the underlying gist of what’s going on, and the comments really help me a lot.

  4. jlc
    Posted Dec 13, 2008 at 5:21 AM | Permalink

    Is there a prize?

  5. kim
    Posted Dec 13, 2008 at 5:57 AM | Permalink


  6. per
    Posted Dec 13, 2008 at 6:49 AM | Permalink

    i will guess at the nrc report as the easy reference; the rest will be statistical references where I cannot begin to understand the words used :-)

    congratulations, and I hope your revisit to the peer-reviewed literature is successful. I suspect you have made many points on the blog that are salient, and it is nice to bring this to attention in the peer-reviewed literature.


  7. Craig Loehle
    Posted Dec 13, 2008 at 7:42 AM | Permalink

    I am guessing the Tiljander paper for using invalid temp proxy. 250 words! what a joke! You can only point out 1 tiny error and without proof.

  8. Craig Loehle
    Posted Dec 13, 2008 at 7:43 AM | Permalink

    By the way, you used up 1% of your allowable words just adding Ross to the header…

  9. MeToo
    Posted Dec 13, 2008 at 7:52 AM | Permalink

    Exactly, I don’t have a PHD, so you can imagine. How I crave an English version of all this, I know intuituively Steve is right. Just please write a little summary for us dummys.

  10. Brent Buckner
    Posted Dec 13, 2008 at 8:50 AM | Permalink

    There were a few blog posts that would have been worth referencing….

  11. Steve McIntyre
    Posted Dec 13, 2008 at 8:54 AM | Permalink

    #6. One point to per on the easy reference. 0 points to 2.

    #3. It’s ironic that I get so much advice from climate scientists on how to get more “scientists” to visit here when it is very obvious that many scientists, PhDs and well-educated professionals visit here – it’s just that they tend to be scientists from other disciplines who, like me, are grappling to understand the issues and who are seeking a level of exposition in between a Grade 5 cartoon and terabytes of output from a GCM, constructed with opaque code over many years.

  12. José Pires
    Posted Dec 13, 2008 at 10:09 AM | Permalink


    • DeWitt Payne
      Posted Dec 13, 2008 at 9:26 PM | Permalink

      Re: José Pires (#13),

      Completely OT, but since the point was raised: The lower two graphs of Arctic area and extent at Arctic ROOS have been incorrect for many weeks. One only had to compare the positions of the end points of the 2008 data or compare the separation of the rising portion of the 2008 line from the 2007 line to see the problem, not to mention comparing to JAXA extent and CT area data. The upper graphs were corrected some time ago and now the lower two, which compare 2008 and 2007 to historical average data, are correct as well.

  13. Mike B
    Posted Dec 13, 2008 at 10:20 AM | Permalink

    I’m guessing Santer et al 2008 for the discussion of how autocorrelation should be handled.

  14. AndyL
    Posted Dec 13, 2008 at 10:35 AM | Permalink

    I’m guessing a Tiljander paper for the upside-down use of sediments – possibly Tiljander, Mia, Matti Saarnisto, Antti E. K. Ojala and Timo Saarinen, 2003.

  15. Steve McIntyre
    Posted Dec 13, 2008 at 10:58 AM | Permalink

    Santer was a good guess and in one draft , but not used. Tiljander is a good guess, but not used; we described the issue though. Woodhouse and Brow – nope.

    • Mike B
      Posted Dec 13, 2008 at 12:19 PM | Permalink

      Re: Steve McIntyre (#17),

      Rats. At least I was warm (no pun intended).

      How about Nychka 2000 or Soon et al 2004?

      I loved the Cobb 2007 reference (wind direction proxies), but I’d be surprised if you used that one.

    • Not sure
      Posted Dec 13, 2008 at 3:16 PM | Permalink

      Re: Steve McIntyre (#17), OK, I’m going to guess that you did not raise any issues with the proxy data at all, then. Like you say, there are many interesting defects in the paper. I’m going to guess you limited yourselves to problems with the statistics. The list of reviewers reinforces that hypothesis. Unfortunately, I can’t follow the stats at all, so all I can offer are a couple of stabs in the dark: Jaynes (1976) and Nychka (2000). Perhaps a cite of Dr. McCulloch’s work as well.

  16. Posted Dec 13, 2008 at 11:18 AM | Permalink

    Here’s my guesses

    Tiljander upside down proxies and Punta Laguna for made up data, Mann 08 for briffa infilling woodhouse and brown as #16.

    I hope something was mentioned about correlation sorting and CPS but it hasn’t been your focus and you can only say so much with two hun……..

  17. Posted Dec 13, 2008 at 11:21 AM | Permalink


    • Geoff Sherrington
      Posted Dec 13, 2008 at 6:23 PM | Permalink

      Re: Jeff Id (#19),


      Guess on PNAS reference. The Bible. You might need to invoke it again.

  18. Posted Dec 13, 2008 at 11:29 AM | Permalink

    I’m not even going to try to guess. :)

    Re: Steve McIntyre (#12),
    On the advice for how to get more scientists to visit:

    I often suspect that sort of advice is suggest the commenter has assumed their “concern troll” incarnation.

    It’s also ironic that those giving the advice seem to ignore the fact that your blog gets lots of traffic, and the non-sock-puppet commenters often have ph.d. in technical fields– sometimes even science! (Many of the anonymous, nearly anon. commenters or just “no last name” commenters also have advanced degrees– often in science. )

    In any case, the advice implies that attracting “scientists” per se should number among your goals. Is there any strong argument that your goal should be attracting ‘scientists” per se? Why would you wish to attract astronomers, chemists or epidemiologists in preference to aeronautical engineers, statisticians or software developers? Maybe you stay awake at night dreaming of attracting the first group and reviling visits form the second. But if you do, it’s not clear that changing your approach would work. After all, is there any strong evidence that the comment section at RC, Tamino or Eli’s blog is visited by zillions of people in those fields? Or climate scientists?

    Ok… I’ll admit I strayed off topic. But…. you tempted me off the “guess the references” path.

  19. Posted Dec 13, 2008 at 11:36 AM | Permalink

    OT, but not irrelevant to CA readers:
    Jose Pires, #13, is correct. For the last few weeks NANSEN indicated that Arctic ice extent was within the ±1 Std Deviation (SD) of the 1979-2007 climatology, and in fact had reached the average level in the past few days, while the ice extent had gotten to within the ±1 SD range. But today NANSEN indicates that neither is within the ±1 SD any longer!! And there is no explanation on the specific page, or the home, observation or background pages that I could locate for this revisionism. [I had checked it only yesterday.] Anyone know what gives?

  20. bmcburney
    Posted Dec 13, 2008 at 12:35 PM | Permalink

    Dodgson, 1865.

    Famous 19th Century mathematician, Charles L. Dodgson’s most famous work.

  21. Phillip Bratby
    Posted Dec 13, 2008 at 1:18 PM | Permalink

    I’m another physics (nuclear) PhD who struggles with the issues and the statistics. But I’m a regular reader here and a strong supporter as there seems to be so much integrity, honesty and seeking after the truth on this site. People aren’t afraid to air their views and have them criticised. That’s good. That’s how progress is made. Thanks Steve.

  22. Steve McIntyre
    Posted Dec 13, 2008 at 1:28 PM | Permalink

    #18. As you well appreciate, there are dozens of different and interesting defects in Mann et al 2008 worthy of being mentioned. Such a banquet of choices. If one wanted to design a lively applied statistics course illustrating all the sorts of pitfalls that people can fall into, one could have built an entire course around MBH98 and now a second one on Mann et al 2008. Someone who understood the problems would become pretty handy at applied statistics – I’m sure that the retention rate among the students would vastly exceed what they get out of some meritorious but dry statistics books.

    #22. Nope.

  23. Richard Sharpe
    Posted Dec 13, 2008 at 1:37 PM | Permalink

    Heh, yes, Alice’s Adventures in Wonderland

    But perhaps this from Through the Looking glass is more appropriate:

    “I weep for you,” the Walrus said:
    “I deeply sympathize.”
    With sobs and tears he sorted out
    Those of the largest size,
    Holding his pocket-handkerchief
    Before his streaming eyes.

  24. AndyL
    Posted Dec 13, 2008 at 1:55 PM | Permalink

    Following the Santer thought, and after googling this site, how about Gershunov et al

  25. Posted Dec 13, 2008 at 2:54 PM | Permalink

    OK, so it’s not Dodgson, 1867. How about Milton, 1667?

    He trusted to have equal’d the most High,
    If he oppos’d; and with ambitious aim
    Against the Throne and Monarchy of [Gore]
    Rais’d impious War [with R] and Battel proud
    With vain attempt. Him the Almighty Power
    Hurld headlong flaming from th’ Ethereal Skie
    With hideous ruine and combustion down
    To bottomless perdition, there to dwell
    In [Mann-o-matic] Chains and penal Fire,
    Who durst defie th’ Omnipotent to Arms.

  26. Patrick M.
    Posted Dec 13, 2008 at 2:54 PM | Permalink

    Knowles et al? link

  27. Red Etin
    Posted Dec 13, 2008 at 3:53 PM | Permalink

    I’ll hae nae hauf-way hoose, but aye be whaur
    Extremes meet – it’s the only way I ken
    To dodge the curst conceit o’ bein’ richt
    That damns the vast majority o’ men.

    MacDiarmid, 1926

  28. cbone
    Posted Dec 13, 2008 at 3:54 PM | Permalink

    Did you referenced Jones, I (1981) again?

  29. anonymous
    Posted Dec 13, 2008 at 4:45 PM | Permalink

    I’m guessing its the one about spurious correlation between rainfall and inflation ?

  30. Terry
    Posted Dec 13, 2008 at 4:58 PM | Permalink

    Re Lucia (#20) and Steve #17, I too have a PhD in physical chemistry, so I have a good understanding of the basic GHG theory. But I vist this site, Lucia’s, Briggs, and JeffID on a regular basis to try to grasp some of the advanced stats, since all that we tradionally used was OLS and PCA at the most. It is tough sometimes, but in many cases an in-depth understanding of the stats theory is not required, and in any event I have come to learn over many years, to leave the advanced theory in disciplines that are not my own, to those that are expert. They likewise leave the in depth understanding of my areas to me. I still try to grasp some of the theory, since “no pain – no gain” is always true.

  31. jae
    Posted Dec 13, 2008 at 8:14 PM | Permalink

    Let’s see. We know NRC is there. It seems MM05 has to be there, and MBH98 should be there. One more. Some statistician that I probably have never heard of before.

  32. Sylvain
    Posted Dec 13, 2008 at 9:34 PM | Permalink

    Briffa and cook white paper must be in there
    the NRC should also be in it

  33. M. Villeger
    Posted Dec 13, 2008 at 9:58 PM | Permalink

    Could it be liked to the Amann SI story? Or the Swindle story judgement?

  34. Dennis
    Posted Dec 14, 2008 at 3:57 AM | Permalink

    In regard to the ice graph issue the response from their website was as follows:

    We made a correction to the 2008 curve yesterday after we found that there had been a too high estimation of the area from 22 October up to yesterday. The new curve for 2008 is now following a more correct slope.

    We appreciate comments to the data presented on the web site.

  35. Stuart Harmon
    Posted Dec 14, 2008 at 5:59 AM | Permalink

    1 De Zheng Sun et al July 2008

    2 Camp and Tung 2007

    3 There’s a hole in my Bucket. Widow Swankey at the New scientist.

  36. Ruth
    Posted Dec 14, 2008 at 6:22 AM | Permalink

    I don’t expect you quoted this, but it seems to sum up the whole approach of this site (and IMO should be the attitude of every scientist):

    1 Thessalonians 5.20-22. Do not despise the words of prophets, but test everything; keep hold of what is good, and avoid all forms of evil.

    [for prophets, read 'the consensus', 'IPCC reports', or 'insert name here'...]

  37. Mike B
    Posted Dec 14, 2008 at 7:53 AM | Permalink

    There has to be something about spurious correlations…and I would it would need to be a climatology-related reference.

    Gershunov, et. al. 2001 on spurious correlations of low frequency smoothed series, i.e. the Slutsky-Yule effect?

    Another one has to be related to data mining tree-ring proxies, particularly those that may be primarily precip proxies. But there are enough to choose from that I’d just be guessing randomly.

  38. PaulM
    Posted Dec 14, 2008 at 12:07 PM | Permalink

    Steve, I’m delighted to hear that you did submit a comment, that’s excellent news, I look forward to seeing it.
    Please heed the comments of Skepto, MeToo and others, and write a longer summary of the Mann etal 08 problems in plain English as you did to good effect in your ohio.pdf

  39. Steve McIntyre
    Posted Dec 14, 2008 at 12:34 PM | Permalink

    I’m quite a way from finishing with Mann et al 2008. I haven’t scratched the surface of their EIV method yet, which is quite interesting statistically in a Little Shop of Horrors sense.

  40. Steve McIntyre
    Posted Dec 14, 2008 at 12:44 PM | Permalink

    #43. Something like that wouldn’t have been out of place, given the insidious role of smoothing. If we’d had 300 words and 6 references, I’d have used something like that (or maybe Santer), but no.

    #41. Not even close. The solar stuff hasn’t got anything to Mann. Jeez.

    Not any of our own papers or MBH98.

    #18. You’re very warm on correlation sorting in CPS. This issue has been discussed on several blogs (passim here on many occasions), with the same thought occurring more or less independently to different people. We included a non-blog reference on this (though CRU and UCAR probably don’t carry the journal.)

  41. Posted Dec 14, 2008 at 1:31 PM | Permalink

    Re Steve #45,

    I’m quite a way from finishing with Mann et al 2008. I haven’t scratched the surface of their EIV method yet, which is quite interesting statistically in a Little Shop of Horrors sense.

    Perhaps you and Ross can put together a permanent “SI” webpage pertaining to Mann et al 2008, with links to your comment and to the many CA posts criticizing specific details, together with some indication of which post(s) pertain to which specific objection you raised in your comment. And which of the many blog comments you endorse as being important. That way, Mann and/or third parties would have a way of finding the details of your objections without having to slog through all of CA.

    As you develop new criticisms, links to these could be added to the SI page in a separate section of “further problems” not discussed in the PNAS letter.

    Ultimately, these could perhaps be consolidated into a paper to CC or some such journal.

  42. whippet
    Posted Dec 14, 2008 at 2:42 PM | Permalink


    you are a one trick pony. The climate change argument is driven by the physical basis, that is, CO2 is a greenhouse gas. You had an awkward stab at it once, then appeared to decide you were way out of your depth and left it alone. Until you get that sorted out, all this obsession with Mann and Hansen is just a waste of your time. Paleoclimatology is NOT the case for AGW. You can rant on all you want, but it’s just a sideshow, and it gets more tedious by the minute.

    • UK John
      Posted Dec 14, 2008 at 2:53 PM | Permalink

      Re: whippet (#48), Could you please provide link to the physical basis for AGW you state.

    • stuart harmon
      Posted Dec 14, 2008 at 3:39 PM | Permalink

      Re: whippet (#48),

      If you are interested in physical evidence regarding global warming/global cooling I suggest you look up Robert Carter on Utube. This series of lectures will provide you with great understanding of the physical world.

      Now regarding ponies?

      The Hockey Stick graph was produced from physical data and then used statistics to arrive at certain conclusions.

      These conclusions flew against all of the evidence both empirical and observed of the past history of the planet. Especially the last 1000 years.

      The service to science which McIntyre and McItrick carried out was to show the methodology and use of data was wrong and hence the conclusions were false. See Wegman et al.

      Wrap up warm its getting colder.

    • Craig Loehle
      Posted Dec 14, 2008 at 4:33 PM | Permalink

      Re: whippet (#48), Why does the Hockey Stick matter? It is used in the IPCC reports to argue that 1) the recent warming is unprecedented in rate and 2) the temperatures we have already reached are unprecedented. That is, it is used to show that it has already reached a point never before seen. Therefore alarm bells go off. If the MWP was warmer, if climate has fluctuated a lot naturally, then the basis for alarm is muted.

    • Posted Dec 14, 2008 at 7:44 PM | Permalink

      Re: whippet (#48)

      Paleoclimatology is NOT the case for AGW.

      The same is said about the GCMs. And basically the same is said about the GMST; only 60 data for the entire globe are needed. And if the ice makes a come-back that too will be listed. If things continue to be added to the no-never-mind list, what’s going to be left?

  43. Dave Andrews
    Posted Dec 14, 2008 at 3:09 PM | Permalink

    I’d like to heartily endorse what Phillip Bratby, #24, said about this blog.

    I can’t begin to even guess the answers to Steve’s quiz but after a year or so of reading climate science blogs you develop a ‘feeling’ for the integrity of what is being written and how the site is conducted. Contributors here rarely resort to the insults that are traded regularly on other blogs. They always seem to try to explore the science and are courteous with those whose point of view may be different – witness the lengthy discussions with Judith Curry and J E-G last year. When have you ever seen anything like that on RC or Open Mind?

  44. Posted Dec 14, 2008 at 3:33 PM | Permalink

    Poincare said, “The scientist must order; science is made out of facts as a house is made out of stones, but an accumulation of facts is no more a science that a heap of stones, a house.”

  45. Posted Dec 14, 2008 at 3:56 PM | Permalink

    Why does it bother you if Steve comments on Mann’s more recent papers?

    If a paleo analyses hypothetically contributes to the literature then comments on that paleo analysis contributes equally. Of course people comment on those analysis that fall in their areas of expertise rather than those outside their areas of expertise. Steve and Ross happen to be famiiar with the sort of statistical analyses Mann applies and so comment on that rather than something else.

  46. Ross McKitrick
    Posted Dec 14, 2008 at 4:32 PM | Permalink

    #48: Whippet… My my, feeling a bit testy today, whoever you are. If paleoclimatology matters so little why was the hockey stick promoted so heavily by the IPCC, and why did the paleo topic continue to get a whole chapter in AR4? Why did the Arctic Climate Impact Assessment and the CCSP Synthesis Report put the hockey stick graph front and center? Etc. etc. You can’t erase the whole record of promotion of paleo-based arguments by those promoting AGW, using the “warmest in a thousand years” tag line.

    Now, if you are conceding that the HS was wrong and climate reconstruction using multivariate calibration between proxies and temperature data yields little more than uninformative noise, but you also think it doesn’t matter for the purpose of arguing AGW, then that’s fine. (However, if, seeing it phrased like that, you feel provoked to defend the HS and multiproxy reconstructions like it, maybe you actually do think it matters.)

    There are lots of topics that get dealt with here beyond paleo climate, maybe you just haven’t looked, or don’t understand the relevance to the big picture. The paucity of attention to the fundamental basis of CO2-induced global warming is because of the paucity of literature actually spelling it out. (But if you’re such an expert, feel free to point us to the papers that establish the physical basis, from first principles, for, say, a 3C mean increase in the surface temperature field due to doubling CO2.) Infrared absorption by CO2 doesn’t get you very far in terms of actual predictions. To get the big global warming effects shown in climate models you need to characterize the water vapour feedback and a host of other secondary processes. They are represented in models using parameterizations of sub-grid scale phenomena. The parameterizations are empirically-based, at best. In other words they depend on things like surface temperature data, sea surface data, balloon and satellite data, etc. Steve and the other contributors to this site have made significant contributions to the work of critically examining these data sets, improving disclosure and accountability, fixing computational errors, documenting flaws and inconsistencies, etc. If you can’t see any relevance to this–if it’s all a tiresome sideshow to you–then you have a strange view of how science should proceed.

    • Mark T.
      Posted Dec 15, 2008 at 10:15 AM | Permalink

      Re: Ross McKitrick (#54), Wow, for being a co-author to a one-trick pony, you sure used some rather detailed theoretical terminology in explaining your position. Ahem… ;)


    • MC
      Posted Dec 15, 2008 at 2:42 PM | Permalink

      Re: Ross McKitrick (#54), On Lucia’s blog there’s reference to the original Hansen paper where the ‘radiative forcing’ (Box 1) explanation is just thrown in. Some equations to start off. Simple planetary heat balance and then what? Where did that equation come from? No reference to the derivation except a paper about aerosols which is fat lot of good. And then the train starts rolling and suddenly there is now a log relationship for ‘forcing’ that makes it way into IPCC and other reports and reviews as if it is correct. Note that this term does not exist in astrophysics but exists in climate science.
      Its taken me some time to notice this and now I see why Steve made the whole doubling CO2 to get increased forcing of 3-4 W/m2 question one of his fundamental issues. And so I’m asking as well. Show me the derivation, Jim.

      Steve: again, I ask people not to try to debate the entire premise of climate science in one paragraph, especially on this sort of thread.

      snip – way OT

  47. Steve McIntyre
    Posted Dec 14, 2008 at 6:00 PM | Permalink

    #48. As a reviewer for AR4, I asked the authors to consider whether the paleoclimate arguments “mattered” for policy and, if they didn’t “matter”, I suggested that the entire chapter be left out if they were anxious to save space and focus on salient arguments. In such circumstances, I suggested that they insert some language saying that the uncertainties in this area appeared to be larger than previously thought and that the arguments were not essential, if that was what they thought.

    I guess that there was a “consensus” that the arguments did “matter” because they left the paleoclimate chapter in. Since IPCC took that position, I’m entitled to discuss it.

    People have sometimes said to me that, if the HS was wrong, then the situation is worse than we think, since sensitivity is greater. My answer is: well, we’d better find out if the HS is wrong then and govern ourselves accordingly. Unlike many readers, I’ve never suggested that policy-makers do nothing or wait for perfect information, though I’ve also encouraged better disclosure and data archiving, better due diligence and that policy makers would be better off if they had the benefit of “engineering quality” reports rather than little articles in Nature in Science.

    You didn’t accuse Rob Wilson of being “obsessed” with tree rings? Or Jud Partin of being “obsessed” with speleothems? They are serious about their specialities and think that they are relevant. Most scientists seem to get serious about things that don’t seem all that relevant to third parties.

    I’m interested in applied statistics. Why should such an interest attract abuse from you or anyone else?

    • Posted Dec 14, 2008 at 7:00 PM | Permalink

      Re: Steve McIntyre (#57),

      People have sometimes said to me that, if the HS was wrong, then the situation is worse than we think, since sensitivity is greater. My answer is: well, we’d better find out if the HS is wrong then and govern ourselves accordingly.

      That’s pretty much my opinion too. If the paleo is wrong we need to know that. If that means things are worse than we think then, we need to know both those things. In both cases, determining whether the paleo is right or wrong seems important to me. It’s even more important to establish this if it’s wrong.

      The same holds for whether or not models predict things correctly etc. If asking whether things are wrong is verboten, then we can’t move forward, and we risk the possibility of establishing that things may be worse than we think.

  48. Steve McIntyre
    Posted Dec 14, 2008 at 7:31 PM | Permalink

    #58. The lack of engineering quality due diligence in this field remains startling.

    One occasional commenter here (who is free to identify himself if he wishes) once ironically suggested to me that the reason in the U.S. was that politicians were not interested in proper due diligence since it would remove excuses for policy inaction, while climate modelers were not interested in proper independent due diligence since there was a risk of adverse reports and negligible benefit as long as they received handsome funding anyway.

    A more likely explanation is that people in the field have no experience with engineering quality reports and no idea what proper due diligence consists of, including the funding agencies.

    But it remains a very odd gap given all the publicity to the issues.

    • Mark T.
      Posted Dec 15, 2008 at 10:07 AM | Permalink

      Re: Steve McIntyre (#59),

      A more likely explanation is that people in the field have no experience with engineering quality reports and no idea what proper due diligence consists of, including the funding agencies.

      Given my experience working with funding agencies connected to the US government (typically DoD), I can say that at least the latter part is probably incorrect, i.e., the funding agencies do know what proper due diligence consists of. They are all subject to the same, or at least very similar, rules and regulations, and they are diligent when they need to be. Whether or not they follow up on such diligence the same way for every contract, however, is a different matter. What gets written into these contracts is an entirely different matter as well.

      Of course, I am coming from a relatively biased engineering position. Even the military contract folk I regularly work with typically have engineering degrees.


  49. Alan S. Blue
    Posted Dec 14, 2008 at 7:43 PM | Permalink

    I’m missing how a failure of the Hockey Stick implies a greater sensitivity.

    If the known LIA of Europe and northern North America was ‘more global than regional,’ then a potentially hefty chunk of the observed 2C/c warming can be sensibly attributed to a rebound from that minima. Adding in the larger-than-previously-thought effect of oceanic cycles, and there’s just not much left for carbon dioxide to contribute.

    Under “the Hockey Stick,” the assumption is that both the MWP and LIA were just regional anomalies. Prior to the Hockey Stick, that wasn’t the assumption at all.

    Steve: IT’s what people sometimes say to me. But it’s not worth the effort to argue against shadows; if someone makes a coherent presentation of the argument, then it’s worth discussing. For now, I’m merely noting the claim and awaiting an adequate exposition.

  50. Posted Dec 14, 2008 at 8:05 PM | Permalink

    I thought bulldogs giving birth to whippets was biologically impossible.

    • Craig Loehle
      Posted Dec 15, 2008 at 7:51 AM | Permalink

      Re: Max (#62), Would that be a bull-whipp?

  51. Steve McIntyre
    Posted Dec 14, 2008 at 8:10 PM | Permalink

    #62. We discussed this here with many excellent contributions. Mine in the head post was on point:

    Bull dogs have little dogs
    That feed on their ferocity
    And little dogs have lesser dogs
    And thus to animosity.

  52. BarryW
    Posted Dec 14, 2008 at 9:12 PM | Permalink

    It’s interesting how many AGW fanatics show up here and take our host’s efforts as an attempt to discredit AGW instead of an attempt to identify bad science and statistics by validating the work that has been done in a particular discipline. It seems that with dogma any attempt to correct errors of fact are met with an assumption of heresy. To keep it in a scientific realm, it would be as if they claimed that Steve was a denier of Quantum Theory when Steve had shown that experimenters that that said they had found the Higgs Boson had, in fact, made a mistake in their math!

  53. Posted Dec 14, 2008 at 10:09 PM | Permalink

    ok, ok, nice rebuttal to the troll and some good exposition of why have climate audit, the role of the hockey stick and other aspects of the meaning of life, but is anyone else still frustrated at not knowing the answer to Steve’s challenge quiz? Just as a game of Monopoly can’t end until everyone else is broke, the only way this thread ends is with the identity of Steve and Ross’ references being revealed.

  54. Terry
    Posted Dec 14, 2008 at 10:33 PM | Permalink

    #59 Steve You have touched on what the fundamental problem is in relatin to engineering quality. However it is probably more fundamental than that, notwithstanding that your comments are also valid. The modelling fraternity have a very strong belief that their models are correct, and that they truly reflect reality, because the fundamental physics are correct. It is an interesting philosophy as they tend to forget their stridency in promoting previous models when an updated version is shown to better reflect reality, and it is simply put down to “improvement and the basic physiscs remain correct”. However the models do become the modellers own interpretation of reality irrespective of the actual. I speak from personal experience in this feild. However there is also the issue of GISS, and as a scientist of 30 years experience, the advocacy of this data set as a reliable metric is nothing short of discraceful.

  55. rafa
    Posted Dec 15, 2008 at 1:40 AM | Permalink

    The difficult “reference” is pretty easy. Real Climate, isn’t it obvious? :-), sorry, just kidding.


  56. stephen richards
    Posted Dec 15, 2008 at 1:28 PM | Permalink

    snip – this sort of personal remark about a fellow poster is not permitted under blog policies, s it just leads to food fights.

  57. Sam Urbinto
    Posted Dec 15, 2008 at 1:50 PM | Permalink

    snip – please resist the temptation to argue the entire physical theory of climate from first principles in one paragraph on a thread on a another topic.

  58. Dave Roberts
    Posted Dec 15, 2008 at 2:19 PM | Permalink

    Why did you limit yourself to a single comment with a 250 word max?

    I used to write environmental permits and we had to respond to every comment. Very often commenters had multiple comments.

    Are you considering offering additional comments?

    Steve: The 250 word limit is PNAS’ not ours. They also have a 3-month time limit. Obviously 250 words is insufficient for something like Mann 2008, but it’s better than nothing. I don’t think that they accept multiple comments on the same article either. So it would have been pointless for, say, Ross and I to submit separately.

  59. Mark T.
    Posted Dec 15, 2008 at 2:42 PM | Permalink

    Thanks, Steve.

    Re: stephen richards (#71), snip – no last words

    As a follow-up to Mark T. (#69), I should add that contract specifics are key to what is required. Full expositions may not be listed as a deliverable. Round 2 of my latest contract actually required a software deliverable in spite of the fact that they can’t run the software without my system, which is not listed as a deliverable. The software we delivered is actually ad-hoc, meant to be a quick replacement for a different solution, and was not originally intended as a deliverable. As a result, it does not contain anything close to the documentation that would otherwise be required to make it work. I imagine that our next round will be different, however. It would seem that the contract specifics might also limit the scope of any FOIA filing, though I’m getting out of my realm of experience with any speculation on that.


  60. Mike B
    Posted Dec 15, 2008 at 3:09 PM | Permalink

    Okay, this is my last try at guessing references:

    Woodhouse and Brown 2001 (tree rings and droughts)

    Jolliffe 2002 (to hammer a few nails in Mannian Centering)

  61. Steve McIntyre
    Posted Dec 15, 2008 at 4:27 PM | Permalink

    #76. Mann et al 2008 didn’t use Mannian centering, so why belabor that point (with only 250 words). So no.

    And no to Woodhouse and Brown.

    per in #6 and Jeff C are the only ones even on the right track.

    • Mike B
      Posted Dec 15, 2008 at 8:52 PM | Permalink

      Re: Steve McIntyre (#77),

      Uncle. I give up. I’ve used more than 250 words guessing at references.

  62. Grammar Tony
    Posted Dec 15, 2008 at 4:41 PM | Permalink

    fewer not less

  63. Sam Urbinto
    Posted Dec 15, 2008 at 5:27 PM | Permalink

    Wegman report?
    A mathematical analysis of the divergence problem in dendroclimatology?

  64. DeWitt Payne
    Posted Dec 16, 2008 at 12:42 PM | Permalink

    von Storch?

  65. DeWitt Payne
    Posted Dec 16, 2008 at 12:47 PM | Permalink

    and a reference to the whole centered vs. non-centered PCA thing.

  66. mccall
    Posted Dec 24, 2008 at 1:03 AM | Permalink

    From the “Meatballs” department of climate journalism, there’s
    Nature freelance science writer Amanda Leigh Mascarelli, who says
    it just doesn’t matter; it just doesn’t matter …


  67. John Norris
    Posted Jan 1, 2009 at 3:37 PM | Permalink

    Not sure if the contest is still going, but I recall from your talk at Georgia Tech you said when Mann produces another hockey stick with proxies, whatever fancy statistical method he uses can be reduced to a weighting of proxies. I believe you went on to predict that the heaviest weighted proxies will be the only ones that hint at a hockey stick. Therefore you could have used the following reference in your PNAS comment:


  68. Raven
    Posted Apr 3, 2009 at 3:31 PM | Permalink


    You need to make your your PNAS response is linked somewhere. I tried google and nothing but Mann’s response shows up.


Get every new post delivered to your Inbox.

Join 3,246 other followers

%d bloggers like this: