Upside Down Tiljander in Japan

Some Japanese articles have been in the news recently. CA readers will be interested in the fact that CA was cited (thanks to a CA reader for the heads up). Here’s a graphic from their SI showing differences between Gaspé versions. As CA readers know, similar discrepancies occur for bristlecones between Ababneh and Graybill or between the Polar Urals updata and Yamal – but you can predict the version used in Team reconstructions with almost total accuracy through a very algorithm. :) Anyway,- I sort of like the look of the citation in Japanese and thought I’d share it with you.

They also circle the uptick in the upside-down Tiljander series, which we discussed here, and again it looks kinda cool in Japanese.

In this context, I thought that I’d briefly review the PNAS exchange on this topic. I reported that the Mann 2008 graphic was upside down from the orientation in the original study. So that the HS goes down in the 20th century. The original authors (Tiljander et al) discounted the 20th century portion as compromised by agriculture, ditches and bridges and so the increased varve thicknesses were not considered to be evidence of global cooling.

Upside down proxies are obviously a bad thing in CPS reconstructions (one of two legs in Mann 2008); and non-climatic contamination is a bad thing for correlation based reconstructions.

We referred to this in our PNAS comment as follows:

Their non-dendro network uses some data with the axes upside down, e.g. Korttajarvi sediments, which are also compromised by agricultural impact (Tiljander, pers. comm.)

To which Mann replied:

The claim that ‘‘upside down’ data were used is bizarre. Multivariate regression methods are insensitive to the sign of predictors. Screening, when used, employed one-sided tests only when a definite sign could be a priori reasoned on physical grounds. Potential nonclimatic influences on the Tiljander and other proxies were discussed in the SI, which showed that none of our central conclusions relied on their use.

I think that even Mann sympathizers should not accept this response. The claim that “upside down” data was used may be “bizarre”, but it’s also true. You can see that the data was used upside down by comparing Mann’s own graph with the orientation of the original article, as we did last year. In the case of the Tiljander proxies, Tiljander asserted that “a definite sign could be a priori reasoned on physical grounds” – the only problem is that their sign was opposite to the one used by Mann.

Mann says that multivariate regression methods don’t care about the orientation of the proxy. But that doesn’t solve the problem for Mann as big problems remain. There are two methods – CPS and EIV. CPS methods directly care about the orientation and the upside down data are directly used in the CPS recons. In the regression methods, the data is also used upside down. The meatgrinder picks up a spurious correlation between agricultural ditches and the proxy and assigns the wrong orientation to the series in the EIV reconstruction as well. All one needs to do is follow the series through.

Mann says “potential nonclimatic influences on the Tiljander and other proxies were discussed in the SI, which showed that none of our central conclusions relied on their use”. These are not “potential” influences; they are clearly identified as actual influences by Tiljander. The SI alludes to problems, but falls well short of providing anything like a rational explanation of why this data was used given the problems. The SI also failed to disclose that the proxies were used upside down.

At this point, there are also issues of whether the SI actually shows that “none of their central conclusions” relied on their use. One of their central conclusions was that they could “get” a stick without dendro proxies – but their non-dendro recon used upside-down Tiljander. Their SI showed that they could “get” a stick without Tiljander but, as far as I can tell, the non-Tiljander comparandum used dendro series and, in particular, relied heavily on a Graybill bristlecone. It’s a large job analyzing the impact of this sort of thing. At the time, I didn’t have a working version of Mannian EIV; one of the reasons for working through Steig RegEM in such detail was to get a handle on Mannian RegEM and I may well re-visit this matter in the near future.

Their non-dendro network uses some data with the axes upside down, e.g. Korttajarvi
sediments, which are also compromised by agricultural impact (Tiljander, pers. comm.)


39 Comments

  1. henry
    Posted Feb 26, 2009 at 9:21 AM | Permalink

    snip – piling-on

  2. Jeff Alberts
    Posted Feb 26, 2009 at 9:50 AM | Permalink

    お祝い!

    er, that’s supposed to be “Congratulations!” in Japanese (thanks to Babelfish)

    • Phil M
      Posted Oct 18, 2009 at 3:38 PM | Permalink

      Re: Jeff Alberts (#2),

      Yes, but おめでとう (omedeto) is the more usual way of saying ‘congratualtions’
      – but I would use すごい! (sugoi!) here – great/fantastic!
      (^_^)

      • Posted Oct 18, 2009 at 5:10 PM | Permalink

        Re: Phil M (#39),

        Hmm, someone’s using my name. This isn’t a pseudonym, folks. It’s the real me. I didn’t write the post at #2.

  3. NormD
    Posted Feb 26, 2009 at 10:16 AM | Permalink

    Link which we discussed >here< is not working

  4. Craig Loehle
    Posted Feb 26, 2009 at 10:21 AM | Permalink

    What part of “contaminated data” is unclear to Mann? In this case the lakes are literally contaminated in recent decades! In medical trials patients are routinely dropped who did something strange or caught some other disease.

  5. jae
    Posted Feb 26, 2009 at 10:40 AM | Permalink

    The claim that ‘‘upside down’ data were used is bizarre.

    If there is anything that is bizarre, it is this statement!

  6. Soronel Haetir
    Posted Feb 26, 2009 at 10:56 AM | Permalink

    This seems like a major shortfall of the word count restriction. If you had been given more space you could have put forward the argument with all the information to back it up. Instead you were given enough room to make the claim but not enough to back it up with all the work you have put into this subject.

  7. Steve McIntyre
    Posted Feb 26, 2009 at 11:05 AM | Permalink

    It’s only partly a word-count issue. The final point can be expressed in a few words.

    To my knowledge, no climate scientist has observed that Mann’s PNAS answers are both unresponsive and untrue. Do climate scientists even care? Doesn’t seem that way.

    • Dave Dardinger
      Posted Feb 26, 2009 at 11:29 AM | Permalink

      Re: Steve McIntyre (#7),

      There needs to be something similar to the Duel from past centuries (without the guns, of course) where you can slap Mann in the face, metaphorically, and make him answer your challenge or be considered unmanly in the eyes of his peers. Something like the Journal of Scientific Jousting.

  8. AJ
    Posted Feb 26, 2009 at 11:45 AM | Permalink

    Has anyone ever attempted to do RegEM analysis on tidal gauge data. I’m sure with a proper selection of PCs, that a hockey stick could be generated for sea level rise. Might be a ground-breaking research paper for anyone willing to do the math.

    ;>

  9. Manfred
    Posted Feb 26, 2009 at 12:08 PM | Permalink

    M.Mann said:
    “Multivariate regression methods are insensitive to the sign of predictors.”

    Then, with reference data having an up trend in 50% of cases and a down trend in 50% of cases, Mr. Mann would still compute “his” hockey-stick.
    Bizarre !

    • Craig Loehle
      Posted Feb 26, 2009 at 1:11 PM | Permalink

      Re: Manfred (#10), M.Mann said:
      “Multivariate regression methods are insensitive to the sign of predictors.”

      This is true (you can have proxies that go down when it warms) BUT in this case something NON-CLIMATIC happened during recent decades (influx of sediment). To use this data anyway one must PROVE that this confounding is not important.

    • Tolz
      Posted Feb 27, 2009 at 4:10 PM | Permalink

      Re: Manfred (#10),
      Manfred, Mann–he was blinded by the light. Wrapped up in a….
      He sure’s been blinded by something, and it ain’t science.

  10. AnonyMoose
    Posted Feb 26, 2009 at 12:40 PM | Permalink

    Is there a word missing in the phrase “very algorithm”, or do I need to look up the work of a Dr. Very?

  11. Scott Lurndal
    Posted Feb 26, 2009 at 12:46 PM | Permalink

    More japanese news here

  12. Mike86
    Posted Feb 26, 2009 at 12:47 PM | Permalink

    Just a thought, but if the sign for the entire series was reversed, then Mann would have a point. As long as the sign didn’t switch mid-stream, it really wouldn’t make any difference to the correlation or regression operation whether the proxy data was positive or negative. You’d have to know the proxy data sign to correctly interpret how the data reflected the real world, but that’s not technically what Mann appeared to comment on.

  13. Pat Frank
    Posted Feb 26, 2009 at 12:48 PM | Permalink

    So let’s see: the non-dendro recon used upside-down Tiljander, and the non upside-down Tiljander recon used Graybill bristle cones. Deploying my full scientific acuity here, it seems to me demonstrated that if one has two independent ways of cheating, one can cheat twice, independently.

  14. Sune
    Posted Feb 26, 2009 at 1:34 PM | Permalink

    In the following I use simplify equations but without loss of generality. Lets say T = temperature and x =proxy data. Then we assume that the Tiljander proxy series instructs that:
    T = a*x. Ok cool then. T = a*x is thus the only correct usage of the proxy as described by the authors of the paper.

    Then it amazes me how Mann, or anyone else, can even begin to claim that: T = – a*x is just as valid? If that is so then one must also support that 42 = – 42. Who would like to raise their hand and support this? Anyone… no?

    Or perhaps someone fell compelled to assert that an error of such gravity does not matter on the grounds that their method is so brilliantly constructed that it can correct for this any type of misuse of proxies? I would like such a magic method for my own work.

    • Dave Dardinger
      Posted Feb 26, 2009 at 1:47 PM | Permalink

      Re: Sune (#16),

      I would like such a magic method for my own work.

      In logic, of course, A and not A implies anything. The team is uncomfortably close to this sort of logic when it makes the sort of random sign selection we’ve become familiar with.

    • steven mosher
      Posted Feb 26, 2009 at 7:20 PM | Permalink

      Re: Sune (#16),

      of course 42 = -42. Like so. The meaning of life is death. death gives meaning to life. therefore life = death. but the sign is changed, which is a small immaterial change. these results are robust.

      professional sophist.

      • Greg F
        Posted Feb 26, 2009 at 7:34 PM | Permalink

        Re: steven mosher (#25),

        The meaning of life is death. death gives meaning to life. therefore life = death. but the sign is changed, which is a small immaterial change. these results are robust.

        The death part is certainly robust.

      • bernie
        Posted Feb 27, 2009 at 7:20 AM | Permalink

        Re: steven mosher (#26), And, left is right in the hall of mirrors.

  15. AJ Abrams
    Posted Feb 26, 2009 at 1:44 PM | Permalink

    |42| = |-42| See that’s all that is needed Sune.

  16. Vernon
    Posted Feb 26, 2009 at 1:51 PM | Permalink

    Sune,

    Your missing the point. The proxy can have an direct or inverse relationship to the temperature but it cannot change. The proxy may go down at temperature goes up but it cannot go down during one period and up during another – which I understand Mann does.

    The other issue of using proxies that the original author says are contaiminated is beyond bad practices unless you can show that what your are using the proxy for is not what is contaiminated, which Mann does not do.

  17. pat
    Posted Feb 26, 2009 at 2:06 PM | Permalink

    Sune #16
    There’s no problem with ’42=-42′. All that is required is the application of the Mannian AVF or Absolute Value Function. Works like magic.

  18. Sune
    Posted Feb 26, 2009 at 3:28 PM | Permalink

    Vernon
    I wish I was missing the point. Unfortunately I’m not. The contamination issue is one thing. Using a proxy by suddenly inverting its relation to temperature to justify a warming trend is another. The issue still stands:

    The proxy relation to climate can only follow T = a*x for x = [1000 to 1800]. Contamination is after x > 1800.

    Mann does two wrong things: 1) He inverts the proxy relation setting T = – a*x and then 2) uses the contaminated, now inversed upward-pointing hokey-stick proxy to claim there is a warming trend. Its upside down and its the data has stopped being valid for the 20th century!

    That is not me missing the point – its the Team missing the point. In fact it is worst than 42 = -42 as I said last time. Mann can get: 42 = -17… even more wrong. Brilliant.
    All the best

  19. KlausB
    Posted Feb 26, 2009 at 4:05 PM | Permalink

    Domo arigato, Steve-san

  20. Vernon
    Posted Feb 26, 2009 at 6:19 PM | Permalink

    Sune,

    Sorry did not realize the team was worse than I expected.

  21. RomanM
    Posted Feb 26, 2009 at 7:34 PM | Permalink

    Mosh, you’ve been drinking too much from the climate science cup of knowledge! ;)

  22. Posted Feb 26, 2009 at 8:48 PM | Permalink

    すごい!

  23. pjm
    Posted Feb 27, 2009 at 5:13 AM | Permalink

    Doesn’t Tiljander’s temperature reconstruction give local temperatures, whereas Mann is producing a global temperature reconstruction. Obviously one rises as the other falls. So … we will know AGW has finally taken the Earth past a tipping point when Tiljander’s lake freezes over.

  24. dearieme
    Posted Feb 27, 2009 at 8:38 AM | Permalink

    “an error of such gravity does not matter”:
    my dear sir, you have surely heard of anti-gravity? And, indeed, anti-matter?

  25. Posted Feb 27, 2009 at 11:17 AM | Permalink

    foxy hokey proxy stopsy maxy vexy crax

  26. Paul Penrose
    Posted Feb 27, 2009 at 12:49 PM | Permalink

    I have no idea what cup Lucy has been drinking from, but I want some of that!

  27. aurbo
    Posted Mar 3, 2009 at 9:04 PM | Permalink

    Point of interest. Can anyone confirm whether or not Ababneh is in the witness protection program? Why no recent papers?

  28. Henry
    Posted Nov 3, 2009 at 1:40 AM | Permalink

    almost total accuracy through a very algorithm.

    Word missing?

5 Trackbacks

  1. [...] Steve writes: Mann didn’t just use one Tiljander series upside down; he used all four of them upside down, a point illustrated in the graphic below from a Japanese language article that rather appealed to me. [...]

  2. [...] Steve writes: Mann didn’t just use one Tiljander series upside down; he used all four of them upside down, a point illustrated in the graphic below from a Japanese language article that rather appealed to me. [...]

  3. [...] upside down, a point illustrated in the graphic below from a Japanese language article that rather appealed to me. Figure 3. Excerpt from Itoh graphic identifying upside down Tiljander [...]

  4. [...] Upside Down Tiljander in Japan [...]

  5. [...] comes this press release that made me recoil when I read the methodology involved because it is Mannian-Tiljanderish in the approach. Instead of doing top down heating (as occurs in Nature), they do bottom up heating [...]

Follow

Get every new post delivered to your Inbox.

Join 3,262 other followers

%d bloggers like this: