Hide the Decline – the Other Deletion

I recently re-visited an article in Science (Briffa and Osborn 1999), that, together with Jones et al 1999 (Rev Geophys), were the first bites of the poison apple of hide-the-decline. I observed that key conclusions in Briffa and Osborn 1999 depended on the rhetorical effect of deleting the decline from their spaghetti graph.

I’ve been looking at the subsequent development of this graphic and, in the process, noticed another curious feature of the figure in Briffa and Osborn 1999 – shown below.

As noted previously, Briffa data was deleted after 1960. (Smoothing was done after the deletion further accentuating the impact of the deletion of post-1960 data.)

In addition, and this point has not been previously discussed, Briffa and Osborn did not show data prior to 1550 for the Briffa MXD reconstruction. I’d previously noticed that an archive for Jones et al 1998 (surprisingly) contained a Briffa version that is linearly related to the Science graphic – the match is shown as a dotted line. (The basis of the linear relationship is not reported and not known to me at present, but can nonetheless be used empirically to show the extensions.)

In the graphic below, I’ve shown (in magenta) not just the hide-the-decline extension, but the deleted data prior to 1550. Take a look. (Update note- see yesterday’s post for provenance.)


Figure 1. From Briffa and Osborn (Science 1999), also with deleted MXD data shown in magenta.

Obviously, the deletion of the pre-1550 portion of the Briffa reconstruction version also makes a large contribution to the rhetorical impression of coherence between reconstructions. Ironically, Briffa and Osborn observe in their running text:

An uninformed reader would be forgiven for interpreting the similarity between the 1000-year temperature curve of Mann et al. and a variety of others also representing either temperature change over the NH as a whole or a large part of it (see the figure) as strong corroboration of their general validity, and, to some extent, this may well be so.

Needless to say, one of the reasons for the reader being “uninformed” is the deletion of adverse data (both before 1550 and after 1960) to give the impression of “corroboration” of the “general validity” of the reconstructions. Note that smoothing after deletion enhances the impact of the deletion – look at the strong divergence pre-1550 as well as post 1960.

The provenance of the Briffa reconstruction was tersely described only as follows ( in the caption): “NH tree-ring densities [1550-1960, from (3 – Briffa et al 1998(Nature); Briffa et al 1998 (Proc Roy Soc London)), processed to retain low-frequency signals]”. Appendix A of Briffa et al 2001 contains some useful further information, showing that this version arose from a composite obtained from averaging MXD chronologies in the large Schweingruber network (scaled over 1700-1994, then opportunistically calibrated over 1880-1960):

In Figure 4, we show different temperature reconstructions of the NH (all land area temperatures north of 20N), each based on a different way of using MXD predictor data. One curve was produced by performing the age-banding procedure on all chronologies in the data set and by using an unweighted mean of all banded series from all locations. This is similar to the curve from 1650-1960 [sic – actually 1550-1960] presented by Briffa and Osborn [1999] (although we have since made very minor modifications to the age-banding procedure and the input data set.) All other curves in Figure 4 were obtained by prior averaging of the age-banded density series into the nine subregions (as defined by Figure 1.)

The next figure shows the Briffa et al 2001 (JGR) spaghetti graph – showing the Briffa 2001 version in limegreen (together with the decline, expurgated in the original figure) and the Briffa-Osborn 1999 version in magenta (including both pre-1550 and post-1960 data):


Figure 2. From Briffa et al 2001 (JGR) Plate 3, also showing the pre-deletion data used in Briffa-Osborn 1999 (magenta). The Briffa 2001 version ends in 1960 and has been extended using data from Climategate emails (showing the decline).

The changes from the 1999 Briffa version to the 2001 Briffa version are instructive for several reasons.

Briffa et al 2001 uses virtually the same population of sites as Briffa and Osborn 1999. The B2001 population was 387 sites, while the Briffa et al 1998 (Nature 393) population (cited in BO99) was 383 sites – immaterially different. The Briffa et al 2001 site count was 19 sites in 1550, 8 in 1500 and only 2 in 1402, but there were enough for Briffa to report a reconstruction. (Readers should bear in mind that the Jones reconstruction, for example, was based on only 3 proxies in the 11th century, one of which was a Briffa tree ring site with only 3-4 cores, well under standard requirements.)

So why were pre-1550 values shown in Briffa et al 2001 and not in Briffa and Osborn 1999? The only reason that I can deduce is that the Briffa 2001 reconstruction had a rhetorical similarity to the Mann and Jones reconstructions in the 1400-1550 period – and therefore was shown, while the Briffa and Osborn 1999 version showed a major discrepancy – and was therefore not shown.

How did Briffa get from the 1999 reconstruction version (magenta) to the 2001 version (limegreen)?

This is an interesting exercise that I’ll describe in more detail on another occasion. But I can’t resist a quick preview. The Briffa-Osborn 1999 version was based on averages of all available sites – a sensible enough procedure. Changes in the Briffa 2001 methodology include the calculation of regional averages followed by stepwise principal components. (The methodological description in Briffa et al 2001 is very sketchy and, unfortunately, the Climategate computer dossier didn’t include Briffa’s programs.)

The effect of using principal components on regional averages is to change the weights for individual sites, including the possibility of negative weights i.e. flipping the regional MXD series. In particular, the closing uptick in the Briffa 2001 reconstruction may well depend on the flipping of data – a point that I’ll try to examine in the future.


77 Comments

  1. Steve Garcia
    Posted Mar 21, 2011 at 12:47 PM | Permalink

    Using the term “cherry picking” is almost inescapable, it seems. Steve is showing a prima facie case for it, in any event – which puts the onus on Briffa to show that this was not done.

    …Since Steve doesn’t specifically point it out, I would note that both the post-1960 and the MWP seem to show exactly the opposite of the current paradigm about tree rings being a good proxy for temperature.

    snip

    Steve: it is an editorial policy of this site that readers refrain from efforts to prove or disprove “big picture” points in a few paragraphs.

    And that readers refrain from arguing things like tree rings at every opportunity. Please stick to specific issues arising out of BO99 and B2001.

  2. Posted Mar 21, 2011 at 12:47 PM | Permalink

    Amazing … it just never ends …

  3. Steeptown
    Posted Mar 21, 2011 at 1:01 PM | Permalink

    Cherry picking; what cherry picking?

    • Dave L.
      Posted Mar 21, 2011 at 1:38 PM | Permalink

      I don’t think this is cherry picking in the usual sense; i.e., selecting only proxies with a desired pattern to use in the temperature reconstruction (such as bristle cones). Rather this is “hiding data” or “data manipulation”; i.e., selecting only a portion of a proxy which has a desired pattern to use in the temperature reconstruction while omitting the remainder which falls outside the desired pattern. The former can be criticized on ethical grounds whereas the latter borders on overt deceit since it is intentionally hidden. Neither is acceptable in the scientific method.

  4. mpaul
    Posted Mar 21, 2011 at 1:10 PM | Permalink

    It should be pointed out that the removal of the pre-1550s data is logically consistent with the Team’s stated approach to temperature reconstructions. The explanation that Jones uses to justify removing the adverse proxy data after 1960 was:

    One of the three temperature reconstructions was based entirely on a particular set of tree-ring data that shows a strong correlation with temperature from the 19th century through to the mid-20th century, but does not show a realistic trend of temperature after 1960.[…] It is because of this trend in these tree-ring data that we know does not represent temperature change that I only show this series up to 1960 in the WMO Statement.

    So logically if the pre-1550’s data also did not represent a “realistic trend” then the Team would be acting in a perfectly consist way by removing it. Generalizing the approach, any data that was deemed ‘unrealistic’ could be removed. In the extreme, individual data points could be removed if they were deemed to be ‘unrealistic’ and replaced with more ‘realistic’ ones.

    The scientific establishment has not seemed at all troubled by this line of thinking.

    • Hu McCulloch
      Posted Mar 21, 2011 at 2:34 PM | Permalink

      The scientific establishment has not seemed at all troubled by this line of thinking.

      Please don’t tar all scientists, or even all climate scientists, by identifying them with the likes of Jones, Briffa and Mann. See, eg the video by Dr. Richard Muller at http://wattsupwiththat.com/2011/03/18/you%e2%80%99re-not-allowed-to-do-this-in-science/

      • mpaul
        Posted Mar 21, 2011 at 7:59 PM | Permalink

        I think Muller is doing what all professionals should do when faced with this sort of situation. Its essential to the proper functioning of science that scientists self-police themselves. But, unfortunately, Muller is a lone voice in the woods. The scientific societies and the academies have largely circled the wagons.

        • Keith W.
          Posted Mar 22, 2011 at 12:36 PM | Permalink

          That, or they do not know that this is such a consistent and wide spread problem. Judith Curry recently had three posts concerning “Hide the Decline” where she repudiated the process. As the light is shown on the less than stellar practices of the few, the more the majority has to react. When the evidence is obviously effected by the decisions to leave outliers out of the system, when the scientists involved cannot explain the outliers, then people are going to start asking why did you make that choice.

        • Posted Mar 22, 2011 at 2:14 PM | Permalink

          Re: Keith W. (Mar 22 12:36),

          Physicist Jonathon Jones, Oxford University comments on “Hide the Decline”:

          “This is not a complicated technical matter on which reasonable people can disagree: it is a straightforward and blatant breach of the fundamental principles of honesty and self-criticism that lie at the heart of all true science.

          The significance of the divergence problem is immediately obvious, and seeking to hide it is quite simply wrong.”

          http://wattsupwiththat.com/2011/02/25/currys-2000-comment-question-can-anyone-defend-%E2%80%9Chide-the-decline%E2%80%9D/

          Hopefully, more will follow these honest scientists.

          Cheers — Pete Tillman
          Consulting Geologist, Arizona and New Mexico (USA)

  5. j ferguson
    Posted Mar 21, 2011 at 1:37 PM | Permalink

    Hoping this question is not outside the confines of this post, has anyone seen any thoughtful discussion among the dendrochronologists in print or in the emails about why the excluded trends seem to vary from the “known” or “strongly suspected” historical trends? I would have thought this variation to be of great interest.

    • Hu McCulloch
      Posted Mar 21, 2011 at 2:29 PM | Permalink

      Your question is indeed off the topic of this post, but you might try asking it over at Real Climate to see what they have to say. (Odds are good your impertinent question will never appear there.)

      • j ferguson
        Posted Mar 21, 2011 at 4:00 PM | Permalink

        Hu,
        I somehow thought asking this sort of question at Real Climate would be a waste of time. Be it impertinent, it still seemed something that should have provoked at least some reflection of the part of the guys who do this sort of thing.

        “The bark that didn’t dog them” for some reason.

        sorry.

        • amac78
          Posted Mar 21, 2011 at 5:49 PM | Permalink

          Re: j ferguson (Mar 21 16:00),

          > …asking this sort of question at Real Climate would be a waste of time.

          Keep a local copy and cross-post. If clearly phrased and polite, such questions sometimes pass moderation, and occasionally prompt informative responses. When this happens, the science is advanced.

        • Henry A
          Posted Mar 22, 2011 at 6:56 AM | Permalink

          Should he also bow three times towards UAE, sit in a lotus position and chant to himself green thoughts before pressing enter.

          When will the Hockey Team final just declare themselves a religion?

          Henry

        • amac78
          Posted Mar 22, 2011 at 9:14 AM | Permalink

          Re: Henry A (Mar 22 06:56),

          > Should he also bow three times towards UAE, sit in a lotus position and chant to himself green thoughts before pressing enter.

          Heh. I’d suggest “no.” He could pose that inquiry to Nicholas Nierenberg, who extracted some key hidden-in-plain-sight information on Mann ’08 using this strategy.

        • j ferguson
          Posted Mar 23, 2011 at 6:53 AM | Permalink

          Henry, Could that be UEA you meant? It’s in a bit different direction from South Florida than UAE which might explain the lack of effect and thus my reluctance to press “enter.”

    • Clark
      Posted Mar 22, 2011 at 2:34 PM | Permalink

      You can check through the many older posts here about tree-ring networks, and the choice of which data sets to use from Yamal and other areas of the Russian north (c.f. climateaudit.org/2009/09/30/yamal-the-forest-and-the-trees/).

      An interesting aspect I learned from all of those posts is that some of the key species can grow on different forms – a bushier, low-to-the-ground form and a primary axis more-like a regular-tree growth. The change from one to the other can produce huge changes in ring width that has more to do with changing growth habits than temperature. (sorry to go OT but those are very useful posts for anyone trying to get up to speed)

  6. Peter S
    Posted Mar 21, 2011 at 1:41 PM | Permalink

    Hide the Precline.

    • ad
      Posted Mar 21, 2011 at 3:51 PM | Permalink

      Rather, Hide the incline, I should think.

      • Hu McCulloch
        Posted Mar 28, 2011 at 10:52 AM | Permalink

        Well put!

  7. Posted Mar 21, 2011 at 1:54 PM | Permalink

    Perhaps they should have hidden the middle as well.

  8. Posted Mar 21, 2011 at 2:45 PM | Permalink

    Steve’s observations here are more evidence that the error bars around the proxy-reconstructions are much wider than the published estimates — a point that Judith Curry has been making, and that McShayne & Wyner demonstrated pretty convincingly in their statistical review of the proxy data.

    Which would seem to leave the published proxy temp reconstructions in limbo — the temperature resolution of the reconstructions appears too coarse to shed much light on the details of temperature record in the past millenium or two.

  9. S. Geiger
    Posted Mar 21, 2011 at 2:54 PM | Permalink

    curious if when the reconstructions are altered in these ways if its ever the case where data from a particular tree is ‘used’ in some timeframes but ‘not used’ in others(not being sure how the data is grouped and analyzed)? I think it would be interesting to hear theories on how a particular tree could fall out (or in?) calibration with the temp field over its lifespan.

  10. RayG
    Posted Mar 21, 2011 at 3:14 PM | Permalink

    Perhaps the dendropaleoclimate folks should be extracting their core samples from prunus cerasus instead of bristlecones and Yamal larch. It would make their cherry picking task easier.

  11. daniel
    Posted Mar 21, 2011 at 3:36 PM | Permalink

    another straw in the back of the British so-called enquiries (Oxburgh, Muir Russell) !

  12. Free Trade
    Posted Mar 21, 2011 at 3:55 PM | Permalink

    The deletion of relevant data from the analysis, without an assignable cause, is simply astonishing. I cannot believe that scientists would be handling their data in such a sloppy and haphazard fashion, and their analysis is given serious consideration in policy decisions.

  13. Alexander K
    Posted Mar 21, 2011 at 3:58 PM | Permalink

    As a layman, The Team have chosen an incredibly silly way to tell porkies. But I’m only a layman, so I could be mistaken.

  14. Jimmy Haigh
    Posted Mar 21, 2011 at 4:23 PM | Permalink

    If you turn all of Briffa 1999 upside down it correlates better…

  15. andy
    Posted Mar 21, 2011 at 4:32 PM | Permalink

    It seems there is another ‘contender’ for the most disconcerting graphic on climate audit.

    Someone tell Delingpole.

  16. andy
    Posted Mar 21, 2011 at 5:21 PM | Permalink

    Or possibly tell Bishop Hill, he’s writing a new book.

    Really that graph is damning.

  17. dearieme
    Posted Mar 21, 2011 at 5:29 PM | Permalink

    Why magenta? Shouldn’t it have been scarlet?

  18. mikemUK
    Posted Mar 21, 2011 at 5:38 PM | Permalink

    “The only reason that I can deduce is that the Briffa 2001 reconstruction had a rhetorical similarity to the Mann and Jones reconstructions . . . etc”

    That being so, and assuming Briffa and Osborn didn’t suddenly go off at a completely different tangent later in 2001, what on earth should one make of Mann’s hysterical(for a scientist) attack on them only a very few months later, evident in the series of Climategate Emails, between 22 March 2002 – 15 April 2002?

    (1016818778.txt – 1018893474.txt group under keyword ‘Briffa’, at ‘eastangliaemails.com’)

  19. Rob R
    Posted Mar 21, 2011 at 5:46 PM | Permalink

    Is any part of that Briffa “curve” actually real?

    • SteveGinIL
      Posted Mar 26, 2011 at 4:30 PM | Permalink

      Not an invalid question to ask, all in all. A correlation over a portion of two curves (or perhaps multiple disconnected portions, even) but not elsewhere along them usually is seen as mere coincidence.

  20. Al
    Posted Mar 21, 2011 at 6:09 PM | Permalink

    Obliviously Briffa knows nothing of the correlation between temp and tree rings and is blindly applying some correlation…

  21. Posted Mar 21, 2011 at 6:31 PM | Permalink

    I see no magenta style plot on these papers – can you direct me to the correct one please?

    Seeing the Wood from the Trees
    Keith R. Briffa and Timothy J. Osborn*

    http://www.sciencemag.org/content/284/5416/926.full?ijkey=c5d71e39f7f85ed34e518d0ce7473549cc903585&keytype2=tf_ipsecsha

    High-resolution palaeoclimatic records for the last millennium: interpretation, integration and comparison with General Circulation Model control-run temperatures
    P. D. Jones, K. R. Briffa, T. P. Barnett and S. F. B. Tett The Holocene 1998

    http://hol.sagepub.com/content/8/4/455.full.pdf

    Low frequencty temperature variations from a northern tree ring density network
    Briffa et al 2001

    http://www.cru.uea.ac.uk/cru/people/briffa/jgr2001/Briffa2001.pdf

    Steve: yup, and why wasn’t the magenta portion in the graphics – that’s the question.

    • Posted Mar 21, 2011 at 11:59 PM | Permalink

      But you show it as originating in Briffa and Osborne 1999. This is clearly not the case.
      So you have obtained and not referenced additional data from some where. Care to say where. Thanks

      Also Briffa 2001 shows in fig 1 that the growth is variable over the life of a tree has this been applied to your data? And in same paper Fig 4 and 6 shows the decline fairly explicitly.

      Steve – the derivation of the deleted data is described in meticulous detail in yesterday’s post,http://climateaudit.org/2011/03/17/hide-the-decline-sciencemag/ the first part of Hide the Decline- Sciencemag. The post includes scripts generating the figures, with the scripts proving the provenance of the data, and thereby establishing what Briffa didn’t show.

      For Briffa et al 2001, the additional data is obtained from a Climategate email – see discussion in Dec 2009.

      As I’ve noted elsewhere on a number of occasions, Briffa showed the decline only in standalone graphics. However, the decline was deleted in spaghetti graphs comparing to other reconstructions, thereby giving false rhetorical coherence – which in turn was used to justify untrue statements in the running text, as shown in these posts.

  22. Posted Mar 21, 2011 at 8:22 PM | Permalink

    Maybe this is why Briffa has been so quiet since the CRU leak. Hoping that keeping his head below the parapet of the castle in the clouds would keep those pesky arrows at bay.

  23. Geoff Sherrington
    Posted Mar 21, 2011 at 11:22 PM | Permalink

    How does this tie in with the comment –
    1123513957.txt (which does not come up on the search engine, so it’s here in full). Refer particularly to the sentence “there is a period around 1400 AD when the proxy records we’ve used in this MWP figure do indicate a warm period – and all records show positive anomalies at the same time.” The magenta curve of 1999 does not fit this description.
    ……………………………….
    From: Jonathan Overpeck
    To: Tim Osborn
    Subject: Re: MWP figure
    Date: Mon, 8 Aug 2005 11:12:37 -0600
    Cc: Keith Briffa , Eystein Jansen ,

    Hi Tim – Decisions, decisions… thanks so much
    for taking the initiative. I think – for the
    reason you state, we should go for the one that
    includes the 20th century. We make clear that
    these are not reconstructed temp, but normalized
    anomalies – this keeps us out of some trouble.
    But, I think the main message is that we’re
    looking at this issue from every angle. And,
    we’re letting others see the issue from every
    angle. It adds punch.

    this means that the MWP box needs to talk about
    the period around 1400 – can you make sure that’s
    on Keith’s radar screen. I believe that
    historians talk about the Medieval Period going
    to at least 1450, so what the heck…

    I you can adjust the caption to work, and then
    send both it and the final fig to Øyvind, me and
    Eystein that would be good – make sure Keith is
    ok with it all first, too.

    Thanks Tim! Best, Peck

    >Hi Peck,
    >
    >there is a period around 1400 AD when the proxy
    >records we’ve used in this MWP figure do
    >indicate a warm period – and all records show
    >positive anomalies at the same time. Thus it
    >couldn’t/shouldn’t be dismissed in the same way
    >as the MWP, as a period of disparate regional
    >behaviour, albeit with more records showing
    >warming than cooling. For 1400, all indicate
    >warming but with smaller magnitude than the 20th
    >century. If the figure were extended to cover
    >the 15th century, then it would also seem
    >necessary to extend it to the present so that
    >the 1400 period could be compared with the 20th
    >century.
    >
    >I’ve attached 3 versions of the figure.
    >
    >850-1350 as originally sent.
    >
    >850-1500 showing warm anomaly in 1400, but
    >cannot tell how warm relative to present-day.
    >
    >850-2000 showing 1400 was not as anomalous as present-day.
    >
    >Take your pick, Peck!
    >
    >Cheers
    >
    >Tim and Keith
    >
    >At 21:57 05/08/2005, Jonathan Overpeck wrote:
    >>Hi Tim and Keith – Hope you’re not going to
    >>kill me, but I was talking with Susan Solomon
    >>today, and she impressed me with the need to
    >>make several points if we can.
    >>
    >>One issue (other to come in a subsequent email)
    >>is whether we can extend the MWP box figure to
    >>include the 15th century. I don’t read the
    >>blogs that regularly, but I guess the skeptics
    >>are making hay of their being a global warm
    >>event around 1450AD. I agree w/ Susan that it
    >>is our obligation to weigh in on issues like
    >>this, so…. can we extend the fig to extend up
    >>to 1500AD?
    >>
    >>Sorry about this, Tim. Of course we need it yesterday.
    >>
    >>Thanks x10**6
    >>
    >>best, peck
    >>
    >>>Dear Eystein, Peck and Keith,
    >>>
    >>>I spotted a minor error in the MWP figure
    >>>(reference period was 1001-2000 but should
    >>>have been 1001-1980 because some series stop
    >>>in 1980) and a typo in the legend, so here is
    >>>a revised MWP figure with these things
    >>>corrected and a slight adjustment to line
    >>>thicknesses and font sizes.
    >>>
    >>>As before I’ve included .ps, .pdf and .gif
    >>>versions because I’m not sure what you prefer.
    >>>
    >>>I’ve also drafted a caption – see attached
    >>>.doc file. Feel free to modify as necessary.
    >>>I think it covers the necessary details
    >>>including normalisation period, but perhaps it
    >>>is a bit “wordy” and unnecessarily repeats
    >>>things already in the MWB box text?
    >>>
    >>>I’m still working on SH figure/caption.
    >>>
    >>>Cheers
    >>>
    >>>Tim
    >
    >
    >Attachment converted: Macintosh HD:mwpbox8502000.pdf (PDF /«IC») (00091133)
    >Attachment converted: Macintosh HD:mwpbox8501500.pdf (PDF /«IC») (00091134)
    >Attachment converted: Macintosh HD:ipccar4_mwpbox 2.pdf (PDF /«IC») (00091135)
    >Dr Timothy J Osborn
    >Climatic Research Unit
    >School of Environmental Sciences, University of East Anglia
    >Norwich NR4 7TJ, UK
    >
    >e-mail: t.osborn@uea.ac.uk
    >phone: +44 1603 592089
    >fax: +44 1603 507784
    >web: http://www.cru.uea.ac.uk/~timo/
    >sunclock: http://www.cru.uea.ac.uk/~timo/sunclock.htm


    Jonathan T. Overpeck

  24. peetee
    Posted Mar 22, 2011 at 12:37 AM | Permalink

    yawn – never-ending audit

    • geronimo
      Posted Mar 22, 2011 at 6:46 AM | Permalink

      Your point is? Are you forced to read these posts or something? What’s going on here is what should be going on in the climate science world, a dissection of the published results to ascertain if they hold water. At first blush this shows that the team had temporary blockages on the holes in their buckets during the preparation of these papers. If that’s of no interest to you then fine, but why waste your energy reading about it? These papers are being used to persuade politicians that the recent warming is unprececented so said politicians will put in place policies that suit some people’s agendas. It is vitally important that they are as robust a statement of the science as they can be, so they should be dissected and understood and, if necessary challenged.

      • peetee
        Posted Mar 23, 2011 at 3:03 AM | Permalink

        my humblest of apologies… meant to say, *yawn* – never-ending (most) selective audit! Just how long can one milk this puppy, anyway?

        • Posted Mar 24, 2011 at 2:22 PM | Permalink

          As long as milk keeps pouring out of it.

          How long will it take you to start caring?

    • oneuniverse
      Posted Mar 22, 2011 at 9:37 AM | Permalink

      yawn – never-ending audit

      The trouble is, the paleoclimate authors left so many ‘easter egg features’ for Steve McIntyre & other scientists to discover that it was always going to take a while.

    • Hu McCulloch
      Posted Mar 22, 2011 at 10:52 AM | Permalink

      peetee
      yawn – never-ending audit

      Bernie Madoff’s sentiment, precisely!

    • stan
      Posted Mar 22, 2011 at 11:10 AM | Permalink

      Auditing is worthless. It doesn’t advance the career of the auditor. It doesn’t bring in any research cash. It doesn’t garner the covers of prestigious journals. It doesn’t lead to interviews by big, mainstream news media outlets.

      In fact, it is actually worse than worthless! It’s a huge negative. It creates headaches for the real scientists because it makes them look bad. It pokes holes in pet theories. It exposes important errors. It creates confusion among the public. It reveals that lots of very prominent people are incompetent. All this is bad, very bad.

      Auditing is a terrible, terrible waste. No real scientist today would ever have anything to do with it.

      • John Tofflemire
        Posted Mar 23, 2011 at 12:22 AM | Permalink

        Here, here!

      • Posted Mar 23, 2011 at 12:58 PM | Permalink

        In essence, the BEST effort is an audit. Muller thought the UEA/Mann efforts were suspicious, and he decided he had the people and wherewithal to double-check their “work.” (I dare not call it “science.”)

        Steve:
        the Berkeley exercise has almost nothing to do with the Climategate emails. They were 99.9% about the proxy reconstructions. The temperature records were barely mentioned. My own concern with the temperature records pertains more to the handling of UHI than homogenization, but has not been a primary concern at this site. I am dubious of the new Menne algorithm as being able to account for UHI and have yet to see evidence that the Berkeley methods are addressing this issue.

        • WA777
          Posted Mar 23, 2011 at 5:44 PM | Permalink

          Re: feet2thefire (Mar 23 12:58),

          If I recall correctly, the purpose of a financial audit is to review procedures to assure they will produce a reasonable answer if followed (the books), and to do appropriate tests to assure that the procedures were, in fact, followed.

          Weaknesses are identified. Recommendations for change may follow (unless the auditor resigns).

          Correction of weaknesses in procedure and record keeping (data) revealed by these audits apply to the Berkeley exercise, as does the need for documentation, data and changes.

        • SteveGinIL
          Posted Mar 24, 2011 at 6:49 AM | Permalink

          Steve M, given what you say here about it, is there some instrinsic reason you have not put any effort into the UHI? Now that you mention it, I don’t recall your having had anything to say about UHI.

          Will you get into that at any point?

          I agree, it is a major factor, especially when some recent work shows the UHI at some sites is as high or higher than their entire graph range as things are reported now. UHI is certainly larger than the 0.05C indicated by Wang’s Chinese study that Jones coauthored – but how much remains a mystery.

  25. EdeF
    Posted Mar 22, 2011 at 1:18 AM | Permalink

    I don’t see why Magenta and Lime-Green can be summarily banished; they are
    in phase for much of the time, both show a marked decline from the 1930s
    peak down to about 1970. This matches standard NH temp swings during that
    time. If the number of samples is good (large) I don’t see why they would
    be excluded. By the way, I now have a craving for rainbow sherbet.

  26. David S
    Posted Mar 22, 2011 at 4:35 AM | Permalink

    As the Cure said in the aptly named “Wrong Number”:

    Lime green, lime green and tangerine
    The sickly sweet colors of the snakes I’m seeing

  27. stacey
    Posted Mar 22, 2011 at 7:38 AM | Permalink

    If the buidling is constructed on sand and the roof is leaking why would anyone reasonable person think that the bit in the middle has been properly constructed?

  28. BRIAN M FLYNN
    Posted Mar 22, 2011 at 8:40 AM | Permalink

    It appears the Briffa-Osborn 1999 version (with both pre-1550 and post-1960 data) would poke holes in the uncertainty margins and in the theory of a medieval warming period.

  29. Robbo
    Posted Mar 22, 2011 at 9:22 AM | Permalink

    A most significant find. It brought me to recall http://climateaudit.org/2010/03/31/tricking-the-committee/ where Steve quoted the Committee as finding, based on a denial from UEA, that:

    ‘CRU never sought to disguise this specific type of tree-ring “decline or divergence”. On the contrary, CRU has published a number of pioneering articles that illustrate, suggest reasons for, and discuss the implications of this interesting phenomenon.’

    It seems to me there are two important issues which CRU need to explain, one with regard to hiding the decline in the peer-reviewed literature, the second with regard to the denial to the committee.

  30. Jeremy
    Posted Mar 22, 2011 at 11:17 AM | Permalink

    This makes their “convergence period” look all the more suspicious, not that that hasn’t been spoken of before either.

    How is it that such separate and “independent” datasets all converge so nicely in the early 1900s? If they were dealing with real noise which exists in all real data, there would be no such convergence. The removal of any divergence before after this period where all plots share sign, slope and offset can only have one purpose, and that is to hide the arbitrary nature of where the convergence exists.

  31. j ferguson
    Posted Mar 22, 2011 at 11:23 AM | Permalink

    Robbo:

    From an inquiry:
    “” …..CRU has published a number of pioneering articles that illustrate, suggest reasons for, and discuss the implications of this interesting phenomenon.’”

    How would one find these articles?

    • Publius
      Posted Mar 26, 2011 at 12:54 PM | Permalink

      “Suggest reasons for” translates to an unknown anthopogenic effect. A couple of probing questions from the “investigators” would have ascertained that the Team really has no clue and is merely speculating. But alas there appears to have been no interest in following this line of inquiry.

  32. Posted Mar 22, 2011 at 11:27 AM | Permalink

    This form of data manipulation exhibits how leverage can cause misrepresentation in something as important as this. In an extremely general sense, I’m all for statistical data when referring to something as natural as this. Yet, somehow this so called “cherry picking” or selective data has ruined the reputation of the data collectors and publishers. Thank you so much for the enlightening post and comments.

  33. Robbo
    Posted Mar 22, 2011 at 12:09 PM | Permalink

    @ j ferguson:

    Link to the report

    http://www.publications.parliament.uk/pa/cm200910/cmselect/cmsctech/387b/38713.htm

    Excerpt with transcribed testimony:

    Q121  Dr Harris: Okay. Moving on to something else, there is this whole “hide the decline” business I want to talk to you about. There was a concession from at least one set of critics that the “trick” is probably not an issue because they recognise that it is a term used.
      Professor Jones: It is the best way of doing it.
      Q122  Dr Harris: It may not be the view of all of your critics but at least you have some on the record saying that that is now not the issue. But then you will recall there was an exchange I had, if you were listening, with them about this question of hiding the decline and I just wanted you to respond to their assertion that when you did that it was not set out in the publications—I must say I have not gone back to the publications to read them so I am relying on your view on this but I am sure it can be done—and that in fact it was never shown that this was going on. Whereas your evidence from the UEA says very clearly that this is part of the published scientific record that you were doing it and the reasons you were doing that, and that can be criticised or agreed with by other scientists. Can you just talk about that?
      Professor Jones: That particular email relates to this document that I produced for the World Meteorological Organisation at the end of the last millennium in 1999. One of the curves was based on tree ring data which showed a very good relationship between the tree rings and the temperature from the latter part of the nineteenth century through to 1960, and after that there was a divergence where the trees did not go up as much as the real temperatures had. We knew that because we had written a paper the year before in 1998 in the journal Nature which discussed this divergence between tree growth and temperatures in recent times. Not all tree ring series show that but this particular one we knew did, so we knew that putting the tree ring series in from 1960 onwards would be wrong because it does not agree with the instrumental temperature. What we did for this simplified diagram was to put the instrumental data on the end from 1960, so that only applies to one of these curves on this cover. We had written about it the year before, in one of the first papers on the divergence problem—I think other groups had actually called it the divergence problem—and, since then, we have been working with other tree ring data trying to improve the way we process the data to try and make sure we keep as much of the low frequency information on longer timescales in the trees because you have to standardise trees in a certain way to produce temperature reconstructions.
      Q123  Dr Harris: My question is: in subsequent papers when that was done was it always explicit, albeit only by reference to the Nature paper to which you were referring?
      Professor Jones: It was always explicit in the subsequent papers because some of the subsequent papers have improved the processing techniques.
      Q124  Dr Harris: Did you understand what those witnesses (if you heard them) meant when they said that they could not see, they thought the hiding of the decline approach—which is a label from an email—the identifying and dealing with the divergence problem, was itself hidden. You do not accept that?
      Professor Jones: We do not accept it was hidden because it was discussed in a paper the year before and we have discussed it in every paper we have written on tree rings and climate.

    I guess, since Jones says “every paper we have written on tree-rings and cimate”, it would be easy to review Jones’s publication list at http://www.cru.uea.ac.uk/cru/pubs/byauthor/jones_pd.htm and check for dendrochronology papers, since every one should have a discussion on the ‘decline’.

    UEA put forward these papers in evidence to the committee:

    Briffa, K R 2000. Annual climate variability in the Holocene: interpreting the message of ancient trees. Quaternary Science Reviews 19, 87-105, doi:10.1016/S0277-3791(99)00056-6.
    Briffa, K R, F H Schweingruber, P D Jones, T J Osborn, I C Harris, S G Shiyatov, E A Vaganov, and H Grudd. 1998a. Trees tell of past climates: but are they speaking less clearly today? Philosophical Transactions of the Royal Society of London Series B-Biological Sciences 353, 65-73, doi:10.1098/rstb.1998.0191.
    Briffa, K R, F H Schweingruber, P D Jones, T J Osborn, S G Shiyatov, and E A Vaganov. 1998b. Reduced sensitivity of recent tree-growth to temperature at high northern latitudes. Nature 391:678-682, doi:10.1038/35596.
    Briffa, K R, V V Shishov, T M Melvin, E A Vaganov, H Grudd, R M Hantemirov, M Eronen, and M M Naurzbaev. 2008. Trends in recent temperature and radial tree growth spanning 2000 years across northwest Eurasia. Philosophical Transactions of the Royal Society B-Biological Sciences 363, 2271-2284, doi:10.1098/rstb.2007.2199.

    • j ferguson
      Posted Mar 22, 2011 at 3:04 PM | Permalink

      Thanks Robbo, this looks like a good start. Imagining that scientists are, in part, in the curiosity business, it seemed odd that there could be these diverging trends without a lot of conjecture as to why in print.

    • Hu McCulloch
      Posted Mar 22, 2011 at 5:16 PM | Permalink

      Robbo:

      UEA put forward these papers in evidence to the committee:

      Bear in mind that UEA, not the Royal Society as originally claimed, picked the 11 papers so as to avoid any controversial ones.

      See eg http://climateaudit.org/2010/09/16/who-chose-the-eleven-an-answer/

    • mpaul
      Posted Mar 22, 2011 at 11:05 PM | Permalink

      “My question is: in subsequent papers when that was done was it always explicit?” Hardly. We now see them hiding the incline prior to 1550. Where was that disclosed and what was the explanation for it? It appears that they simply cherry-picked the middle part where there happened to be the appearance of correlation. Surely any reasonable observer would get an entirely different impression as to the usefulness of the proxies when looking at the cherry-picked bits vs the complete series. How can anyone argue that the removal of this data does not give the reader a false impression?

  34. Stacey
    Posted Mar 22, 2011 at 12:27 PM | Permalink

    Hockey Sticks

    No

    Fiddle Sticks

  35. Graeme
    Posted Mar 22, 2011 at 4:44 PM | Permalink

    wow…and they never commented on the divergence problem at the other end of the scale! so they knew about post-1960 but not pre-1600. that is duplicity

    • Posted Mar 22, 2011 at 5:37 PM | Permalink

      Re: Graeme (Mar 22 16:44), Re: Graeme (Mar 22 16:44),

      What I think this is showing is that a small change in calculations and/or tree selection leads to a large change in the proxy reconstruction. Another clue that the reconstructions are so noisy that, if the investigator runs the calculations often enough, he’s likely to come up with a result he likes. The old problem of Confirmation Bias. No duplicity required, just sloppy science.

      If you (or other readers) haven’t read M&W’s analysis, you should:

      http://www4.stat.ncsu.edu/~reich/st740/AOAS.pdf

      Well-written and straightforward, very interesting paper.

  36. kim
    Posted Mar 22, 2011 at 8:08 PM | Permalink

    So please inform me
    Who would live in such a Hell,
    Deliberately?
    ========

  37. Bill
    Posted Mar 23, 2011 at 1:57 AM | Permalink

    A self-described software developer after reviewing the climategate emails comments in part to an article by Richard Muller as follows:
    “In the module called briffa_sep_98_d.pro, one can read these lines of code:
    …..
    yyy=reform(compmxd(*,2,1))
    ;mknormal,yyy,timey,refperiod=[1881,1940]
    ;
    ; Apply a VERY ARTIFICAL correction for decline!!”

    http://www.technologyreview.com/energy/13830/page2/

    If this apparent Briffa admission regarding a “very artificial correction for decline” is off-topic, or has already been discussed, please delete and accept my apolgies.

    Steve- the comments in the code were noted at the time, but haven’t, to my knowledge, been placed into a firm context, beyond the obvious eyebrow raising. My own (thus far unreported) parsing of the code indicates that the code contains a “Briffa bodge” similar to that used in Briffa et al 1992 – see CA discussion on the topic. I’ll report on this parsing some time. Osborn said that he did not use the bodges in published articles. Although it was an active question, the “inquiries” did not report on what was going on nor, it appears, did they even ask CRU scientists about the topic.

  38. kim
    Posted Mar 23, 2011 at 8:57 AM | Permalink

    Chocolate Delight.
    Cookin’ up a nice thick batch.
    Branwenn-Fudge Factor.
    ==========

  39. dgh
    Posted Mar 23, 2011 at 1:37 PM | Permalink

    The unmodified Plate 3 of Briffa 2001 is interesting.

    http://www.ncdc.noaa.gov/paleo/pubs/briffa2001/briffa2001.html

    The beginning of the decline is apparent, indeed it is highlighted by the standard error shading. But that declining portion corresponds with a decline in the other series including the temperature record. This leaves the impression that the series might have begun an increasing trend with the others.

    And has been noted, the data description is silent regarding the truncation.

    Steve: It’s interesting to contrast the end effect of Plate 3 – where the incipient decline is noticeable, with other versions e.g. IPCC TAR, where it isn’t. It’s not easy to figure out the differences.

    • dgh
      Posted Mar 23, 2011 at 3:28 PM | Permalink

      Not easy until you have the several versions of data in hand.

      I’ll look forward to your post further detailing how Briffa persuaded the series from magenta to lime-green.

  40. Kenneth Fritsch
    Posted Mar 24, 2011 at 10:28 AM | Permalink

    Briffa and Osborn appears as an outlier when both the early and late declines are shown. Why would not some team member make a case for B&O as an outlier and merely rid the spaghetti of that noodle? Exposing the declines leads to further questions about coherence of proxies and reconstructions. Could a reasonable case be made for B&O as an outlier? What is different about B&O? Are there any implications here about partial selections of the other spaghetti noodles?

  41. clarence wilson
    Posted Mar 25, 2011 at 1:53 AM | Permalink

    Please excuse me if I have missed something obvious –

    Can someone please tell me what is the heavy black line in Figure 1 (at teh right hand end of the series, in vertcal ascent a la hockey stick?)

  42. Mike Jonas
    Posted Mar 25, 2011 at 6:40 AM | Permalink

    About the Briffa data after 1960 : Some months ago, I had an email dialogue with a prominent (in Australia) pro-AGW scientist. I challenged him repeatedly over the deletion of Briffa’s post-1960 proxy data, and his replies included (in sequence):
    “the “trick” and the “hiding” is not a scientific problem. What was being done was legitimate and was clearly acknowledged in the resulting graphs. It is an utter misrepresentation to suggest otherwise”
    “You are misunderstanding the context in which that word [hide] was used.”
    “there was simply no hiding. Just read the papers where the material was published.”
    “proxy data are proxy data. They cannot be used up to the present day. That is why instrumental data were added.”
    “terminating the proxies early was because the people to collect the proxies say that they are invalid if brought to the present day.
    So – I am sorry – but it is simply wrong to NOT terminate the proxies.”

    Maybe I should have pressed on,but I felt I had reached the brick wall.

    • kim
      Posted Mar 25, 2011 at 6:52 AM | Permalink

      I like the term ‘Antidisestablishuniformitarianism’, but I think the ‘too many negatives’ rule met the rule of holes with it.
      ============

  43. Mike Jonas
    Posted Mar 25, 2011 at 6:45 AM | Permalink

    clarence wilson – the heavy black line at the right hand end of the series is the instrumental temperature record. All the other lines are temperature proxies. By deleting the segments shown in red – they deleted the initial downstroke in the black line too – the various lines all appeared to be consistent and therefore convincing. Had they left in all the deleted bits, Blind Freddie could have seen straight away that the proxies were completely useless and therefore the whole study was useless.

10 Trackbacks

  1. By Climate Audit on Mar 23, 2011 at 5:57 AM

    […] Climate Audit by Steve McIntyre Skip to content Hockey Stick StudiesStatistics and RContact Steve McProxy DataCA blog setupFAQ 2005Station DataHigh-Resolution Ocean SedimentsSubscribe to CAEconometric ReferencesBlog Rules and Road MapGridded DataTip JarAboutCA Assistant « Hide the Decline – the Other Deletion […]

  2. […] day before yesterday, I reported that Briffa and Osborn (Science 1999) had not just deleted the post=1960 decline (see also CA […]

  3. […] day before yesterday, I reported that Briffa and Osborn (Science 1999) had not just deleted the post=1960 decline (see also […]

  4. […] du graphique: http://climateaudit.org/2011/03/21/hide-the-decline-the-other-deletion/ Briffa et Osborn ont décidé de ne garder que la portion de la série qui est en accord avec les […]

  5. By Hide the Decline – igen | The Climate Scam on Mar 25, 2011 at 6:11 PM

    […] här, här och […]

  6. By Exposed to the light on Mar 29, 2011 at 6:09 AM

    […] This data comes from a report by Briffa and Osborn in 1999 and was used as evidence to support the Michael Mann hockey stick warming data. Steve McIntrye explains how their computer code shows that they knew this data existed and chose to clip it off. Read the details […]

  7. […] 3http://climateaudit.org/2011/03/21/hide-the-decline-the-other-deletion/ […]

  8. […] […]

  9. […] en lo de Jeff Id. Es de donde he sacado los emails. El primer gráfico viene de Climate Audit [–>], un trabajo también de esta año sobre el […]

  10. […] COOK: ClimateGate just a few out-of-context emails — Hardly, Steve McIntyre has plunged into it with mind-blowing detail about how it shows a pattern among that bunch to hide inconvenient truths. […]

Follow

Get every new post delivered to your Inbox.

Join 3,421 other followers

%d bloggers like this: