Proxy Data

Here are some collations of reconstructions and data. I’ll try to add some annotations and read scripts at some point.

Spaghetti Graph of Reconstructions

  • http:/www.climateaudit.info/data/spaghetti.data.txt

Proxy Collations


19 Comments

  1. jame j engel
    Posted Dec 14, 2009 at 12:49 PM | Permalink | Reply

    Maybe someone can help me with some points I cannot get from MSM or Web.

    1) Why do the proxies end at 1980, this 2009 ?

    2) Did Mann exclude declining tree ring proxies he had ? i.e. ‘Hide the Decline’

    3) Did Mann splice in real temps into the proxies ? (I would find this incredible)

    4) If we have increasing real temperatures and decreasing proxies, and if we add the decreasing proxies (1960 or 1980 on) would the regression equation on temp on the proxies be statistically insignificant ?

    5) If the regression equation is statically insignificant; a) we cannot infer prior temps, or b) the recently reported real temps may have been massaged ?

    • G
      Posted Nov 23, 2010 at 3:50 PM | Permalink | Reply

      I do not know as much as some others but…
      1) try harder. Check when each proxies was physically collected. I would expect that few researchers would want to collect proxies for decades where much more sophisticated ways of measuring things were around. The point of using a proxy is to understand the past.

      2) The context of the hide the decline phrase/comment does not suggest this. It is supposed to be shorthand for some sophisticated statistical manipulation (not deleting real data). Most of the data he used for his analysis was not even collected by him or his colleges but something that was shared (original copy not in his control).

      3) If I understand correctly that was part of the type of stat trick but I do not understand that area of statistics well enough. If you search through the scientific papers that Mann cited, there should be one or two explanations for pros to understand.

      4) As far as anyone knows proxies in general are not declining but specific types may be. There are also so many different proxies to keep track of that it makes my head spin to think about it. Search Science-direct or some other database. Depending on which proxy data you look at, from which location, you will get different results (only a fraction of which will match your 1960-1980 decline claim).

      5) a mismatch could mean a problem with anything. The most reasonable explanation would be that the techniques for inferring past temps you are using are not so great. to my knowledge, none of the individuals accused of dishonesty in climategate had much of any involvement with measuring the temperatures. The CRU does not control the thermometers around the world. They just contact researchers in other countries and ask for data. Whenever P. Jones gets lucky, the foreigners read their e-mails and respond if they feel like it.

  2. Jonathan Carson
    Posted Dec 18, 2009 at 5:46 PM | Permalink | Reply

    Perhaps naive question: Today’s temperatures are recorded from stations all over the world, and any lack of coverage is considered a weakness. But aren’t proxies generally confined to limited geographical areas? A tree’s rings, assuming they tell us anything at all, could tell us only about the temperature in the one place where the tree grew. So why do people take these proxies so seriously?

    • D.Ellison
      Posted Dec 19, 2009 at 6:33 PM | Permalink | Reply

      Hi, Jonathan,
      Proxy data is not really confined to limited geographical areas. Looking at tree rings over the whole earth can present a partial picture of past climate in all of these different locals. Tree ring data can only hint at the temperature during the time of the tree’s growth. Tree ring growth says more about the water supply to the tree, in other words the conditions under which the tree grew, or the climate. Other proxy data must be used to complete the picture. The reason tree ring data is taken so seriously is because the data is isolated or incapsulated and cannot be contaminated by outside variables. Once the ring has completed its formation or seasonal growth , “something” cannot enter the tree and alter that formation. That “something” might be able to alter a subsequent, new developing tree ring, but that alteration would clearly be seen when the tree’s rings were examined. This is why tree rings are an important part of understanding climate and climate changes.

  3. Posted Jan 8, 2010 at 11:13 PM | Permalink | Reply

    The links on the page http://climateaudit.org/proxy-data/ do not work. It seems that there has been a change to the directory structure and the links not updated.

    • RomanM
      Posted Jan 9, 2010 at 8:11 AM | Permalink | Reply

      Re: Ray Tomes (Jan 8 23:13),

      Try substituting the letters “info” for the letters “org” in the URLs. The data and script pages were moved by WordPress in the Great Transfer of 2009.

      • Posted Jan 10, 2010 at 5:00 AM | Permalink | Reply

        Thanks Roman.

        • Martin Hendel
          Posted Nov 9, 2010 at 12:52 PM | Permalink

          Hi,
          unless I missed something, I believe that the two proxy data links are broken, even after replacing .org by .info:
          – MBH99
          – Crowley and Lowery 2000

          Does anyone else have any information about them or a copy of those files?

          Many thanks,
          Martin

  4. mike seward
    Posted Feb 7, 2010 at 11:14 PM | Permalink | Reply

    Re Tree Ring “Proxies”
    Tree growth rates are a compound of temperature and watersupply. Any home gardener could figure that out.

    On what possible basis then could people with PhD’s and professorships propose tree rings as a temperature proxy without at the very least a concurrent record of water supply at the same location?

    No wonder the Saudi’s think this is all so much doo doo, its not their vested interest in selling oil, its their long experience in lots of temperature with no water = no trees. To them it is a no brainer.

    • Adam Collyer
      Posted Jul 7, 2010 at 4:42 PM | Permalink | Reply

      Even more interestingly, wouldn’t tree rings depend on CO2 concentration over long timescales? Plants grow faster as CO2 concentrations rise. So couldn’t tree rings be a proxy for CO2 rather than temperature?

      In which case, it wouldn’t really be surprising if the tree ring “temperature” record correlated with CO2 – because it would really be a CO2 record!

      • Gogo
        Posted Nov 23, 2010 at 3:58 PM | Permalink | Reply

        The leaves of trees are used as a CO2 proxy. There are also more than half a dozen other proxies for CO2. Which variable correlates most strongly with growth rates depends on which species of tree you are looking at. Theoretically if you had a species for each variable, you would be able to get really good ideas of past atmospheres.
        In general, how fast a plant grows depends of what variable is limiting (like a scarce nutrient).

        • Gogo
          Posted Nov 23, 2010 at 3:59 PM | Permalink

          do not understand what reasoning your “over long timescales comment was based on”.

        • Posted Nov 24, 2010 at 1:20 PM | Permalink

          Over long timescales I meant more than yearly. In other words, obviously the CO2 concentration varies across the year, but the tree ring thickness presumably would depend on average CO2 concentrations over periods of more than one year.

          But the fact remains – if more CO2 means that tree rings get wider, then thicker tree rings could be the result of higher CO2 as well as higher temperature.

  5. mike seward
    Posted Feb 7, 2010 at 11:27 PM | Permalink | Reply

    the thing that always got be about the “hockey stick” analysis was that planetary albedo was not modelled as a variable yet consideration of greenhouse effect first principles says it is and a significant one. Even if Stephen McIntyre & Ross McKindrick and others had not analysised the mathematical modelling Mann et al’s work would remain fundamentally and fatally flawed.

    The only decent analysis I have come across which models albedo is by Ermakov, Okhlopkov and Stozhkov and which uses a fourier transform to model the constituent contributions to temperature variation over time.

  6. mike seward
    Posted Feb 7, 2010 at 11:34 PM | Permalink | Reply

    and another thing…Ifyou compare the temperature record since about 1850 of the globe, the two hemishperes and the difference between the two an interesting pattern emerges. The SH ( where I live) is on avereage about 0.1C cooler than the NH but over the past decade that has opened up to about 0.3 degrees. This suggests to me that the two hemispheres act quite differently in their contribution to the global average. That the two are geographical opposites, that is NH is largely land with a polar sea surrounded by land mass and the SH the opposite seems to me that their albedo behaviour is likely quite different and that perhaps the NH tends to drive a warming and the SH a cooling phase. This is also consistent with the various dynamic interactions of the seas and atmoshere from N to S.

  7. John Murphy
    Posted Feb 18, 2010 at 2:13 AM | Permalink | Reply

    Steve

    Have you seen the new paper by Keith Briffa at http://www.cru.uea.ac.uk/cru/people/briffa/yamal2009/

    Regards

    Steve:
    that was out in late October. See contemporary posts for much discussion.

  8. wooden spoon
    Posted Feb 26, 2010 at 5:41 PM | Permalink | Reply

    A new paper that casts doubt on the findings of Berkelhammer & Stott’s oxygen isotope bristlecone departure. Abstract is here:

    http://hol.sagepub.com/cgi/content/abstract/20/1/3

  9. ButchKelly
    Posted May 10, 2010 at 4:55 PM | Permalink | Reply

    Of course CO2 has been increasing also, any gardener will tell you that an increase in fertilizer will increase the rate of growth.

  10. Posted Jan 18, 2011 at 12:21 PM | Permalink | Reply

    Has the following e-mail from the CRU/climategate regarding Yamal treerings been commented on CA? It clearly shows that Briffa have tried to alter the conclusion delt with in Hentemirov 2002 in accordance with the teams effort to show the last centuries “unprecedented warmth”

    From: Rashit Hantemirov
    To: Keith Briffa
    Subject: Re: Yamal paper for The Holocene special issue
    Date: Wed, 21 Aug 2002 17:56:18 +0500
    Reply-to: Rashit Hantemirov

    Dear Keith,
    thank you very much for editing our paper.
    It’s a pity you strike your name off the list of authors, you
    make an important contribution to writing paper. Your corrections
    and additions surely improve paper.

    I would only notice the next sentence (page 8):

    ‘The low interannual variability and the minimum occurrence of
    cold extremes during the 20th century, argue that the most recent
    decades of this long summer record represent the most favourable
    climate conditions for tree growth within the last four
    millennia.’

    I’m not sure that this statement follows unambiguous from results
    presented in this paper. Because mean temperatures during last
    decades, according presented reconstruction, are not exceptional.
    Besides, e.g. period about 1700 BC, according this
    reconstruction, represent probably the same conditions taking
    into account low variability, low occurrence of extremes and high
    mean temperature.
    May be to soften this statement and replace ‘the most favourable’
    with something like ‘highly favourably’ or ‘probably the most
    favourable’?

    Thank you once more for invaluable assistance.

    Best regards,
    Rashit M. Hantemirov

    (I’m sorry for the late answer, I just come back from the trip to
    the north.)

One Trackback

  1. [...] Steve McIntyre, Jeff Id, and others have done a vast amount of work analyzing how Mann, et. al., used this data in hockey stick construction.  I am just starting to look at it in my own ignorant fashion, and expect to have a lot a fun [...]

Post a Comment

Required fields are marked *

*
*

Follow

Get every new post delivered to your Inbox.

Join 3,297 other followers

%d bloggers like this: