Up-to-Date White Spruce Ring Widths

Bring the proxies up to date” was the title of one of my earliest posts. Michael Mann had explained that doing so required the use of heavy equipment (like tree ring borers) and travel to out-of-the way sites such as Bishop, California or even Niwot Ridge, a full 45 minute drive from UCAR world headquarters in Boulder CO. As a result, Mann explained that few proxies were available after 1980 and it was therefore necessary to keep using bristlecone and other series ending in 1980 or so. [See next post in sequence.]

Obviously, given the warm temperatures in the 1990s and 2000s, up-to-date proxies reaching 1998 and later offer an ideal opportunity to test the validity of tree ring proxies out-of-sample. I’ve done a quick calculation of contributions to the North American tree ring data base at WDCP; I counted no fewer than 250 sites where there is data for 1998 or later.

The species with the most new sites is Douglas fir (PSME); there are also many new white spruce (PCGL) site – a species held to be a temperature proxy. There are 48 sites where new measurements have been archived, many due to the work of the Jacoby group in Alaska and northern Canada. Jacoby’s archiving is frustratingly incomplete however. Also presumably to be annoying, virtually none of the Jacoby chronologies are archived with the measurements. (This is the opposite of Briffa for Yamal, Taymir and the Tornetrask update, where the RCS chronologies are archived, but not the underlying measurement data.) The Team never makes it easy. In the new data, there are no bristlecone (PILO, PIAR) or foxtail (PIBA) sites with archived data for 1998 and later, although we know that Hughes carried out measurements at Sheep Mountain in 2002. (One of my standing predictions is that Sheep Mountain bristlecone ring widths did not go off the charts during the warm 1990s.)

Even without any Jacoby chronologies, there are still 14 PCGL chronologies reaching 1998 or later. The figure below shows the average of chronologies (the average of available chronologies.) I have trouble discerning a HS in this average. (I didn’t use Mannian principal components.)


Figure 1. Average of 14 North American PCGL chronologies

If the proxy hypothesis is that PCGL ring widths increase with increasing temperatures, I would say that these non-Jacoby chronologies are evidence against the hypothesis. Indeed, one might even be inclined to say that the average of these 14 chronologies is inversely related to temperature, as ring widths were low in the warm period in the late 1930s and increased to higher levels in the mid-1960s when temperatures were supposedly cooling.

Update: Rob Wilson and Mike Pisaric have observed that the Meko white spruce chronologies in northern Alberta are considered to be precipitation proxies rather than temperature proxies and argued that this post is “flawed!” In fairness, everything in this post is expressed conditionally: “if the proxy hypothesis is that PCGL ring widths increase with increasing temperatures”, then these particular updated chronologies are not evidence for that proposition. If they are not believed to be temperature proxies, then they are likewise not evidence for the proposition either. The logic may be a little subtle for a climatologist. I am quite prepared to accept their opinion that the white spruce chronologies from northern Alberta are not temperature proxies. However, as I point out in some subsequent point, applying the same logic requires dendroclimatologists to take a stand against some of the proxies used in multiproxy studies.

Update2: Discussion of Alberta sites by Meko here


103 Comments

  1. Posted Mar 19, 2007 at 11:02 PM | Permalink

    If it’s soooooooo damned difficult to get the updates on the proxies, then wasn’t it that much harder to get the original samples in the first place?????

  2. JMS
    Posted Mar 20, 2007 at 2:09 AM | Permalink

    Repeat after me: global warming does not mean a uniform temperature response in every gridcell.

  3. Rob Wilson
    Posted Mar 20, 2007 at 2:40 AM | Permalink

    Steve et al.,

    Sigh…………!

    Location, location, location!

    or even

    ecology, ecology, ecology!

    you are confusing the issue again.
    Your little ‘spiel’ above is meaningless without some site information.
    Where are the spruce sites located (?Map)?
    How close are these sites to the high elevation or high latitude tree-line?
    For what purposes were these sites originally sampled?

    For readers of this blog, PLEASE understand that one cannot randomly sample trees from any location and expect there to be a valid climate signal (temperature or precipitation).

    Dendroclimatologists are guilty of biased sampling – we have to be. To derive a valid temperature signal we must go to environments where tree growth is limited by temperature. The high elevation or latitude tree-lines being a good starting point. Even when going to these environments, site ecology (i.e. aspect, angle of slope and drainage) also needs to be taken into account as this will also influence how the trees respond to temperature.

    Dendrochronology is NOT just a study of climate. Tree-rings are used to study fire history, geomorphological processes and ecological aspects of tree growth (e.g. stand dynamics, insect attacks, pollution effects etc). Therefore, one cannot assume that ALL tree-ring data in the “North American tree ring data base at WDCP” was sampled with a climate related question in mind. Therefore, a simple average of 14 sites, without knowing why they were sampled in the first place, is arguably a meaningless exercise.

    Regards
    Rob

  4. Nicholas
    Posted Mar 20, 2007 at 3:45 AM | Permalink

    “For readers of this blog, PLEASE understand that one cannot randomly sample trees from any location and expect there to be a valid climate signal (temperature or precipitation).”

    We’re all familiar with the Bristlecone/Foxtail schlemozzle. I think it’s understandable if we are collectively extremely cynical and skeptical about dendroclimatology at this point.

  5. TAC
    Posted Mar 20, 2007 at 3:57 AM | Permalink

    Rob,

    IMHO, when you state:

    you are confusing the issue again

    you touch directly on SteveM’s point: Standard dendrochronological practice make it nearly impossible not to “confuse the issue.”

    Many tree-ring datasets are essentially un-auditable. Critical meta-data are unavailable (except, possibly, to an elite group). Under the circumstances, Why should a skeptic (read “scientist”) believe analyses based on these data?

    Well, here’s one possible (potential) reason: Because the methods have been shown to work in controlled experiments employing acceptable sampling methodologies. For example, dendro-competent practitioners could identify temperature-sensitive sites, publish the complete list of site locations, wait a while — say 25 years — and then resample the trees and publish the complete re-sampled tree-ring data along with a corresponding record of temperatures. If the tree-ring data during the 25-year period correspond to the temperature signal, that would provide a basis for believing that the old-wood ring data contain useful information about historical temperatures.

    Of course, this “experiment” could be conducted today. All that need happen is to follow SteveM’s mantra and “bring the proxies up to date.” How hard can that be? Really? I, for one, appreciate SteveM’s sense of frustration.

    Finally, I think all regular CA readers are fully aware

    that one cannot randomly sample trees from any location and expect there to be a valid climate signal (temperature or precipitation)

    The nagging question for some of us is whether there are any conditions under which “a valid climate signal (temperature or precipitation)” can be obtained from tree-ring data. While I tend to believe that such conditions do exist, at this point I could use some convincing.

  6. fFreddy
    Posted Mar 20, 2007 at 4:22 AM | Permalink

    Re #3, Rob Wilson

    To derive a valid temperature signal we must go to environments where tree growth is limited by temperature. The high elevation or latitude tree-lines being a good starting point. Even when going to these environments, site ecology (i.e. aspect, angle of slope and drainage) also needs to be taken into account as this will also influence how the trees respond to temperature.

    Is this other data also stored at WDCP ?
    If not, why not ?

  7. MarkW
    Posted Mar 20, 2007 at 5:00 AM | Permalink

    Rob,

    If location is so important, then why does Mann use Bristlecone as his only proxy for some of his time spans?

  8. bernie
    Posted Mar 20, 2007 at 5:57 AM | Permalink

    Whoa guys:
    Let’s enlist Rob to help out obtaining the additional in situ data needed for these 78 samples. So if Rob has friends among the dendroclimatologists, he can help get the relevant data archived. I am assuming that folks here aren’t out to prove or disprove AGW per se, but to (a) improve the quality of data and (b) the quality of certain scientific processes. My skepticism is an out growth of poorly or non-archived data, unspecificed or undocumented data manipulations, excessive defensiveness on the part of some in the climate fields and an abundance of hyperbole from the proponents of AGW.

  9. Jean S
    Posted Mar 20, 2007 at 6:04 AM | Permalink

    Rob:

    Therefore, one cannot assume that ALL tree-ring data in the “North American tree ring data base at WDCP” was sampled with a climate related question in mind. Therefore, a simple average of 14 sites, without knowing why they were sampled in the first place, is arguably a meaningless exercise.

    Does this apply also to MBH as the key components at 1000-1500 steps are the “PCs” of all tree-ring data in the “North American tree ring data base at WDCP”? If not, then what is the magic property of the (Mannian) PCA that avoids the above mention selection bias?

  10. Jean S
    Posted Mar 20, 2007 at 6:22 AM | Permalink

    Rob:

    For readers of this blog, PLEASE understand that one cannot randomly sample trees from any location and expect there to be a valid climate signal (temperature or precipitation).

    Oh dear, if you had been reading this blog for a longer time, you’d noticed that this is pretty obvious for the most of the readers of this blog. IMO you should be educating your collegues making multiproxy reconstructions who, not only assume that all series contain valid temperature signal, but assume that the trees are linearly related to the hemispheric mean temperature. See here, part E, for the latest clarification of the principle:

    The model assumes there is a linear dependence in the composite, not in the individual proxies.

  11. Edouard
    Posted Mar 20, 2007 at 6:48 AM | Permalink

    Hello,

    Have you seen this entry on Pielke’s blog?

    http://climatesci.colorado.edu/2007/03/16/two-papers-on-the-urban-heat-island-effect-on-temperatures/

    ““A 1250-Year History of Summer Temperature in the European Alps

    Reference
    Buntgen, U., Frank, D.C., Nievergelt, D. and Esper, J. 2006. Summer temperature variations in the European Alps, A.D. 755-2004. Journal of Climate 19: 5606-5623.”

    “What was learned
    Among a number of other things, notable features identified by the authors were high temperatures in the late tenth, early thirteenth, and twentieth centuries and a prolonged cooling from ~1350 to 1700, or as they describe it: “warmth during medieval and recent times, and cold in between.” Also of great interest, they report that the coldest decade of the record was the 1810s, and that even though the record extended all the way through 2004, the warmest decade of the record was the 1940s. In addition, they observed that “warm summers seem to coincide with periods of high solar activity, and cold summers vice versa.” Finally, they report that comparing their newest temperature record with other regional- and large-scale reconstructions “reveals similar decadal to longer-term variability.”

    What it means
    Buntgen et al. conclude – in the final sentence of their paper – that based upon their findings and the similar findings many others, “the twentieth-century contribution of anthropogenic greenhouse gases and aerosol remains insecure.””

  12. Douglas Hoyt
    Posted Mar 20, 2007 at 6:57 AM | Permalink

    Interesting that tree rings show the present time is not as warm as earlier years in Canada. A similiar result is found in Europe where the 1940s are warmer than now as described in a paper by Buntgen et al. and I copy the review from CO2science.com:

    “A 1250-Year History of Summer Temperature in the European Alps

    Reference
    Buntgen, U., Frank, D.C., Nievergelt, D. and Esper, J. 2006. Summer temperature variations in the European Alps, A.D. 755-2004. Journal of Climate 19: 5606-5623.

    What was done
    An annually-resolved mean summer (June-September) temperature record for the European Alps, covering the period AD 755-2004 and based on 180 recent and historic larch (Larix decidua Mill.) maximum latewood density series, was developed using the regional curve standardization method that preserves interannual to multi-centennial temperature-related variations.

    What was learned
    Among a number of other things, notable features identified by the authors were high temperatures in the late tenth, early thirteenth, and twentieth centuries and a prolonged cooling from ~1350 to 1700, or as they describe it: “warmth during medieval and recent times, and cold in between.” Also of great interest, they report that the coldest decade of the record was the 1810s, and that even though the record extended all the way through 2004, the warmest decade of the record was the 1940s. In addition, they observed that “warm summers seem to coincide with periods of high solar activity, and cold summers vice versa.” Finally, they report that comparing their newest temperature record with other regional- and large-scale reconstructions “reveals similar decadal to longer-term variability.”

    What it means
    Buntgen et al. conclude – in the final sentence of their paper – that based upon their findings and the similar findings many others, “the twentieth-century contribution of anthropogenic greenhouse gases and aerosol remains insecure.” Since these words appear to be particularly carefully chosen, we feel we should not attempt to rephrase or clarify them any further, assuming that most rational people will be able to correctly determine for themselves both what they mean and what they imply.”

    Note that, in the review above, they say “the warmest decade of the record was the 1940s”. This result is quite different than what you get using thermometers which have the present day as warmest. The obvious conclusion is that European thermometers are UHI contaminated compared to rural trees and that the thermometers are not recording climate change accurately.

  13. Rob Wilson
    Posted Mar 20, 2007 at 7:01 AM | Permalink

    Hi
    All I wanted to communicate with my response, is that there was no useful information in what Steve M had posted. If he had made a mean series of tree-ring data that have been interpreted as being temperature proxies, and stated perhaps the region that mean series represented, it would have been much more meaningful.

    To be fair, I think on the whole, dendrochronologists are not too bad at archiving data. Just go to the relevant pages at the WDC:

    http://www.ncdc.noaa.gov/paleo/treering.html

    and see how much data have been archived over the past 30 years.

    Submitting data is very easy:

    http://www.ncdc.noaa.gov/paleo/itrdb_submit.html

    and you will see that there are form fields where relevant meta data can be inserted.

    Accessing these meta data is admittedly not so easy and one must go into the FTP directory and look for the relevant *.txt file with the code of that site or look at the relevant COFECHA output.

    Meta files do not exist for all sites. This mostly reflects that such information was not requested in early versions of the ITRDB.

    However, there are two further ways to acquire meta information:
    1. Look at the original relevant publication – hopefully the information is detailed in the paper.
    2. Contact the original primary scientist.

    In the latter scenario, I have never had a problem acquiring information from colleagues if such information was never detailed in the original articles.

    Unfortunately for Steve M, he has not made too many friends in the palaeo-community over the last few years and so it is not really surprising that researchers are not willing to help or work with him.

    My 10 cents
    Rob

  14. Rob Wilson
    Posted Mar 20, 2007 at 7:05 AM | Permalink

    This did not come through the first time – let’s try again

    Hi
    All I wanted to communicate with my response, is that there was no useful information in what Steve M had posted. If he had made a mean series of tree-ring data that have been interpreted as being temperature proxies, and stated perhaps the region that mean series represented, it would have been much more meaningful.

    To be fair, I think on the whole, dendrochronologists are not too bad at archiving data. Just go to the relevant pages at the WDC:

    http://www.ncdc.noaa.gov/paleo/treering.html

    and see how much data have been archived over the past 30 years.

    Submitting data is very easy:

    http://www.ncdc.noaa.gov/paleo/itrdb_submit.html

    and you will see that there are form fields where relevant meta data can be inserted.

    Accessing these meta data is admittedly not so easy and one must go into the FTP directory and look for the relevant *.txt file with the code of that site or look at the relevant COFECHA output.

    Meta files do not exist for all sites. This mostly reflects that such information was not requested in early versions of the ITRDB.

    However, there are two further ways to acquire meta information:
    1. Look at the original relevant publication – hopefully the information is detailed in the paper.
    2. Contact the original primary scientist.

    In the latter scenario, I have never had a problem acquiring information from colleagues if such information was never detailed in the original articles.

    Unfortunately for Steve M, he has not made too many friends in the palaeo-community over the last few years and so it is not really surprising that researchers are not willing to help or work with him.

    My 10 cents
    Rob

  15. Jean S
    Posted Mar 20, 2007 at 7:26 AM | Permalink

    Rob:

    Unfortunately for Steve M, he has not made too many friends in the palaeo-community over the last few years

    What do you think is the reason for that? What I’ve been communicating with Steve, he seems to be a nice fellow.

  16. Steve McIntyre
    Posted Mar 20, 2007 at 7:28 AM | Permalink

    Hi, Rob. If ring widths for a tree and species are to be used as a temperature proxy, then you should be able to specify the species and locations ahead of time, and evidence the results in a population of different sites without picking after the fact. If you find that there are factors that cause some sites to respond differently afterwards, you can’t just pick sites after the fact, you have to modify your hypothesis and collect from a new population and show that your modifed hypothesis works.

    As to using updated results, the Jacoby group has much the best record in the field and, as I observed have collected much new data. I agree that there has been a considerable amount archived. The frustrating aspect, as you know, is that the archiving isn’t complete and, as long as that isn’t the case, one wonders why other sites haven’t been archived, whether there has been some criterion applied to the decision to archive – as has occurred with Jacoby in the past, where he only archived sites with the “Jacoby signal”. I have no idea why Jacoby and d’Arrigo didn’t archive their chronologies for the new data. Yeah, one can calculate one’s own chronology, but the chronology as used by your group is also relevant and you’ve both not archived it and, as you know and through no fault of yours, d’Arrigo has refused to make them available on request.

    As we’ve discussed in the past, the metadata is frustratingly incomplete. For example, altitudes of individual samples are not recorded even in sites with substantial variations in altitude of sample and inhomogeneity of altitude over time (e.g. Polar Urals, bristlecones).

    It shouldn’t be a matter of whether someone has made “friends” that entitles an examination of the data. Business auditors are not supposed to be “friends” of the promoters; one of the problems with Arthur Andersen and Enron is that the auditors got to be friends with the executives. If tree ring studies are to be relied upon for climate policy, then data and metadata should be available independently and not depending on whether you’re a “friend” of the cause.

    I used PCGL in this little exercise because this species seems to have been widely used as a temperature proxy, including D’Arrigo et al 2006 of which you were a coauthor. I’d be happy to re-do the calculation with the D’Arrigo Jacoby chronologies if and when they are archived are made available to me.

    The 14 sites in this average were:

    ak038 Grasshopper Hill PCGL 64.55 -163.40 150 1790 1999 Andrea Lloyd
    ak039 Monahan Flats PCGL 63.16 -147.55 884 1697 1998 Andrea Lloyd
    ak040 Nome Creek below treeline PCGL 65.22 -146.37 884 1785 1999 Andrea Lloyd
    ak075 Hawkins Hill PCGL 61.09 -142.05 1030 1580 1998 Gordon Jacoby Nicole Davi Greg Wiles
    ak072x Windy Wolf PCGL 61.46 -141.38 1036 1573 1998 Gordon Jacoby Nicole Davi Greg Wiles
    cana212 Point Providence 2001 Recollect PCGL 59.00 -112.00 209 1698 2000 David Meko Charles Stockton Hal Fritts
    cana213 Quatre Fourches 2001 Recollecti PCGL 58.48 -111.30 209 1712 2000 David Meko Charles Stockton Hal Fritts
    cana214 Horseshoe Slough PCGL 58.54 -111.36 209 1801 2000 David Meko
    cana215 Revillon Coupe 2001 Recollectio PCGL 58.54 -111.24 209 1742 2000 David Meko Charles Stockton Hal Fritts
    cana216 Peace/Rouchers 2001 Recollectio PCGL 59.00 -111.24 209 1687 2000 David Meko Charles Stockton Hal Fritts
    cana217 Athabasca River 2001 Recollecti PCGL 58.24 -111.30 209 1708 2000 David Meko Charles Stockton Hal Fritts
    cana218 Birch River PCGL 58.30 -112.30 209 1757 2000 David Meko
    cana219 Mamawi Lake PCGL 58.36 -111.18 209 1801 2000 David Meko
    ak085 Dalton Highway Mile Post 235 PCGL 68.02 -149.41 810 1806 2005 Andrea Lloyd Chris Fastie

  17. Steve McIntyre
    Posted Mar 20, 2007 at 7:30 AM | Permalink

    For general information, here’s a barplot showing the major species for which 1998 and later information is at WDCP.

  18. Ron Braud
    Posted Mar 20, 2007 at 7:33 AM | Permalink

    Does anyone know whether research has been done to determine whether, or to what degree, CO2 fertilization effects should be considered on Black Spruce(and any other tree) Ring Width that are or could be used in a temperature proxy study? Are Bristle cones and Foxtails the only plants so effected?

  19. Steve McIntyre
    Posted Mar 20, 2007 at 7:38 AM | Permalink

    Rob, a question. I’ve looked at the metadata and it typically adds rather little. For example, the meta data for the Almond Butter AK site ak057 here shows the COFECHA results of a run by Bruce Bauer of WDCP, but no metadata contributed by D’Arrigo et al that is not on the rwl form.

    I noticed that the metadata does refer to a chronology ak057.crn. However the chronology ak057.crn is not archived (as noted previously in the context of non-archiving of site chronologies in the recent D’Arrigo tranche.)

  20. Posted Mar 20, 2007 at 7:47 AM | Permalink

    #3

    To derive a valid temperature signal we must go to environments where tree growth is limited by temperature.

    Unless you use CFR methods. Then, any local correlation with the climate field of interest is not required. We can assume that tree growth is limited by global temperature, for example.

    #13

    Unfortunately for Steve M, he has not made too many friends in the palaeo-community over the last few years and so it is not really surprising that researchers are not willing to help or work with him.

    They’re scared they may lose their jobs?

  21. Jean S
    Posted Mar 20, 2007 at 7:51 AM | Permalink

    All I wanted to communicate with my response, is that there was no useful information in what Steve M had posted..

    No, to paraphrase gavin, you are wrong. There is plenty of useful information. My summary:

    Mann has explained that a reason for ending the series in 1980, is the lack of series extending afterwards. This allows him to use Bristlecones and other not so meaningful series. Steve is here simply showing that if you take those series extending to 1998, there is no hockey stick. If there is no hockey stick in the North American tree ring data set, there won’t be any hockey stick in MBH-style reconstruction (at early steps). It is simple as that.

    Furthermore, Steve shows above that the mean of the series does not seem to correspond to the temperature. This implies that either the corresponding cronologies do not contain a valid tempereture signal and/or the positive linear response assumed by the multiproxy studies is invalid (for those cronologies).

  22. Gator
    Posted Mar 20, 2007 at 7:55 AM | Permalink

    It shouldn’t be a matter of whether someone has made “friends” that entitles an examination of the data. Business auditors are not supposed to be “friends” of the promoters; one of the problems with Arthur Andersen and Enron is that the auditors got to be friends with the executives. If tree ring studies are to be relied upon for climate policy, then data and metadata should be available independently and not depending on whether you’re a “friend” of the cause.

    Auditors shouldn’t have an axe to grind either. Given your apparent contempt for “The Team” why should anyone consider you a disinterested auditor? And if you’re not a disinterested auditor, why should someone help you?

  23. Steve McIntyre
    Posted Mar 20, 2007 at 8:11 AM | Permalink

    #22. I’m not asking for “help”. I’m asking that people archive their data and methods. As to the Team, they were unco-operative long before anyone had ever heard of me. IT isn’t just me – if a random reader tries to get detailed sample data from Lonnie Thompson, he won’t succeed.

  24. bernie
    Posted Mar 20, 2007 at 8:16 AM | Permalink

    You are right auditors should not assume the role of prosecutors. Independent of the attitudes of anyone towards anyone else, the principles should be the same: the data should be archived in such away as the analysis undertaken by one researcher can be replicated by another. That is fundamental to the scientific endeavor. To continue the auditor metaphor, if the books to not follow standard accounting principles then they do not follow standard accounting principles. The issue remains how do we get the available data archived in a way that enables replication.

  25. Steve McIntyre
    Posted Mar 20, 2007 at 8:17 AM | Permalink

    #21.

    to paraphrase Gavin

    LOL. I wonder what the recording will say (and why Gavin said he was sorry).

  26. jax
    Posted Mar 20, 2007 at 8:22 AM | Permalink

    #2,3,
    Exactly! That is one of Steve and Ross’s criticisms of the MBH hockey stick:

    “Further examination showed that the early portion of the series consisted of only one tree; that the 20th century portion was highly non-linear and the site was located in a forested area nowhere near the northern treeline (of which it is supposed to be a member).”

  27. Steve McIntyre
    Posted Mar 20, 2007 at 8:22 AM | Permalink

    For reference, Mann’s comment about the need to use stale proxies is located here and was as follows:

    Most reconstructions only extend through about 1980 because the vast majority of tree-ring, coral, and ice core records currently available in the public domain do not extend into the most recent decades. While paleoclimatologists are attempting to update many important proxy records to the present, this is a costly, and labor-intensive activity, often requiring expensive field campaigns that involve traveling with heavy equipment to difficult-to-reach locations (such as high-elevation or remote polar sites).

    My Op Ed on bringing the proxies up to date is here and includes the following picture of the “heavy equipment” involved in bringing tree ring samples up to date.

  28. Rob Wilson
    Posted Mar 20, 2007 at 8:28 AM | Permalink

    My last response as I need to get some work done:

    1. The Buentgen paper is available here (http://www.wsl.ch/staff/jan.esper/publications/Buentgen_2006_JoC.pdf). The statement “The obvious conclusion is that European thermometers are UHI contaminated compared to rural trees and that the thermometers are not recording climate change accurately” is ridiculous. The MXD TR record tracks the instrumental record very well – see also Buentgen’s earlier paper using the RW data (http://www.wsl.ch/staff/jan.esper/publications/Buentgen_2005_CD.pdf). The new MXD series is really quite special and highlights the need to develop more long MXD data-sets.
    The coolest year over the whole period is 1816 (year without a summer in Europe) and the warmest is 2003 – direct response to the hot summer of that year. There are no residual problems in the calibration of this record with the instrumental data except for the early instrumental data (pre 1818) – but that is an interesting story in itself. Steve M may use D’Arrigo et al. (2004) as an example of non-linear response and threshold problems, but this Buentgen record clear shows no divergence and simple linear modelling works quite well in this situation – as it does in many TR based studies.

    2. Although I do not know the details of the relevant studies, 8 of the 14 records were archived by Dave Meko. He normally works with moisture stressed trees.

    3. re. META data files. I think all the relevant META data will be included in the *.txt file rather than the COFECHA output, but if the data were not provided in the first place (it is not enforced to do so), then it cannot be made available.

    4. I do not want to be drawn into a debate about Mike Mann’s NH reconstruction. If you do not like it, fine – forget it. There are other NH records now available.

    Over and out
    Rob

  29. Gerald Machnee
    Posted Mar 20, 2007 at 8:30 AM | Permalink

    Re Heavy Equipment – Maybe he meant the air-conditioned Hummer?

  30. Steve McIntyre
    Posted Mar 20, 2007 at 8:48 AM | Permalink

    Rob, The Buentgen study looks interesting. Can you direct me to the location of the archive of the Buentgen data?

  31. Hans Erren
    Posted Mar 20, 2007 at 9:04 AM | Permalink

    re 23:

    Indeed, he is busy collecting new samples for his freezer.

  32. Stan Palmer
    Posted Mar 20, 2007 at 9:08 AM | Permalink

    Unfortunately for Steve M, he has not made too many friends in the palaeo-community over the last few years and so it is not really surprising that researchers are not willing to help or work with him

    When General Groves, who managed the Manhattan project, visited the group of physicists at the University of Chicago who were wrking on the atomic bomb, he quickly determined that there was little possibility of this group ever creating a workable design. This group consisted of some the greatest physicists of the 20th century. It had some of the best minds that the 20th century produced. Yet the nature of academic atmosphere in which that group worked prevented it from succeeding. Groves determined that a more formal and organized project was essential to the success of the effort.

    This is why the above quotation is highly relevant to the entire question of AGW. It is claimed that AGW is a serious threat to future of civilization and perhaps even to the future of the environment. Yet research into it is being treated as a purely academic matter and is being mediated by the peer review and grant process. As Groves found with the bomb, this is not an effective way of addressing a very serious and pressing issue.

    The philosopher Imre Laktos in this seminal book “Proofs and Refutations” showed that there are two aspects of mathematical research. There are the formal proofs which are published in peer-reviewed journals. This is an essential aspect of mathematics. However Lakatos showed that it is the informal mathematics, the unstructured interaction between mathematicians, that drives this froward. It is informal mathematics, the idle conversations in hallways, at conference etc, that allow the sharing of ideas, the settings of problems. Formal mathematics archives the results of mathematical research. Informal mathematics drives it froward. Later sociological reserahc has extended this result across many orehr fields

    This is the essence of peer review. It why the quotation above is important. Informal interaction takes place among friends not adversaries. Yet General Groves found the Chicago group to be disorganized and unlikely to succeed. Success in a large project requires organization and it is organization that entire AGW effort is lacking. The peer review/grant process is not capable of tacking a large issue which requires a speedy resolution. Can one imagine that the space program would have succeeded if NAA had managed peer reviewed grants. NASA needed equipment that would meet stringent specfications. Groups that could produce such equipment got the contracts. Groups that couldn’t, didn’t. Friendship was not a consideration.

    I think that it is this point that Steve McIntyre is making. Results are published and relied upon in relation to AGW that have far ranging public policy and economic implications. These results have to be verified. These results have to be shared. Friendship among researchers is not an efficient way for organizing this. As a matter of fact, as evidence byt eh Chicago group and the current AGW mess, it is not an effecive way of organizing this as well.

    The AGW debate has been marked by claims that peer review is the mark of true science. The Manhattan Project and the NASA moon program have indicated that the peer review/grant process is not essential (and is indeed counterproductive) to success. Having a critical project depend upon friendship among researchers in nonsensical. Putting a critical project at risk because someone’s feelings are hurt is unconscionable.

    What AGW needs is the equivalent of NASA or the Manhattan Project.

  33. Steve McIntyre
    Posted Mar 20, 2007 at 9:21 AM | Permalink

    We start to hear of the phrase “post-normal” science. If you change the nuance slightly, I would submit that post “normal science” comes engineering. Engineers check parameters; they do detailed studies; they don’t publish little articles in academic journals, where no one can check data, parameters, methodology. It’s fine if no one is using the results, but, since we are invited to rely on these results, these precious little academic games should have ended long ago.

  34. Bill F
    Posted Mar 20, 2007 at 9:55 AM | Permalink

    Rob,

    For readers of this blog, PLEASE understand that one cannot randomly sample trees from any location and expect there to be a valid climate signal (temperature or precipitation).

    Please speak with Martin Juckes regarding this. According to his responses to comments on his recently submitted paper, it is not necessary for tree rings to match local temperature in order for them to display a global temperature signal. While I am well aware that your statement is the correct one, the field of dendroclimatology has been taken over by people like Juckes and Mann who believe that any tree anywhere in the world is a thermometer if it shows a hockey stick shape in the tree ring widths over time. As long as serious researchers like yourself defend the science as a whole against people like Steve who are trying to audit their results, you will ALL be lumped together in the minds of cynics who see their methods for what they are…junk science. I am not saying you should throw your colleagues under the bus and join Steve…I am just saying that when you attack Steve for trying to shed light on what is surely a bogus claim by Mann about updating the proxies, and then turn around and refuse to criticize those who won’t share data and who refuse to archive useful metadata, you are building a wall around your science as a whole and inviting cynical observers to assume that the whole dendroclimatology field is full of BS artists whose work can’t withstand honest scrutiny and review by outsiders.

    It is up to you, but if you want to silence those who believe dendroclimatology is 21st century alchemy, attacking people like Steve for trying to replicate methods is not the way to go about it in my mind.

  35. Mark T.
    Posted Mar 20, 2007 at 9:58 AM | Permalink

    The level of peer review that most engineers involved with design are subjected to is ridiculous. Generally, we’re standing in front of peers (read: more experienced engineers) that have a goal of finding problems. For one design, say a digital PCB with 18 layers and 10k nets, said engineer is getting vetted 3 or 4 times, with ever increasing scrutiny, eventually standing in front of the customer (with tight purse strings). The reason? Design is cheap, while building a flawed system can easily dwarf the non-recurring (i.e. design) costs 100:1 or even 1000:1.

    Mark

  36. Posted Mar 20, 2007 at 10:13 AM | Permalink

    Rob wrote:

    Dendroclimatologists are guilty of biased sampling – we have to be. To derive a valid temperature signal we must go to environments where tree growth is limited by temperature. The high elevation or latitude tree-lines being a good starting point. Even when going to these environments, site ecology (i.e. aspect, angle of slope and drainage) also needs to be taken into account as this will also influence how the trees respond to temperature.

    I understand the samples are not easy to get, but I would like to know why it is suddenly so much more difficult to to get those samples now than it was 27ish years ago? The AGW proponents, due to high profile of the theory, not to mention consequences to all if the theory is indeed correct, must have access to much more $$$$ presently than they did two plus decades ago. So cost can’t be a reason not to get the updates. It can’t be government regulation preventing the retrieval of up-to-date samples. Those government types love you guys right now. Would probably kiss your feet if you ask them (and yes, I’m a little jealous of your scope of influence).

    Why does the reluctance to update the proxies remind me of the Barry Bonds vs Steroid debate.

    Baseball Fan: “I think you used steroids to help hit more home runs”.

    Bonds: “No I didn’t! I have never tested positive for steroids”.

    Baseball Fan: “Those old tests didn’t test for the stuff you are accused of using. Plus there were many ways to mask the Balco brand steroids. Why don’t we retest the old samples with current tests that are more sensitive, or maybe test you for HGH”.

    Bonds: “No Way! That’s an invasion of my privacy, and you people aren’t smart enough to understand what it’s like being Barry Bonds. The matter is settled – we should just move on…”.

  37. Doug
    Posted Mar 20, 2007 at 10:15 AM | Permalink

    Audit done by a friend contributes no credibility to a study. Audit by a skeptic does.

  38. Steve Sadlov
    Posted Mar 20, 2007 at 10:27 AM | Permalink

    RE: Steve M – “(One of my standing predictions is that Sheep Mountain bristlecone ring widths did not go off the charts during the warm 1990s.)”

    I’d urge some caution. In much of California, especially the part north of 35 deg N latitude, the 90s were unusually wet. With the exceptions of 1991 and 1994, snow pack was good and lasted late. During a number of 90s years, we had good monsoons. Ergo, greater-than-normal moisture availability in the late July – late September time frame.

  39. Douglas Hoyt
    Posted Mar 20, 2007 at 11:55 AM | Permalink

    Buentgen says: “It seems interesting, however, that the observed decoupling between proxy and target [i.e., instrumental] data coincides with the timing of homogenization changes applied to the instrumental temperature data (warming before 1840 and after 1960, with cooling in between), with meteorological observations generally providing higher quality during the late twentieth century (A06; Böhm et al. 2001).”

    I think the instrumental data is of lower quality in recent years due to UHI effects that have not been removed and that is why the proxies and instrumental data are no longer tracking together.

    In China, we have a similar situation. CRU claims a 0.00055 C/year warming there due to UHI. A recent study by Ren et al. puts the UHI effect at 65% to 80% of the trend since 1960, or about 0.0155 to 0.019 C/year, a factor of 27 to 34 times larger. If CRU can make a mistake that large in China, then assuredly they can do so elsewhere.

    Ren G. Y., Z. Y. Chu, Z. H. Chen, Y. Y. Ren (2007), Implications of temporal change in urban heat island intensity observed at Beijing and Wuhan stations, Geophys. Res. Lett., 34, L05711, doi:10.1029/2006GL027927.

  40. Steve McIntyre
    Posted Mar 20, 2007 at 12:03 PM | Permalink

    Rob, you pointed to the many archived sites at WDCP. However, to take the prominent and interesting study D’Arrigo et al 2006 with which you were involved, the archival record is very deficient. In the Alaska sites and northern Canadian sites, 8 groups of sites are listed in Table 1.

    Seward – I can pretty much reconcile this listing to the sites identified in the reference D’Arrigo et al Clim Dyn 2005, although the site ak046 appears to have been added to the cited lists. This should be mentioned in a SI.2.

    NW North Alaska – no citation is given; as far as I can tell, none of the sites are archived nor are the sites identified anywhere.

    Yukon- there is no citation. again I am unable to locate any listing of sites or archive of sites. Twisted Tree has obviously been updated, but there is no archive. Is Ha

    Central Alaska – again there is no citation nor any identification of sites, nor do the sites appear to have been archived.

    Wrangells – Davi et al 2003 is a citation and most of the chronologies from that study appear to have been archived. There are some puzzling inconsistencies: for example, one of the sites Windy Wolf reports MXD but not RW – do you know why? Also 3 sites do not report MXD, while all the others do. There is no statement that MXD was or wasn’t done at these 3 sites and I presume that RW was rone at Windy Wolf. The number of cores from the listed sites exeeds the number of cores said to have been used in DArrigo et al. Why is that?

    Cosatal AK – The citation here is to G. Wiles et al., Tree-ring evidence for a medieval warm period along the southern coast of Alaska, manuscript in preparation, 2005. This is a title that will interest many readers of this site. Do you know the status of this manuscript in presentation that you cited?

    Central NWT- again no citation, listing of sites, or archived sites.

    Southern Alaska – again no citation, listing of sites or archiving.

    Your own data from Icefields has not been archived either.

  41. John Hekman
    Posted Mar 20, 2007 at 12:39 PM | Permalink

    Rob Wilson seems to like some things about the Buentgen study, citing the 2003 being the warmest year and saying that the proxies do track with the instumental record. Yet Buentgen is saying that the long term proxy records are tracking solar activity. Also, there is clearly no hockey stick in Buentgen’s study. So does Rob respect this study or not?

  42. Follow the Money
    Posted Mar 20, 2007 at 1:31 PM | Permalink

    #3 Rob:

    “Dendroclimatologists are guilty of biased sampling – we have to be.”

    False humility will get you … not very far.

    My suggestion: a flurry of grant applications implicitly promising papers with wild, disasterous predictions.

    Might as well cash in before the Carbon Lobby cashes out in a few years.

  43. Michael Jankowski
    Posted Mar 20, 2007 at 2:08 PM | Permalink

    one cannot randomly sample trees from any location and expect there to be a valid climate signal (temperature or precipitation)

    Of course, the trees and site in question are from the World Data Center for Paleoclimatology. One would normally assume that “climatology” data contains a “valid climate signal,” or it wouldn’t be a climatology dataset, would it?

  44. Steve McIntyre
    Posted Mar 20, 2007 at 2:29 PM | Permalink

    one cannot randomly sample trees from any location and expect there to be a valid climate signal (temperature or precipitation)

    I guess that would apply to carrying out principal components on the ITRDB North American data set as well, but I guess no dendrochronologist would want to grasp that nettle. He wouldn’t make any “friends”.

  45. Don Keiller
    Posted Mar 20, 2007 at 4:09 PM | Permalink

    Re “The Hockey Stick”
    Mann et al make extraordinary claims. These require extraordinary proof.
    Proof that I have yet to see….

  46. Ken Fritsch
    Posted Mar 20, 2007 at 4:19 PM | Permalink

    Re: #3

    you are confusing the issue again.
    Your little ‘spiel’ above is meaningless without some site information.
    Where are the spruce sites located (?Map)?
    How close are these sites to the high elevation or high latitude tree-line?
    For what purposes were these sites originally sampled?

    RW, Steve M has raised an issue and you as an expert in the area are posing questions that one would hope that you would instead be answering. Give us some insights here and direct us to what you think these TR samples consist of and let us know specifically what is being done to cover the recent 20 odd years of calibrating TR versus temperature (locally preferably). You, like Judith Curry do not need any reassurance from this board and the crowd here is not about to give any. We are looking for answers and insights when workers in the field such as yourself appear and are not particularly interested in your assessments of the persona here.

  47. Steve McIntyre
    Posted Mar 20, 2007 at 4:43 PM | Permalink

    I was doing some calculations on the Jacoby sites for which no chronologies were archived. My chronology failed on one of the Jacoby sites ak053. I followed the calculation piece by piece – this sort of babysitting is very time intensity – and eventually tracked the problem back to the original file ftp://ftp.ncdc.noaa.gov/pub/data/paleo/treering/measurements/northamerica/usa/ak053.rwl
    where all the ring widths of series P3-254 are negative.

    Now this sort of error probably doesn’t “matter” in the total scheme of things, but how does this sort of error occur. Shouldn’t it cause a problem for Rob Wilson as well?

    P3-254 1889 -680
    P3-254 1890 -1231 -779 -661 -734 -947 -994 -841 -725 -787 -690
    P3-254 1900 -700 -910 -851 -1222 -963 -838 -1354 -1307 -1313 -905
    P3-254 1910 -701 -593 -781 -921 -1209 -1030 -920 -928 -693 -641
    P3-254 1920 -929 -582 -732 -1309 -683 -990 -960 -711 -648 -749
    P3-254 1930 -1186 -807 -789 -682 -755 -727 -617 -429 -402 -576
    P3-254 1940 -695 -535 -502 -485 -836 -644 -578 -263 -353 -500
    P3-254 1950 -391 -242 -346 -458 -270 -255 -180 -242 -300 -120
    P3-254 1960 -188 -172 -165 -186 -235 -441 -340 -219 -180 -114
    P3-254 1970 -125 -80 -238 -162 -118 -157 -126 -110 -160 -64
    P3-254 1980 -91 -168 -287 -278 -239 -176 -373 -331 -204 -182
    P3-254 1990 -263 -342 -406 -351 -266 -227 -1 -263 -234 -187
    P3-254 2000 -207 -9999
    P3-217 1870 401 532 532 646 604 532 1132 779 790 558
    P3-217 1880 482 460 811 737 449 655 519 646 697 609

    There is a COFECHA report for this site at ftp://ftp.ncdc.noaa.gov/pub/data/paleo/treering/measurements/correlation-stats/ak053.txt which shows clearly that the mean ring width for P3-254 is negative.

    A couple of years ago, I pointed out to Rosanne D’Arrigo that the chronologies for several sites (ones used by Mann) did not match the measurement data – something that Mann would have noticed had he performed the quality control procedures that he claimed to have carried out. D’Arrigo agreed and said that the chronologies would be corrected, but they remain uncorrected to this day.

  48. John Lang
    Posted Mar 20, 2007 at 7:06 PM | Permalink

    It does look like these chronologies are inversely correlated with temperature.

    If you redid the Mann analysis with the updated data (particularly after 1980), you might think the hockey stick should then be upside down or bent upwards more on the stick end (versus the blade for you non-Canadians).

    But Mann would just insert these tree ring chronologies with a negative coefficient, therefore, reinforcing the hockey stick. Such is life with data selection.

  49. Curious
    Posted Mar 20, 2007 at 7:32 PM | Permalink

    4. I do not want to be drawn into a debate about Mike Mann’s NH reconstruction. If you do not like it, fine – forget it. There are other NH records now available.

    Okay – can someone suggest a NH reconstruction that has been done with a valid sample and methods? Which one is currently recommended?

  50. Mike Pisaric
    Posted Mar 20, 2007 at 8:29 PM | Permalink

    As Rob points out in 3 and 28, Steve’s analysis is flawed!
    You can easily search out for yourself where the Meko sites are located using Google Earth. They are a long way from latitudinal tree line or the mountains! As Rob indicates much of Meko’s work has focused on moisture sensitive trees and these are no exception. All of these sites are located in and around the Peace-Athabasca delta in north eastern Alberta, just north of Fort McMurray and the oil sands development. Most of them appear to update previous collections by Stockton and Fritts; these were originally used to reconstruct past lake levels and not temperature (Stockton, C.W., Fritts, H.C. 1973. Long-term reconstruction of water level changes for Lake Athabasca by analysis of tree rings. Water Resources Bulletin 9(5): 1006-1027). My suspicion is that the updated files and the new collections are for the same purpose and were never intended to be used for temperature reconstructions as Steve has implied and done. As it is routinely suggested on these pages that MBH’s hockey stick is driven by suspect chronologies, I would suggest that Steve’s analysis is exactly that. I hope Steve will acknowledge this and concede that his analysis is flawed. The eight Canadian sites he has included are not going to show a temperature signal and yet they will of course dominate the chronology he developed given they represent almost 60% of the data he included in his analysis.

    A couple of corrections to other posts above:

    Steve’s title is incorrect: PCGL is white spruce, PCMA is black spruce (Picea mariana). Different species with varied ecological likes and dislikes. I am sure this was just an oversight though.

    #43 – Tree data comes from the International Tree Ring Databank (ITRDB) which is maintained (housed) by the NOAA Paleoclimatology Program and World Data Center for Paleoclimatology. The ITRDB contains tree ring chronologies that have been used for a multitude of purposes and not just paleoclimatology. So no, you would not expect all tree ring chronologies from the ITRDB to contain a valid climate signal.

  51. Jim Clarke
    Posted Mar 20, 2007 at 9:21 PM | Permalink

    Stan Palmer,

    The difference between AGW and building an atomic bomb or sending men to the moon is that the latter two are definable goals with real scientific solutions. What is the ‘solution’ to man-made climate change? How do you solve a problem that can not even be quantified and only exists as a potential threat in computer models? What exactly would the researchers do in such a project?

  52. Steve McIntyre
    Posted Mar 20, 2007 at 9:31 PM | Permalink

    #43 – Tree data comes from the International Tree Ring Databank (ITRDB) which is maintained (housed) by the NOAA Paleoclimatology Program and World Data Center for Paleoclimatology. The ITRDB contains tree ring chronologies that have been used for a multitude of purposes and not just paleoclimatology. So no, you would not expect all tree ring chronologies from the ITRDB to contain a valid climate signal.

    So why didn’t you folks make that observation about MBH98?

  53. Steve McIntyre
    Posted Mar 20, 2007 at 9:47 PM | Permalink

    OK, can we stipulate that the JAcoby PCGL (white spruce – thanks) sites are supposed to be good temperature proxies. Unfortunately, Jacoby didn’t archive his chronologies although these chronologies are referred to in the quality control. However, Jacoby’s usual method is to fit negative exponentials which he calls “conservative” standardization. I’ve written an R function and been able to benchmark my routine against COFECHA (which has some odd numerical features in its nonlinear routines but that’s a story for another day.) So just for fun, I did Jacoby “conservative” standardization on the 30 PCGL sites for which no chronology was archived (the calculation failed on the 31st site which had negative ring widths and this site is excluded from my average.) These are mostly Jacoby sites and would all have been used in D’Arrigo et al 2006. Here’s the result:

    Yeah, yeah, Rob will say that you should use RCS standardization – OK, but, to my knowledge, there isn’t a single RCS chronology archived at the ITRDB databank. Also, Jacoby “conservative” standardization should preserve 20th century relationships. I’ll try to get to RCS standardization on another occasion. The difficulty here is that D’Arrigo et al 2006 (example) mentions a lot of variation in how this is done; Rob has criticized by implementation of RCS in the past. In the absence of any objective documentation of what was done, I do the best I can.

    In any event, the 20th century portion of this series looks not dissimilar to the earlier series, which critics have said reflects moisture-stressed sites.

  54. Steve McIntyre
    Posted Mar 20, 2007 at 10:14 PM | Permalink

    Here’s the script and list of sites in the above figure. I wouldn’t want Rob to criticize me for not listing sites (although, by an oversight, such a listing is absent from D’Arrigo et al 2006). The script is not self-executing online as it relies on pre-calculated chronologies and measurements in R-format. It documents the calculation and runs on my computer.

    ##AVERAGE OF JACOBY-TYPE CHRONOLOGIES FOR JACOBY PCGL SITES ENDING GE 1998

    ### this script is not self-executing on the web as it relies on library of tree ring measurement series and chronologies
    ### in R-format which I have on local drive in my computer
    ### which have taken a fair bit of time to prepare and quality control
    ### I would urge ITRDB to archive this information in a consistent and QC’ed format as well

    ###LOAD INFO
    library(nlme)
    load(“d:/climate/data/tree/northamerica.details.tab”) #this is prestored

    ##IDENTIFY SERIES ENDING GE 1998
    temp=1998)&!is.na(details$end);sum(temp)
    test< -details[temp,]
    N<-nchar(as.character(test$id))
    temp1<-!(substr(as.character(test$id),N,N)=="a") &!(substr(as.character(test$id),N,N)=="r")
    test<-test[temp1,1:8]
    id<-as.character(test$id)
    id<-sort(id) ;length(id)#181
    x<-tapply(details$id[temp],details$type[temp],length)
    #name0<-names(x);name0[x<15]<-NA
    #ABAM ABLA ACRU CHLA CHNO JUOC LALY PCEN PCGL PCMA PCSP PIAL PIBN PICO PIEC PIED PIFL
    # 2 4 1 1 3 12 1 6 48 7 1 1 6 6 2 25 9
    #PIPO PIRE PIST PLRA PPGR PSME QUAL QULO QUMA QURU QUST TAMU TSCA
    # 35 1 1 1 1 66 7 1 2 2 1 3 1

    ##NOW IDENTIFY PCGL SERIES ENDING GE 1998
    type0<-"PCGL" #PSME
    temp=1998)&!is.na(details$end)
    test< -details[temp,]
    N<-nchar(as.character(test$id))
    temp1<-!(substr(as.character(test$id),N,N)=="a") &!(substr(as.character(test$id),N,N)=="r")
    test<-test[temp1,1:8]
    id<-as.character(test$id); N<-length(id);N #45 sites

    #MAKE AVERAGE OF ARCHIVED CHRONOLOGIES
    example="northamerica";
    chron<-NULL;stat1<-rep(TRUE,N)
    for (i in 1:N) {
    loc<-file.path("d:/climate/data/tree/chronologies/r",example,paste(id[i],"crn.tab",sep=".") )
    test<-try(load(loc))
    if( class(test)=="try-error") stat1[i]=FALSE else chron<-ts.union(chron,chron.crn)
    }
    dimnames(chron)[[2]]<-id[stat1]
    dim(chron) #433 14
    #missing many crn series
    details[!is.na(match(details$id,id[stat1])),]
    #all from non-Jacoby except two with Wiles ak075; ak072x

    #MAKE JACOBY-TYPE CHRONOLOGIES FOR THE SERIES WITH MEASUREMENT DATA AND NO ARCHIVED CHRONOLOGIES
    indexid<-id[!stat1];N<-length(indexid)
    chron<-NULL;stat4<-rep(TRUE,N)
    for (i in 1:N) {
    loc<-file.path("d:/climate/data/tree/measurements/r",example,paste(indexid[i],"rwl.tab",sep=".") )
    test<-try(load(loc))
    test<-try(jacoby.chronology(tree))
    if(class(test)=="try-error") stat4[i]<-FALSE else chron<-ts.union(chron,test$series)
    }
    dimnames(chron)[[2]]<-indexid[stat4]
    chron.jacoby<-chron
    #save(chron,file="d:/climate/data/blog/chron.jacoby.tab")

    ##ANALYSIS OF PROBLEMATIC SERIES WHERE CALCULATION FAILED
    i<-23;
    loc<-file.path("d:/climate/data/tree/measurements/r",example,paste(indexid[i],"rwl.tab",sep=".") )
    test<-try(load(loc))
    #the Jacoby chronology was run one core at a time until it failed at:
    #"P3-254"
    #all the values of this core were negative !!

    ##CALCULATE MEAN OF JACOBY SERIES
    chron.mean<-ts(apply(chron.jacoby,1,mean,na.rm=TRUE),start=tsp(chron.jacoby)[1])
    index<-tsp(chron.jacoby)[1]:tsp(chron.jacoby)[2]
    ylim0=c(.4,1.800)
    chron<-chron.jacoby

    ##PLOT FIGURE POSTED AT BLOG
    #png("d:/climate/images/proxy/pcgl.jacoby.png",width=600,height=300)
    nf<-layout(array(1:2,dim=c(1,2)),widths=c(3,1.3))
    par(mar=c(3,3,1,1))
    plot(tsp(chron)[1]:tsp(chron)[2],chron.mean,type="l",xlab="",ylab="",col="grey80",ylim=ylim0)#xlim=c(800,2000))#,ylim=ylim0)
    f<-function(x) {f<-filter.combine.pad(x,truncated.gauss.weights(21))[,2];f}
    lines(index,f(chron.mean),col="red",lwd=2)

    par(mar=c(3,0,1,1))
    plot(1900:tsp(chron)[2],chron.mean[(1900:tsp(chron)[2])-tsp(chron)[1]+1],type="l",xlab="",ylab="",col="grey80",ylim=ylim0,axes=FALSE)
    lines(1900:tsp(chron)[2],f(chron.mean)[(1900:tsp(chron)[2])-tsp(chron)[1]+1],col="red",lwd=2)
    axis(side=1);axis(side=2,labels=FALSE);box()
    #dev.off()

    ##LIST SITES IN THIS CALCULATION
    temp<-!is.na(match(details$id,indexid))
    details[temp,]

    Now here is the list of sites:

    id location type lat long altitude start end author
    11011 ak057 Almond Butter Lower PCGL 65.11 -162.12 168 1607 2002 R. D’Arrigo E. Mashig D. Frank R. W
    11012 ak058 Almond Butter Upper PCGL 65.11 -162.13 213 1406 2002 R. D’Arrigo E. Mashig D. Frank R. W
    11013 ak059 Alpine View PCGL 65.07 -162.11 282 1542 2002 R. D’Arrigo E. Mashig D. Frank R. W
    11014 ak060 Burnt Over PCGL 65.13 -162.15 259 1621 2002 R. D’Arrigo E. Mashig D. Frank R. W
    11015 ak061 Bye Rosanne PCGL 65.05 -162.11 282 1575 2002 R. D’Arrigo E. Mashig D. Frank R. W
    11016 ak062 Death Valley PCGL 65.12 -162.16 239 1358 2002 R. D’Arrigo E. Mashig D. Frank R. W
    11017 ak063 Echo Slope PCGL 65.06 -162.09 229 1590 2002 R. D’Arrigo E. Mashig D. Frank R. W
    11018 ak064 Frost Valley PCGL 65.05 -162.09 229 1611 2002 R. D’Arrigo E. Mashig D. Frank R. W
    11019 ak065 Gordon’s Cat PCGL 65.12 -162.12 168 1400 2002 R. D’Arrigo E. Mashig D. Frank R. W
    11020 ak066 Hey Bear PCGL 65.13 -162.13 213 1533 2002 R. D’Arrigo E. Mashig D. Frank R. W
    11021 ak067 Hey Bear Upper PCGL 65.13 -162.13 229 1383 2002 R. D’Arrigo E. Mashig D. Frank R. W
    11022 ak068 Mt. Mole PCGL 65.05 -162.11 229 1550 2002 R. D’Arrigo E. Mashig D. Frank R. W
    11023 ak069 Ptarmigan Hill PCGL 65.04 -162.14 244 1718 2002 R. D’Arrigo E. Mashig D. Frank R. W
    11024 ak070 Windy Ridge PCGL 65.11 -162.13 251 1556 2002 R. D’Arrigo E. Mashig D. Frank R. W
    11038 cana209 Eyeglass Lake PCGL 57.55 -61.36 100 1713 2001 Brendan Buckley Rosanne D’Arrigo
    11054 cana211 Hazel Creek PCGL 61.54 -140.43 731 1679 1999 Gordon Jacoby Nicole Davi Greg Wiles
    11001 ak047 Firth River PCGL 68.06 -141.06 100 1676 2002 Martin Wilmking
    11002 ak048 Chimney Lake PCGL 67.07 -150.05 100 1725 2001 Martin Wilmking
    11003 ak049 North Fork Koyukuk PCGL 67.09 -150.05 100 1715 2001 Martin Wilmking
    11004 ak050 Nutirwik Creek PCGL 67.09 -149.08 100 1731 2000 Martin Wilmking
    11005 ak051 Nutirwik Creek Floodplain PCGL 67.09 -149.08 100 1828 2001 Martin Wilmking
    11006 ak052 Hunt Fork of John River PCGL 67.08 -152.04 100 1730 2002 Martin Wilmking
    11007 ak053 Sheenjek River PCGL 68.05 -143.08 100 1715 2000 Janet Joergenson Martin Wilmking
    11008 ak054 Cabin Creek, Denali NP PCGL 63.06 -150.00 100 1735 2000 Martin Wilmking Glenn Juday
    11009 ak055 Slime Creek, Alaska Range PCGL 63.05 -148.08 100 1788 2002 Martin Wilmking
    11010 ak056 Rock Creek, Denali NP PCGL 63.07 -149.00 100 1735 2000 Martin Wilmking, Glenn Juday, Jens Ibendo
    11026 cana197 Bolen Lake PCGL 57.52 -103.48 425 1852 2002 A. Beriault D. Sauchyn J.Stroich
    11027 cana198 Doupe Bay, Jan Lake PCGL 54.56 -102.47 315 1839 2001 A. Beriault D. Sauchyn J.Stroich
    11031 cana202 Kinapik Island, Sandfly Lake PCGL 55.42 -106.26 390 1840 2001 A. Beriault D. Sauchyn J.Stroich
    11034 cana205 Otter Rapids PCGL 55.38 -104.44 360 1879 2001 A. Beriault D. Sauchyn J.Stroich
    11035 cana206 Patterson Peninsula, Lac La Ron PCGL 55.13 -104.32 370 1822 2001 A. Beriault D. Sauchyn J.Stroich

  55. Observer
    Posted Mar 21, 2007 at 1:32 AM | Permalink

    As a lawyer, I am still absolutely stunned by the lack of simple data and methodology that has not been disclosed STILL TO THIS DAY on an issue like this.

    In the most simple cases, both civil and criminal, every single piece of data is disclosed. I’m not referring about capital murder and billion dollar class actions. This is in even slip and fall and B&E trials. If it is withheld, lawyers and clients can be sanctioned. Licenses to practice altogether can be revoked.

    Even in these tiny hearings, every piece of data and every methodological step is disclosed and subject to examination by all.

    By contrast, this is literally a multi-trillion dollar issue. If the UN IPCC Working Group 2 is to be believed “Hundreds of millions of Africans and tens of millions of Latin Americans who now have water will be short of it in less than 20 years.”

    Yet there are no available sites that simply contain all of the actual data used in the studies upon which the U.N. IPCC is basing it’s reports.

    No sites that contain which temperature-sensitive trees have been selected.

    No sites explaining the exact methodologies used in detail for any of these studies. One must go to the original broad and extremely general acadenuc “papers” on the subject just to get a very generalized picture of the methods and data. No replication files are provided. No even after-the-fact step-by-step methodologies are produced now to support their findings. Usually, nothing even as basic as “here is every data file I used, here is every script, and here were the steps I took.”

    It seems OUTRAGEOUS — and I used that term because I know of none greater — that the collection of world governments is asked to make multi-TRILLION dollar decisions without even the miniscule first step of providing this basic underlying data and detailed methodological steps. As would be provided by experts in the most simple criminal and civil hearings.

    So much so that I’m still, years later, absolutely stunned about this fact. I’ve told lawyers about this at times. That they can’t simply obtain all of the data and scripts used by the experts putting together the studies underlying the IPCC. No such site exists. And the study originators STILL TO THIS DAY provide no step by step methodologies and scripts showing exactly what was done. They’re blown away when they hear it. Many don’t even believe that to be the case.

    I think some people on both sides of this issue have spent so much time in, or examining, this industry that they forget just how outrageous this appears to people operating in the outside world, where even tiny transactions and hearings must be documented to the line and letter.

  56. Willis Eschenbach
    Posted Mar 21, 2007 at 1:36 AM | Permalink

    Rob W., thank you kindly for contributing to this thread. I was interested in your comment:

    Dendroclimatologists are guilty of biased sampling – we have to be. To derive a valid temperature signal we must go to environments where tree growth is limited by temperature. The high elevation or latitude tree-lines being a good starting point. Even when going to these environments, site ecology (i.e. aspect, angle of slope and drainage) also needs to be taken into account as this will also influence how the trees respond to temperature.

    Rob, do you pick the sites first based on the criteria and then take use all of the available samples from that site, without first looking at whether the tree rings track local (or global) temperature? Or do you pick a general location that fits the criteria, and then see which trees or stands in that location show a correlation with local (or global) temperature, and just use those?

    Also, as the Juckes example shows, some dendroclimatologists are picking sites/trees that have no correlation with local temperature, but are nonetheless claimed have a correlation with global temperature.

    In a field which has been around as long as dendroclimatology, surely there must be some text or study that lays out the general rules for site selection. What are those rules?

    Many thanks,

    w.

  57. Observer
    Posted Mar 21, 2007 at 1:50 AM | Permalink

    First, I apologize for some of the typos above.

    Second, were I a combative skeptic — and I’m actually not — I would make this my SOLE issue. Forget everything else. I would repeat it like a mantra to voters.

    A skeptical politician(s) repeating over and over and over “I’m sorry, but I’m not going to pledge a single dime of taxpayer money to this issue if we can’t simply get all of the underlying data and detailed line-by-line methodologies and scripts of the studies relied upon. As we’d get even in any simple slip and fall court case. Until this miniscule first step is taken, it is absolutely outrageous that one would even ask to spend incredible amounts of taxpayer money and institute incredibly onerous regulations. Period.”

    Make that speech 10 times a day when the issue is raised, say nothing else (don’t cloud the issue) and then focus upon how even this ridiculously simple request is not being met as time passes.

  58. Don Keiller
    Posted Mar 21, 2007 at 3:11 AM | Permalink

    #56 “Also, as the Juckes example shows, some dendroclimatologists are picking sites/trees that have no correlation with local temperature, but are nonetheless claimed have a correlation with global temperature.”
    Forgive me, I must have stumbled onto the wrong blog by mistake. I thought this was about paleoclimatology. Instead I find that some people (Juckes) are talking about Quantum Physics- namely “Spooky action at a distance”.

  59. Andrey Levin
    Posted Mar 21, 2007 at 3:29 AM | Permalink

    Re: 55,57

    Observer, I share your outrage, as many others.

    Note, that most temperature and other data used by IPCC are supplied by East Anglia University, and two US government agencies: NOAA and NASA. US federal agencies are subject to Data Quality Act:

    http://library.findlaw.com/2003/Jan/14/132464.html

    Regards.

  60. Jean S
    Posted Mar 21, 2007 at 3:53 AM | Permalink

    re #50: Mike P.:

    never intended to be used for temperature reconstructions

    Why don’t you give an authorative comment on validity of these series as temperature proxies:

    http://www.ncdc.noaa.gov/paleo/ei/data_supp.html

    If you are consistent with your views, I’d expect you to state loud and clear that MBH network is flawed for the use in temperature reconstructions. I won’t hold my breath though.

  61. Jean S
    Posted Mar 21, 2007 at 4:11 AM | Permalink

    What really bugs me about comments like #14 (Rob) and #50 (Mike P.) is the fact that these guys have nerve to come to nitpick Steve’s blog-posts about mistakes commonly found in published papers (even in journals like Nature and Science) in their own field. Yet, these same guys keep their mouths shut about these papers, and even worse, they might have been peer reviewers for those papers. I wish they’d consider pasteing their own pictures next to the word “hypocrite” in a dictionary.

  62. fFreddy
    Posted Mar 21, 2007 at 4:36 AM | Permalink

    Re #55, #57, Observer
    I thoroughly agree.
    Which raises the obvious question, can you think of any legal basis to bring some sort of action against the Hockey Team, so that they can be made subject to discovery ?

  63. bruce
    Posted Mar 21, 2007 at 5:48 AM | Permalink

    Re #62: In Australia, we have a Trade Practices Act which makes it a crime to engage in “False and misleading conduct”. It is abundantly clear to any observer that “False and misleading conduct” has been engaged in. And it is also arguable that the reason for that happening has been to secure significant benefits, being access to grant moneys. A lawyer may correct me, but I think that the technical term for that is ‘Fraud’.

    I am not a lawyer, but I would be amazed if similar provisions are not fundamental to the laws of USA, Canada, UK, Europe etc. Building on the cogent points made by “Observer” in posts #55 and #57, perhaps the way to rapidly secure commitment to truth might be to press charges against the Team for “False and misleading conduct”. No doubt those characters would indignantly profess that they are doing the right thing. However, as Observer points out. Courts are designed to deal with focussed forensics designed at getting at the truth.

    I would suggest that if the Team were to consult lawyers, they might find that they are in fact in jeopardy.

    Of course, I have asked the question before. But are not climate scientists responsible to some professional body for maintaining professional standards? Would they not face difficult ethics charges if they breached those standards??

  64. Steve McIntyre
    Posted Mar 21, 2007 at 7:24 AM | Permalink

    Observer, I have much experience in what’s involved in business audits, prospectuses and litigation, together with all the standards that you correctly describe. That perspective very much animated my initial sense of astonishment at the lack of due diligence and the inadequate disclosure involved in Team climate science. Academics without that perspective simply didn’t understand it. From what you write, it sounds like my sense of shock has become attenuated over the last couple of years. I hope not.

    Academics described me as an “amateur”, but I had a lot of experience that was simply different from working in universities. I knew what actual due diligence was like. I used to say that I felt like an anthropologist in the world of academics. I viewed peer review as carried out by journals in more general terms – not as an end in itself, but as a form of due diligence. Viewed in those terms, it is a very cursory form of due diligence; yes, it’s better than nothing, but not much. Also, much of the energy of academic peer review is devoted, not to what I view as useful activities, like ensuring that the results are replicable and documentable, but towards whether the results are consistent with the POV of the reviewer.

  65. Michael Jankowski
    Posted Mar 21, 2007 at 7:51 AM | Permalink

    Re#50

    #43 – Tree data comes from the International Tree Ring Databank (ITRDB) which is maintained (housed) by the NOAA Paleoclimatology Program and World Data Center for Paleoclimatology.

    “Maintained,” as the site words it, and “housed,” which you equate to be the same thing, seem like two drastically different things to me.

    Regardless, it makes zero sense to me for a Paleoclimatology website to either maintain or even house tree ring data that is not safe to assume is not related to climate.

    That’s not to say its safe to assume the data directly correlates specifically to temperature, etc.

  66. Steve McIntyre
    Posted Mar 21, 2007 at 7:53 AM | Permalink

    Jean S, Rob Wilson is a good guy who I’m on good terms with, for which Rob gets enormous criticism within his trade guild. So please tone down the rhetoric a little.

    Having said that, I agree 100% with you that it is absolutely remarkable that a blog post should attract such sharp criticism – that it was erroneous to use ITRDB tree ring data without screening – while failing to make the same criticism of MBH98. To your link, I would also add the following link describing MBH procedures. It also includes the following statement, which I’ve referred to in the past, but deserves repetition:

    MBH98 found through statistical proxy network sensitivity estimates that skillful NH reconstructions were possible without using any dendroclimatic data, with results that were quite similar to those shown by MBH98 based on the full multiproxy network (with dendroclimatic indicators) if no dendroclimatic indicators were used at all.

    How this is said with a straight face given their prior knowledge of bristlecone dependence just amazes me. And we don’t see any dendrochronologists calling them out on this – instead I get lectured about not making “friends” and get that as an excuse as to why they won’t disclose their data.

  67. Vasco
    Posted Mar 21, 2007 at 8:16 AM | Permalink

    AGW has become a religion. A conviction based on faith. All those heretics who suspect The Word of Gaia will be damned by the righteous. There is no need to explain a religion as long as you believe in it and follow Rev.Gore.

    Thou will be punished by the Revenge of Gaia!

  68. Steve McIntyre
    Posted Mar 21, 2007 at 9:21 AM | Permalink

    Here’s a plot of the up-to-date archived PCGL chronologies in my head post versus my Jacoby-type chronology calculation for the Jacoby etc sites for which measurements are archived (but not chronologies.) There are obviously points in common. Curiously the levels are particularly high in the mid-1960s, often regarded as a cool period. It’s odd that some periods synchronize and others don’t.

  69. Steve Sadlov
    Posted Mar 21, 2007 at 9:51 AM | Permalink

    RE: #64 and previous relevent posts – let us call the problem what it is, lawlessness. There is a certain extra-legal aspect to what the mass of hysterics and those stirring them up are doing in their zeal to “prosecute” killer AGW. If we don’t stop this, a certain unfortunate precedent will have been set, and the legal systems of nearly all nations who follow the English Common Law tradition will have been made less potent by it.

  70. Rob Wilson
    Posted Mar 22, 2007 at 3:32 AM | Permalink

    Morning,
    I was away yesterday so missed all the ‘fun’ on the blog.
    There are obviously lots of points that have been raised and I will briefly address those ones I think relevant. However, I cannot afford any more time to this thread, so this will be my last response. Unfortunately, I feel that whatever I or others might say will never be enough to sate your lust for criticism. I can only urge the readers of this blog to spend more time reading the original articles and spend a little less time on this or other blogs. I perhaps should take my own advise to heart.

    1. For the record, I personally think that a proxy record (tree-ring or otherwise) should represent local climatic conditions (e.g. temperature). However, there are situations where a proxy site may not be situated close to a meteorological station. Therefore, if a tree-ring chronology, sampled taking ecology in mind (i.e. high elevation tree-line, north facing slope etc) to optimise its ‘potential’ temperature signal, does not correlate with the nearest met. station, which could be several hundred kilometres away, does that mean the site should not be used?
    However, if a site is close to a met. station and still does not correlate with climate, then it should not be used for proxy climate reconstructions. But this is my opinion and others may not agree with this. It is perhaps an interesting story in itself to explore why a site does not portray a climate signal when it perhaps should – but this is more of an ecological question.

    2. As far as I am aware, data-sets are being updated. Many of the North American sites were updated for DWJ2006. I am currently funded through the EU and we (many TR labs) are updating many sites in Europe as well as sampling new ones.

    3. The Buentgen Alpine (RW and MXD) records are very important. They track the local temperature record well and the reconstructed cold periods coincide well with periods of solar minima and also major volcanic events – not unlike the Luckman and Wilson (2005) Icefields study. For the next generation of NH reconstructions, I believe we need many more records of similar quality.

    4. The negative values in AK053 are obviously incorrect. Please contact Bruce Bauer (Bruce.A.Bauer@noaa.gov) and he can rectify this situation.

    5. Thanks to Mike Pisaric for also stepping into the fray. I wish more dendrochronologists would occasionally address the issues raised in CA. Criticism is fine, but misinformation is counter productive.

    6. The Icefields reconstructions and STD and RCS (oops there is one archived!!) chronologies are archived. The raw data are not.
    The Wiles study on the MWP for the Gulf of Alaska is not yet published. It is a student’s thesis and is still in preparation. However, I have recently utilised these data to look at decadal and multi-decadal variability in the region:
    Wilson, R., Wiles, G., D’Arrigo, R. and Zweck, C. 2007. Cycles and Shifts: 1300-years of multi-decadal temperature variability in the Gulf of Alaska. Climate Dynamics. 28:425’€”440. DOI 10.1007/s00382-006-0194-9.
    The PDF can be downloaded from the publications link at my website (http://www.geos.ed.ac.uk/people/person.html?indv=930)

    7. This whole blog started with me criticising Steve M about averaging 14 sites without expressing where the sites came from and why they were originally sampled. Your 30 chronology mean is no different I am afraid. Not all the data are from the Jacoby group. Many of these series are from Wilmking. Again, I do not know the details of the individual sites, but Wilmking is a dendroecologist and has been study the ‘divergence’ issue in recent years with respect to positive and negative responder trees.
    Also, Steve, your time-series is a real mess. Surely you can see that the variance of this time-series needs to be stabilised to account for the decrease in N back in time. But that is for another thread (http://www.climateaudit.org/?p=1279)
    Steve, if you are really interested in exploring the utility of these data as temperature (or other) proxies, then I would recommend you do a site by site analysis and see which of these chronologies express a valid climate signal. At the same time, for those sites which are apparent temperature proxies, you could then also address the ‘divergence’ issue at the site scale. This would truly be an interesting piece of work and may actually be a positive contribution to science.

    Back to being a hypocrite
    Rob

  71. Posted Mar 22, 2007 at 5:36 AM | Permalink

    Thanks for your time, Dr. Wilson!

    However, I cannot afford any more time to this thread, so this will be my last response.

    If you have time in the near future, I’d like to hear more about this variance stabilization:

    Also, Steve, your time-series is a real mess. Surely you can see that the variance of this time-series needs to be stabilised to account for the decrease in N back in time.

    Specially, what is wrong in our understanding, see the comments in

    http://www.climateaudit.org/?p=1279

    I understand you are busy, but there is EU money involved, after all. No one will complain to EC if you spend some time answering the questions raised here.

  72. Willis Eschenbach
    Posted Mar 22, 2007 at 6:15 AM | Permalink

    Rob, thanks as always for your participation. One of my questions didn’t get answered:

    In a field which has been around as long as dendroclimatology, surely there must be some text or study that lays out the general rules for site selection. What are those rules?

    Any insights on that?

    My best to you,

    w.

  73. Steve McIntyre
    Posted Mar 22, 2007 at 6:50 AM | Permalink

    Hi, rob, thanks for your comment. At the NAS Panel hearing, Malcolm Hughes contrasted two styles of reconstruction – what he called the Schweingruber approach, where one took samples of many sites expected to be “temperature sensitive” sites, and reported on the population; and what he called the Fritts approach, where you took a whole bunch of sites with a “climatic signal” and relied on software to extract a “signal”. Obviously Mannian PCs are an example of the latter. MAnnian PCs impart a strong bias to the result, but any sort of ex-post picking starts imparting bias, just as much as Mannian PCs.

    I’m quite happy to do averages using any form of ex ante criterion that you care to propose. I did this quick average to see what an average of PCGL updated series looked like. If you say that the relevant criterion is north-facing PCGL sites above a certain latitude or following an ex ante latitude-altitude relationship characteristic of temperature-sensitive sites, then fine. But enough of this after-the-fact picking.

  74. Steve McIntyre
    Posted Mar 22, 2007 at 7:34 AM | Permalink

    1. For the record, I personally think that a proxy record (tree-ring or otherwise) should represent local climatic conditions (e.g. temperature). However, there are situations where a proxy site may not be situated close to a meteorological station. Therefore, if a tree-ring chronology, sampled taking ecology in mind (i.e. high elevation tree-line, north facing slope etc) to optimise its potential’ temperature signal, does not correlate with the nearest met. station, which could be several hundred kilometres away, does that mean the site should not be used?

    It’s interesting that in the first tranche of Jacoby treeline sites in the 1989 study, Jacoby selected 10 of the 36 sites (and we never heard anything more of the other 26 on which Jacoby refused to provide information), Jacoby mentions in passing that they were allsouth-facing. This seemed to open the possibility of other influences besides temperature being at work.

    However, if a site is close to a met. station and still does not correlate with climate, then it should not be used for proxy climate reconstructions. But this is my opinion and others may not agree with this. It is perhaps an interesting story in itself to explore why a site does not portray a climate signal when it perhaps should – but this is more of an ecological question.

    If the site meets ex-ante criteria, then you should include it. OR you need to refine objective selection criteria. But you can’t do ex post elimination. Think about this sort of practice in the hands of drug tests or mining promotions. You have to report everything. The failure to understand this is perhaps the biggest single obstacle to proper statistical procedure in dendroclimatology.

    2. As far as I am aware, data-sets are being updated. Many of the North American sites were updated for DWJ2006. I am currently funded through the EU and we (many TR labs) are updating many sites in Europe as well as sampling new ones.

    Well, they are not being updated in the archives concurrent with publication. Twisted Tree Heartrot Hill has no information more recent than 1992 in the archive. Other Jacoby sites no information more recent than 1979. The Gaspe update isn’t archived. The measurements for the important Tornetrask update, or Taymir or Yamal aren’t archived. D’Arrigo et al 2006 does not even provide a list of sites.

    3. The Buentgen Alpine (RW and MXD) records are very important. They track the local temperature record well and the reconstructed cold periods coincide well with periods of solar minima and also major volcanic events – not unlike the Luckman and Wilson (2005) Icefields study. For the next generation of NH reconstructions, I believe we need many more records of similar quality.

    The Buentgen data is not archived. Do you know whether the altitudes of the samples changed over time; the altitude from which historic timbers came from? At Polar Urals, the medieval treeline was higher and samples came from higher altitudes. It seems like an important potential bias to take modern samples from lower altitudes. Is this allowed for in the Buentgen study? Didn’t think so. Did you record and archive altitudes of samples in the Icefields study? Didn’t think so. If you want the next generation of studies to be better than the last generation of studies, then you should annotate and archive a broader spectrum of relevant information for each sample. Take a geologist with you on your next sampling expedition and ask him to take notes and map things; he’ll know what to do. OR take a geology course and find out how to do proper maps. I can’t specify all the things that need to be done. It’s just that I’ve seen a lot of geological reports in my life; they are expected to be useful 15 years later or even 50 years later and usually are.

    4. The negative values in AK053 are obviously incorrect. Please contact Bruce Bauer (Bruce.A.Bauer@noaa.gov) and he can rectify this situation.

    I already had done this. I’m on excellent terms with Bruce Bauer and have done whatever I can to support the work that this archive does. The issue is really more towards the computer programs used in dendro: how did the negative values get introduced into the data bank? what is the defect in QC? It shouldn’t be a matter of ad hoc noticing negative values. Not that my observing errors causes anyone to change things: there is an error in the mesurement file for ftp://ftp.ncdc.noaa.gov/pub/data/paleo/treering/measurements/asia/russ021w.rwl Polar Urals that really puzzled me a couple of years ago; I pointed out the problem both to Schweingruber and to WDCP, but the error remains three years later, as do D’Arrigo chronologies which she said were wrong.

    5. Thanks to Mike Pisaric for also stepping into the fray. I wish more dendrochronologists would occasionally address the issues raised in CA. Criticism is fine, but misinformation is counter productive.

    I have no desire to convey any misinformation. If I’m reporting the results of some calculation, I try to show what I did. In this case, I provided scripts. IF either of you can suggest an ex ante criterion for site selection from PCGL sites, I’ll re-do the calculation. Of course, what’s sauce for the goose is sauce for the gander and I’d ask you about the inclusion of the Gaspe site in D’Arrigo et al 2006 (and MBH98). As you say, misinformation is counter-productive.

    6. The Icefields reconstructions and STD and RCS (oops there is one archived!!) chronologies are archived. The raw data are not.

    Well, to pick a nit, if you go to the ITRDB tree ring site http://hurricane.ncdc.noaa.gov/pls/paleo/fm_createpages.treering where tree-ring data is archived, the Icefields reconstructions and chronologies are not archived there, although, as you say, they are archived. As you say, the measurement data is not archived. It’s the same situation as Briffa with Yamal and Tornetrask. These reconstructions are archived at BRiffa’s website, but the measurement data is not. To my knowledge, there is NO archived RCS reconstruction with accompanying measurement data from which someone can benchmark RCS calculations.

    The Wiles study on the MWP for the Gulf of Alaska is not yet published. It is a student’s thesis and is still in preparation. However, I have recently utilised these data to look at decadal and multi-decadal variability in the region: Wilson, R., Wiles, G., D’Arrigo, R. and Zweck, C. 2007. Cycles and Shifts: 1300-years of multi-decadal temperature variability in the Gulf of Alaska. Climate Dynamics. 28:425’€”440. DOI 10.1007/s00382-006-0194-9.The PDF can be downloaded from the publications link at my website (http://www.geos.ed.ac.uk/people/person.html?indv=930)

    I look forward to reading another one of your excellent publications. IF I criticize things, please don’t interpret this as a lack of respect for your work.

    7. This whole blog started with me criticising Steve M about averaging 14 sites without expressing where the sites came from and why they were originally sampled. Your 30 chronology mean is no different I am afraid. Not all the data are from the Jacoby group. Many of these series are from Wilmking. Again, I do not know the details of the individual sites, but Wilmking is a dendroecologist and has been study the divergence’ issue in recent years with respect to positive and negative responder trees.

    Well, other groups beside Jacoby should be able to obtain valid results. My understanding is that Wilmking was trying to obtain temperature responses and his point about positive and negative responders came from following the results.

    Also, Steve, your time-series is a real mess. Surely you can see that the variance of this time-series needs
    to be stabilised to account for the decrease in N back in time. But that is for another thread (http://www.climateaudit.org/?p=1279)

    Well, “variance stabilization” is a method that has been used in recent dendro work, but there’s always a value in looking at data prior to adjustment. I am unaware of any third-party statistical reference to this methodology and until I see such a reference in a time series text and see a discussion of the properties of the method, I don’t accept the Briffa-Osborn adjustment as an essential statistical technique. I’m not saying that it’s not a good idea, I’m just saying that I don’t understand all the properties of this transformation right now and I’d be surprised if any of the dendro people do either. Maybe UC can shed a little light on it for me.

    Regardless, I like to see unadjusted data and personally, in this sort of sitation, I’d rather look at the unadjusted series in a “Grey70″ or “grey80″ tone with a red smooth overlaying it.

    Steve, if you are really interested in exploring the utility of these data as temperature (or other) proxies, then I would recommend you do a site by site analysis and see which of these chronologies express a valid climate signal. At the same time, for those sites which are apparent temperature proxies, you could then also address the divergence’ issue at the site scale. This would truly be an interesting piece of work and may actually be a positive contribution to science.

    Rob, I’d be happy to disaggregate this a little and would welcome some suggestions. Can you recommend an ex ante objective criterion expressed in terms of (say) latitude, longitude, altitude, facing direction by which PCGL sites should be included in a proxy average. But I’m not sold on ex post site-by-site picking.

  75. Edouard
    Posted Mar 22, 2007 at 7:46 AM | Permalink

    Hello,

    @Rob Wilson

    I’m from Europe and it’s very sad, that you don’t want to answer these questions anymore. Very many people read this blog, an they really want to know what is going on. If anyone would like to convince as many people as possible, that the climate-reconstructions are not a joke, he ought to do it on this blog. EVERYBODY who is interested in this science, wants to read answers from a real expert, and everybody has the same questions, that Steve asks. Please believe us, that we are just interested in climate science, and that we cant believe, that Co2 should have more influence on climate than for example the clouds in the European summer 2006 in August. There was a difference in temperature compared to July of 10 degrees (or something similar). We are no cranks or crazy sceptics. We just have some questions!

    Thank you very much for your answers

    Edouard

  76. Paul Penrose
    Posted Mar 22, 2007 at 8:50 AM | Permalink

    Dr. Wilson,
    Thank you for spending some of your valuable time coming here and answering our questions. Hopefully in the future you will be able to stop by again. Please understand that most of us are seriously trying to understand the issues raised in this blog, but find the relevent papers on the subject confusing and sometimes quite vague. Being able to ask the experts in the field some basic (and sometimes very specific) questions can be a great benefit, especially when they are clear and forthcoming as you have been (which is in stark contrast to some of your colleagues – Martin Jukes comes to mind).

  77. Dave Dardinger
    Posted Mar 22, 2007 at 9:43 AM | Permalink

    Rob,

    I recognize why you want to withdraw from responding. Each time you do respond there are a spate of messages (of which this would count as one I suppose), mostly negative, and each of which would seem to demand a response. This is the danger of message boards or blogs.

    I’ve been posting on boards or blogs for 15 years or so and one conclusion I’ve reached is that there’s a positive feedback to responding. The more you respond,the more others reply too and you end up with either exhaustion or a food fight. But that’s not how it has to be. The trick is to realize that you’re in charge. You don’t have to post any oftener than you want and you don’t have to respond to everything, and most importantly you don’t have to feel guilty for doing so.

    I personally think this is the cleanest board I’ve ever seen in terms of people being polite and desiring to know the truth. But you do have to realize that people of all sorts read and post here. So when you see someone with less knowledge / experience / ability saying something stupid according to your lights, it’s easy to transfer that to the whole group which gather here. Yet people learn, knowledge grows and sometimes even humility rears its beautiful head.

    I’d say that an “expert” who posts here should quickly develop a plan for what his/her purpose for being here is. Possibilities are:

    1. Check things out and then leave.
    2. Counter, occasionally, especially bad science / logic.
    3. Become a resident expert in some narrow field of interest and stick to ones knitting.
    4. Try to educate those who definitely need it and occasionally engage in a cutting-edge discussion with peers.
    5. Try to slay the dragon (Steve M). Personally I’d advice against this as the blowback has tended to scorch the people who have tried. (see #74 in response to your #70 as a good example.)
    6. Just have some fun and rile up the regulars. (The Dano technique or more generally the Troll Technique.)

    In summary, I’d suggest you NOT respond to everything thrown at you, nor do so to those things you do respond to at all quickly. But also don’t say “this is my last post here”, or even adding “… on this thread”, as it makes you look silly if you break your pledge or will raise your blood pressure if you stick to it and can’t post a zinger which comes to mind. As I said, you’re in charge of your own time; act like it!

    Fellow regulars: Consider this when you deluge a visiting expert with “everything I’ve always wanted to know about AGW but was afraid to ask.”

  78. John Hekman
    Posted Mar 22, 2007 at 10:41 AM | Permalink

    Rob Wilson in #70:

    However, if a site is close to a met. station and still does not correlate with climate, then it should not be used for proxy climate reconstructions.

    As an economist, I am stunned by this statement as to the use of data. If I am studying the demand for shoes, I would be regressing the quantity of shoes purchased on the price per pair. So I would be looking for the “response” of shoe purchases to price. I would normally have data from various consumers as to the number of pairs of shoes that they had bought at different prices. Now, following Rob Wilson, should I look at the behavior of individual shoe buyers and say, well this one here doesn’t respond to price in his purchases. There is no good “signal” there. So I’ll just eliminate him from the dataset.

    No. If economists had licenses, I would lose mine for that. The whole point of the exercise is to see WHETHER OR NOT shoe purchases respond to price, not to find the ones that do and measure their response. If I cherry-pick the data and then get my econometric estimate of the response of shoe-buying to price, and then use that “model” to predict shoe-buying, I will be absolutely wrong.

  79. MarkW
    Posted Mar 22, 2007 at 11:15 AM | Permalink

    #78,

    The purpose of dendroclimatology is to use tree rings to figure out what the climate was like back before there was anyone taking
    readings.

    What Dr. Wilson is saying is that those sites which approximate the temperature record in the present, can be assumed to approximate
    the temperature in the past. Those that show no correlation to the temperature record in the present, are not likely to correlate
    with temperature in the past.

    We are not trying to determine if tree rings correlate with temperature. We already know that under the right circumstances, they do.

    The purpose of these studies is to use tree ring data to determine past climate, more specifically, past temperature. It makes sense
    that we would want to limit our data to that subset of trees that are growth limited by temperature, not those that are growth limited
    by say, moisture.

    If you are doing a study on the price sensitivity of shoes, it makes sense to exclude stores that sell only pants. This is all the good
    doctor is talking about.

  80. cbone
    Posted Mar 22, 2007 at 11:23 AM | Permalink

    Dr. Wilson,

    I too would like to show my appreciation for your participation here. I primarily lurk around here, but I find the discussions to be quite informative. I appreciate you taking time to answer questions about your field of expertise here.

    CB

  81. Dave Dardinger
    Posted Mar 22, 2007 at 11:45 AM | Permalink

    re: #79 MarkW,

    You’re seriously wrong here.

    The purpose of these studies is to use tree ring data to determine past climate, more specifically, past temperature.

    The fact that some trees correlate with current temperature in certain situations doesn’t give us the temperature by inversion. In the first place the correlation is never going to be high enough give a useful thermometer for a given tree. Therefore it’s necessary to use statistical analysis to get a useable temperature. But if you cherry-pick your trees, sites, etc. then you’ve lost the data which would allow a valid statistical model to be created.

    In the second place… Well, I don’t need a second place. The first one is plenty.

  82. MarkW
    Posted Mar 22, 2007 at 11:57 AM | Permalink

    I said that the purpose of the study was to get temperature. I never said that the study was any good.

  83. Ken Fritsch
    Posted Mar 22, 2007 at 1:57 PM | Permalink

    3. The Buentgen Alpine (RW and MXD) records are very important. They track the local temperature record well and the reconstructed cold periods coincide well with periods of solar minima and also major volcanic events – not unlike the Luckman and Wilson (2005) Icefields study. For the next generation of NH reconstructions, I believe we need many more records of similar quality.

    Per Rob Wilson’s final post, I feel much disappointed that he was not able to offer many details or specifics regarding queries and concerns expressed at this blog concerning tree ring temperature proxies. He did however offer a couple of papers by Buentgen, that I have downloaded and plan on reading in more detail. In the meantime, would not a more productive approach to this exchange with RW, albeit in absentia, include a review and discussion of these two papers here and here.

    http://www.wsl.ch/staff/jan.esper/publications/Buentgen_2005_CD.pdf

    http://www.wsl.ch/staff/jan.esper/publications/Buentgen_2006_JoC.pdf

  84. Willis Eschenbach
    Posted Mar 22, 2007 at 2:29 PM | Permalink

    Re #79, MarkW, you say:

    What Dr. Wilson is saying is that those sites which approximate the temperature record in the present, can be assumed to approximate the temperature in the past. Those that show no correlation to the temperature record in the present, are not likely to correlate with temperature in the past.

    While I do think that this is the assumption upon which dendroclimatology is based, I do not find it at all convincing. If I take one hundred random 1000 year red-noise series and compare them to the 50 – 150 year local temperature record (depending on location), a good number of them will “approximate the temperature record in the present”. But we know, since they are random, that there is absolutely no reason to assume that they will approximate the temperature in the past.

    The need, as Steve M., myself, and others keep saying, is for ex ante criteria for picking sites (north facing, elevational or latitudinal treeline, etc.), and then using each and every record from each and every site that meets those criteria.

    Otherwise, we may well be just picking random red noise that happens to correlate with the recent past.

    This is why I asked Rob if there was any standard text in tree ring studies that covered site selection procedures.

    w.

    PS – It strikes me writing this that we should be able to distinguish red noise from signal by comparing the average inter-series correlation during the calibration period to the same statistic during the rest of the earlier record … but I haven’t a clue what would be a valid test statistic for the comparison, particularly since the number of data points decreases with age … UC, Jean S … HELP!

  85. Posted Mar 22, 2007 at 2:38 PM | Permalink

    #83

    Hmmm, both articles cite that same OsbornBriffaJones1997 variance adjustment paper.

    EU Sixth Framework Programme budget, Sustainable development, global change and ecosystems 2 329 EUR million. Can’t be that hard to find someone to give a detailed description variance adjustment. Sorry, this is populistic post ;)

  86. Steve McIntyre
    Posted Mar 22, 2007 at 2:48 PM | Permalink

    #84. I agree with your surmise that greater correlation in a calibration period than in the series as a whole is evidence of biased selection. Rob complains about this characteristic of series being studied in Wilson et al 2007 as follows:

    Although these records generally agree well with each other during the twentieth century, there is relatively weak between-series coherence before this time.

    He hypothesizes that the lack of early coherence may be evidence of a real diversity:

    However, the early lack of agreement among the PDO reconstructions may also reflect a real pattern of less coherency in the Pacific atmosphere’€”ocean system prior to the twentieth century (Gedalof et al. 2002; D’Arrigo et al. 2005).

    In my opinion (as yours), it’s more likely that the lack of coherence results from the proxies not being very good proxies. It’s exactly the same situation with Crowley’s use of proxies in the MWP (or Osborn and Briffa) – the proxies have a bit of a common pattern in the 20th century because of how they are selected rather than because they are measuring a common signal.

  87. MarkW
    Posted Mar 22, 2007 at 3:10 PM | Permalink

    Maybe I should have been more clear. #78 complained that the very notion of selecting sites for their sensitivity to temperature was
    flawed. I pointed out that this technique was valid. I never meant to imply that trees were, or were not good proxies for temperature.

    Assuming trees could be found that had good temperature to growth characteristics. Assuming that we could demonstrate that this
    characteristic extended to the past. Then the technique in question would yeild valid data.

  88. John Baltutis
    Posted Mar 22, 2007 at 4:39 PM | Permalink

    Re: #79

    We are not trying to determine if tree rings correlate with temperature. We already know that under the right circumstances, they do.

    The purpose of these studies is to use tree ring data to determine past climate, more specifically, past temperature.

    Really? Exactly what are the right circumstances? I.e., the ones where the confounders (noise), such as fertilization, location, soil erosion, and precipitation, have been uniquely identified and removed from the signal, leaving only temperature? IMHO, there is no correlation in the tree-ring data that can be directly linked to temperature. It’s only speculation, hypothesization, or a gut feeling that it must exist; otherwise, one would conclude that dendroclimatology serves no useful purpose .

  89. Kevin
    Posted Mar 22, 2007 at 5:34 PM | Permalink

    I am curious how to distinguish a valid climate signal from an invalid one. What exactly is needed to differentiate coincidence from correlation regarding tree rings and temperature variation, for instance, and to isolate all other possible factors that could influence tree growth in a given span of time?

  90. IL
    Posted Mar 22, 2007 at 6:08 PM | Permalink

    Willis, #84. I completely agree with you. What is more, if the case is that a particular tree responds principally to temperature today because it is on the tree line or is otherwise ecologically stressed so that temperature is the dominant factor in its growth – this assumption surely cannot be made about that tree through the past! Tree lines vary in latitude and altitude through time and if it was in the right position to respond to temperature in the past, its either well below the treeline now and not responding so well, or above the current tree line – and dead. It would be incredible coincidence for a particular tree to have remained in the ecologically stressed condition where it responds dominantly to temperature for hundreds of years. It could only do that if global climate was completely stable (until humans messed it up) (oh yes, that was the hockey stick wasn’t it?).

  91. MarkW
    Posted Mar 22, 2007 at 6:40 PM | Permalink

    #88,

    Basically the “right conditions” would be a situation where everything else that a tree needs to maximize growth is available in best quantities. Enough nutrient. Enough, but not too much water. Etc.

    As IL mentions, just because such conditions exist during the last few years, is not evidence that the same situation existed hundreds, much less thousands of years ago.

  92. John Baltutis
    Posted Mar 22, 2007 at 7:42 PM | Permalink

    Re: #91

    Basically the “right conditions” would be a situation where everything else that a tree needs to maximize growth is available in best quantities. Enough nutrient. Enough, but not too much water. Etc.

    Know of any, except in a controlled greenhouse?

    As IL mentions, just because such conditions exist during the last few years, is not evidence that the same situation existed hundreds, much less thousands of years ago.

    Agreed. However, IL said that was an assumption:

    …if the case is that a particular tree responds principally to temperature today because it is on the tree line or is otherwise ecologically stressed so that temperature is the dominant factor in its growth – this assumption surely cannot be made about that tree through the past!

    I don’t know of any case where that assumption’s outside of a controlled environment.

  93. bender
    Posted Mar 22, 2007 at 9:20 PM | Permalink

    Re #68
    The high 1960s growth is due to a rebound following a spruce budworm outbreak. Budworms were virtually absent from eastern North America throughout the 1960s (Hardy et al. 1983). This 1960s growth rebound has been well documented by many researchers, but especially Jardon et al. (2003). So much for inferring climatology from disturbance-dominated forest types.

  94. Willis Eschenbach
    Posted Mar 22, 2007 at 9:42 PM | Permalink

    MarkW, you say:

    Basically the “right conditions” would be a situation where everything else that a tree needs to maximize growth is available in best quantities. Enough nutrient. Enough, but not too much water. Etc.

    Unfortunately, even that is not enough to make a tree act like a thermometer, because when it gets too hot, even though everything else is perfect, growth still slows. There is no way that I know of to distinguish narrow rings=too cold from narrow rings=too hot.

    In addition, since at high altitude treelines moisture is often scarce, the “too hot for the available water” scenario is all too common.

    w.

  95. Andrey Levin
    Posted Mar 23, 2007 at 1:21 AM | Permalink

    Tang at al, Global Change Biology, 1999:

    Abstract
    Dendrochronological work at Sheep Mountain in the White Mountains, CA has demonstrated that bristlecone pine trees in two forms, full-bark and strip-bark, have experienced different cambial growth rates over the past century or longer. The strip-bark trees showed a greater growth increase than the full-bark ones. A calculation of the plant water-use efficiency (W) in response to anthropogenic CO2 released into the atmosphere shows that W of trees in both forms has increased for the past 200 years. However, there is no significant difference between the two tree forms in the rate of increase in W. This implies at least two possibilities with respect to the CO2 fertilization effect. First, the biomass in both tree forms might have increased, but carbon distribution among different parts of a tree was different. Second, the biomass may increase without causing any corresponding change in the plant water-use efficiency.

    http://www.blackwell-synergy.com/doi/abs/10.1046/j.1365-2486.1998.00204.x?journalCode=gcb

    How carbon fertilization effects were accounted for during calibration time interval? What historical CO2 time series were used for last 1000 years? Siple curve from ice cores or current stomata proxies?

  96. MarkW
    Posted Mar 23, 2007 at 4:47 AM | Permalink

    #95,

    “How carbon fertilization effects were accounted for during calibration time interval? ”

    From the studies that I have seen, it hasn’t been accounted for at all.

  97. bernie
    Posted Mar 23, 2007 at 5:49 AM | Permalink

    The last bit of this thread seems to me to be a potent critique of dendroclimatology, notwithstanding Rob Wilson’s patient responses. Like many others, I simply don’t see how the models can work statistically because of the number of unknowns and the interactions among factors. (In part this comes down to how do define a PC as being the temperature signal as opposed to anything else. From a far less rigorous field, opinion surveys, where I spend much of my time, the first two factors out of poorly structured or defined dataset tend to be a “good” factor and a “bad” factor – but these are seldom useful explanatory variables and typically mean you have to restructure your sample to find any meaningful underlying structure. This type of approach means you need plenty of samples for confirmation purposes. )
    The potential flaws are too obvious for the issues not to have been raised before. Does anyone have references to earlier debates on these issues?

  98. Posted Mar 23, 2007 at 9:07 AM | Permalink

    #84

    PS – It strikes me writing this that we should be able to distinguish red noise from signal by comparing the average inter-series correlation during the calibration period to the same statistic during the rest of the earlier record … but I haven’t a clue what would be a valid test statistic for the comparison, particularly since the number of data points decreases with age

    That is a good question. Monte Carlo simulations are one option.

    But let’s try to write down what should happen: Assume N independent red-noise processes, and those processes have identical autocorrelation structure. Then it is easy to show that the average process (*) has similar autocorrelation structure, but the process variance is reduced by a factor of 1/N. If the autocorrelation structure varies, then computations are not that straightforward. But the variance of the average process is 1/N of the mean variance of the original processeses. And if those original processes are standardized, the 1/N reduction works, whatever the autocorrelation structure is. Thus, if for given data the variance gets down by 1/N in the verification period, but not that much in the calibration period, we should be worried about overfitting.

    And this here’s the funny part, it means that we should be worried about overfitting if the result is hockey-stick shaped!!

    See also Dave’s comment

    http://www.climateaudit.org/?p=1283#comment-98150

    (*) A new stochastic process that we obtain by taking average over N processes.

  99. Mark T.
    Posted Mar 23, 2007 at 9:29 AM | Permalink

    #78 complained that the very notion of selecting sites for their sensitivity to temperature was
    flawed. I pointed out that this technique was valid.

    The problem with the analogy that you used is that you knew a-priori that the store did not sell shoes, relying on pants for income instead. We don’t have that a-priori knowledge w.r.t. trees and their ring widths, hence statistical analysis. Therefore, selecting one tree because it responds well to a specific climate variable, while excluding another, is really an end-run around what should really be done: exclude the hypothesis (or accept the null, which is no correlation).

    I never meant to imply that trees were, or were not good proxies for temperature.

    Yes, I got that part. Should we manage to contrive a controlled experiment, they probably could be.

    Mark

  100. Willis Eschenbach
    Posted Mar 23, 2007 at 3:25 PM | Permalink

    UC, thanks for your posting supporting my idea that we can determine the statistical relevance of a group of proxies by comparing their average cross-correlation over time. Being nothing if not experimentally minded, I took a look at Juckes MITRIE data, since all of the proxies extend for the full record. Here are the results:

    The calibration period runs from about 1860-1980, and the 100 year trailing cross-correlation data for that period is shown by the blue lines … I don’t think we need a lot of sophisticated math to determine whether the MITRIE reconstruction is statistically significant …

    The pre-calibration period average cross-correlation is 0.04, while that of the calibration period is more than three times that, 0.14.

    w.

  101. Posted Mar 24, 2007 at 2:11 AM | Permalink

    #100

    Look interesting. Let’s continue:

    Inspired by Dave’s comment , I made a hockey-stick plant, code here . It produces this kind of pictures:

    Millennium project cost is 15 million euros (2006-2009). They say

    Our multi-disciplinary team will use innovative and developing technologies to extract quantitative palaeoclimate information from documentary and natural archives, including trees, lakes, mires and ice cores

    If the questions raised here are not handled properly, I want my money back. That will be 0,03 euros.

  102. Steve McIntyre
    Posted Apr 14, 2007 at 3:48 PM | Permalink

    Update2: Discussion of Alberta sites by Meko here

  103. bender
    Posted Apr 14, 2007 at 5:22 PM | Permalink

    Not sure there’s much to say on these Alberta spruces. As I said before, Charles Stockton made the first collection in the early 1970s, and published a report in 1973 showing that these spruces were a proxy for water table height. Anyone claiming they’re a temperature proxy would have a hard time refuting Stockton’s interpretation.

3 Trackbacks

  1. [...] http://www.climateaudit.org/?p=1278 Up-to-Date White Spruce Ring Widths [...]

  2. By Wilson, Pisaric and Gaspé « Climate Audit on Jun 6, 2011 at 8:16 PM

    [...] Pisaric of Carleton University in Canada writes: As Rob points out in 3 and 28, Steve’s analysis is flawed! You can easily search out for [...]

  3. By Wilson, Pisaric and Gaspé « Climate Audit on Jun 6, 2011 at 8:16 PM

    [...] see post here with further thoughts on the location of this site. Next post in category] Rob Wilson wrote, using an uncharacteristic Gavin-esque sigh, as follows: Sigh — Location, location, [...]

Follow

Get every new post delivered to your Inbox.

Join 3,299 other followers

%d bloggers like this: