Spaghetti Graph Data

Some of you are looking for digital versions of the multiproxy results. For convenience, here is a digital collation of major reconstructions used in spaghetti graphs. The collation text is here, and URLs to source data can be extracted from line items. It’s not been made user-friendly.


9 Comments

  1. Posted Jan 9, 2006 at 1:46 PM | Permalink

    Excellent, thanks!

  2. Willis Eschenbach
    Posted Jan 9, 2006 at 1:53 PM | Permalink

    Steve, many thanks for the data, and for the R script. I learn more by reading your scripts than from my texts on R.

    A question on an unrelated subject. Constructal theory is much in the news, including the application of constructal theory to climate. Do you know a good source on the statistics of constructal geometries? Are there any R scripts involving constructal geometries?

    All the best,

    w.

  3. Steve McIntyre
    Posted Jan 9, 2006 at 4:08 PM | Permalink

    Willis, my R techniques are pretty homemade; I know enough now that I could benefit from a course in it from first principles. I noticed your posts on constructal theory at RC. I googled and didn’t see any articles that were very helpful and can’t look at the references until I go back down to the university. If you’ve got some pdf’s handy, could you email them to me? I haven’t seen anything at R, but I’ll keep it in mind. Cheers, Steve

  4. Posted Jan 9, 2006 at 6:31 PM | Permalink

    Perhaps I’m being naive or totally missing something, but when I look at Mob05, I can’t figure out how someone could look at the data with a critical eye and believe that it supports the GW theory for several reasons:

    1) Measurements over the last 45+ years are flat
    2) Measurements before the so-called “mini-ice age” are equal or higher to recent measurements
    3) The drop in temperatures leading to the mini ice age appears to be as dramatic as the depicted increase starting in the mid 19th century.

    Comments appreciated.

  5. Willis Eschenbach
    Posted Jan 9, 2006 at 6:37 PM | Permalink

    Steve, the only paper on good constructal theory I can find is “The constructal law of organization in nature: tree-shaped flows and body size, by Adrian Bejan (the man who invented constructal theory), available at

    http://jeb.biologists.org/cgi/content/full/208/9/1677

    It’s a fascinating paper, but not about climate. I have found references to a paper on constructal theory of climate, but cannot find the paper online. I’m about to order Bejan’s book from Amazon, but I’m in Fiji so it will take a while to get here.

    If climate is well described by constructal theory (which would not surprise me, as constructal theory covers anything that flows, which includes climate), then it would be fascinating to understand the statistics of the constructal universe. Regarding that question, I find nothing …

    w.

  6. Dano
    Posted Jan 9, 2006 at 8:34 PM | Permalink

    Googleicious:

    1.

    applications:

    2.
    C.
    Q.

    No R. Plenty more on ISI/Ingenta if’n you have a sub, w.

    D

  7. nospam Dano
    Posted Jan 9, 2006 at 8:39 PM | Permalink

    Googleicious:

    1.

    applications:

    2.
    C.
    Q.

    No R. Plenty more on ISI/Ingenta if’n you have a sub, w.

    Best,

    D

  8. Dano
    Posted Jan 10, 2006 at 11:16 AM | Permalink

    When John A is done looking for opportunites to make up phrases, perhaps he can delete one of the duplicate posts above. Thanks!

    D

  9. Posted Jan 24, 2006 at 1:53 PM | Permalink

    Hi, I have attemped a dummies guide to analysing scale-invariant
    series and put it on my site. The H values on the proxy data Steve supplied are very sobering.

    Scale Invariance of Natural Series (http://landscape.sdsc.edu/~davids/enm/?p=32)

    Below is an investigation of scaling or long term persistence (LTP) in time series including tree-ring proxies — the recognition, quantification and implications for analysis — drawn largely from Koutsoyiannis [2] (preprints available at http://www.itia.ntua.gr/e/byauthor/Koutsoyiannis/0/. In researching this topic, I found a lot of misconceptions about LTP phenomena, such as LTP implying a long term memory process, and a lack of recognition of the implications of LTP. As to implications, using the estimate s.e. = sigma/n^(1-H), and the degree of LTP of the CRU global temperature series of H = 0.95 the standard error of the mean is 4 times larger than the usual estimate for normal errors. Given that LTP is a fact of nature — attributed by Koutsoyiannis to the ME principle — this suggests as standard practise, (1) LTP behavior should be our first assumption, (2) H estimated for all data and (3) estimates such as the generalized s.e. above should be used unless classical statistics can be shown to apply.

%d bloggers like this: