Jacoby Archiving

Jacoby is a key player on the Hockey Team. Jacoby and d’Arrigo have received nearly $8 million from the National Science Foundation (collated from public NSF records)- see here ; in addition, they have received an undisclosed amount of funding from other public agencies (collated from publication references).

As far as I’m concerned, US federal policy on data archiving states quite clearly that such data should be archived; it seems self-evident that NSF, as the granting agency, has an obligation to ensure that it’s archived. Jacoby and d’Arrigo have done work in North America (especially Alaska), Mongolia, New Zealand, Russia, Indonesia, Tasmania and perhaps elsewhere. They have travelled widely and undoubtedly worked hard. However, exploration geologists travel widely to remote places and work hard too and aren’t prima donnas.

The NSF stated that Jacoby had been “have been responsive to the NSF policy on access to data”. Let’s check this. See the summary in Table 1.

In North America, Jacoby archived RW measurements for 10 of 36 sites used in his northern treeline composite — all measurements taken on or before 1984 (!). What happened since? I’ve seen references to 34 MXD chronologies; none have been archived). Updates of 4 RW series (of the 10 sites) have been mentioned, but no update measurements have been archived. This includes two different updates at Twisted Tree Heartrot Hill; the most recent of which caused the authors to wonder whether ring widths had a quadratic (upside-down U) response to temperature since ring widths were not going up with higher temperature. I learned privately that an update of the Gaspé site was carried out, which did not yield a hockey stick shaped series. As discussed elsewhere, Jacoby refused to provide this data, refused to provide information on the site location and even denied that the update had taken place. I’ve seen references to 40 new Jacoby RW chronologies in North America, of which only 3 have been archived. Of these 3, two were actually the responsibility of another PI (Wiles) in connection with a grant not listed here, with Jacoby a secondary author. Thus, only one set of North American measurements (Okak) after 1984 has been directly archived by Jacoby. In our review of MBH98 material, we noticed that some of the site chronologies (which are a form of index constructed from the measurements) appeared to be incorrect. We notified d’Arrigo of this early last year, who promised to re-issue corrected site chronologies, but does not appear to have done so.

Elsewhere, Jacoby has cited 5 Russian sites; so far he has only 1 incomplete RW archive and no MXD archive. There are mentions of 5 Indonesian chronologies, but only 1 measurement archive. (An Indonesian chronology is used in MBH98 which lacks any primary archiving.)

There are mentions of 9 New Zealand chronologies, but only 1 measurement archive. There are mentions of 13 Tasmania chronologies, of which 3 archived by Cook.

For Mongolia, Jacoby has archived 9 measurement files, but the Sol Dav file used in Mann and Jones [2003] does not go back as far as the record in Mann and Jones [2003].

I may have missed a nit here and there, but it is evident that Jacoby and d’Arrigo have not adequately archived their material. One can hardly avoid wondering what their criteria were in each case for which measurements were archived and which were not. Has the selective archiving itself distorted the record? After all, Jacoby has said that he is "mission-oriented".


  1. John A
    Posted Jul 2, 2005 at 10:01 AM | Permalink

    Has the selective archiving itself distorted the record? After all, Jacoby has said that he is “mission-oriented”.

    Let me guess. The goal of the “mission” has not been archived either.

  2. Paul
    Posted Jul 2, 2005 at 10:18 AM | Permalink

    From Jacoby’s response to you in the mission oriented link, “A lesser amount of good data is better without a copious amount of poor data stirred in.” This is true, but to paraphrase Bill Clinton, it depends on your definition of good data. I could understand it if his description of bad data included things like tree rot or whatever. But reading his full response, he is clearly filtering the data by the signal he sees, not the physical quality of the sample. ” Is there a common low-frequency signal among the trees? At a good temperature- sensitive site with good trees, there is.” This clearly biases his results.

    Of course, this still begs the whole question of how you can extract a temperature signal from tree ring widths. The width of a ring depends on at least five variables, temperature, water, sunlight, CO2, and soil fertility but the width only provides a single equation. Basic math requires four more equations to solve for all five unknowns. What are the other independent measurements that will provide them?

    Oh well, Semper Fi

  3. Posted Jul 2, 2005 at 11:46 AM | Permalink

    Of Jacoby et al. maybe we should say they are “omission oriented”.

  4. Jim Carson
    Posted Jul 4, 2005 at 12:15 PM | Permalink

    The drug trial analogy is incomplete. Jacoby clearly states that the 26 patients should not be heard from again.

    Dr. McIntyre will have fully arrived when he receives correspondence that begins, “Steve, you ignorant slut…”

  5. Jeff Alberts
    Posted Mar 6, 2013 at 10:09 AM | Permalink

    “Wizard of Oz” science at it’s best (don’t look behind the curtain, just be in awe of the results).

One Trackback

  1. [...] In 2005, I tried to get the NSF to intervene and require Jacoby to archive his data completely. They refused. [...]


Get every new post delivered to your Inbox.

Join 3,203 other followers

%d bloggers like this: