“AGU Journals Should Ask Authors to Publish Results”

This is the title of a current op ed in EOS drawn to my attention by Leif Svalggard. The policies advocated in the op ed are obviously ones that I endorse.

AGU actually does have data policies that, on paper, would deal with many of the disputes that I’ve had with paleoclimate authors. From time to time, I’ve tried to get AGU editors to enforce even their present policy, but to date AGU editors have simply ignored such correspondence – not even acknowledging.

For example, I tried to get Colin O’Dowd, editor of JGR, to enforce AGU policy on data. My exchange with O’Dowd is mentioned by Climategate correspondents who felt confident that my initiative would be rebuffed. O’Dowd never even acknowledged any of multiple emails (though I’m a member of AGU.) I later wrote a member of the AGU board who acknowledged my email but did nothing either.

More recently, in connection with Gergis et al, I asked Eric Calais, editor of GRL, to require one of the Gergis coauthors to archive data published in GRL (and considered in Gergis et al) in accordance with existing AGU data policies.

I am writing in respect to data for Neukom et al 2010, Multi-centennial summer and winter precipitation variability in southern South America, published in GRL.

There has obviously been considerable adverse publicity about authors of paleoclimate temperature reconstructions failing to archived data and several committees have recommended that such practices end. This has occurred once again with Neukom et al 2010. Could you please ask the authors to archive the proxy data used in their reconstruction?

No answer. No acknowledgement.

The existing AGU policy is as follows:

1. Data sets cited in AGU publications must meet the same type of standards for public access and long-term availability as are applied to citations to the scientific literature. Thus data cited in AGU publications must be permanently archived in a data center or centers that meet the following conditions:

a) are open to scientists throughout the world.
b) are committed to archiving data sets indefinitely.
c) provide services at reasonable costs.

The World and National data centers meet these criteria. Other data centers, though chartered for specific lengths of time, may also be acceptable as an archive for this material if there is a commitment to migrating data to a permanent archive when the center ceases operation. Citing data sets available through these alternative centers is subject to approval by AGU.

2. Data sets that are available only from the author, through miscellaneous public network services, or academic, government or commercial institutions not chartered specifically for archiving data, may not be cited in AGU publications. This type of data set availability is judged to be equivalent to material in the gray literature. If such data sets are essential to the paper and authors should treat their mention just as they would a personal communication. These mentions will appear in the body of the paper but not in the reference list.

What tends to happen is that authors disregard rule (2) in a typical case where a dataset is described in a print publication but not archived, instead being passed hand to hand among pals. As I interpret AGU rule 2, the citation of the print article should not permitted if the data itself has been obtained through gray channels. Unfortunately, neither AGU editors or reviewers pay the slightest attention to the policy or take the slightest interest in breaches.

In addition to recommending that the existing rules be enforced, I would be inclined to toughen up rule 2 so that the obstacles to the use of gray versions were much more severe than at present.


  1. Gary
    Posted Jul 14, 2012 at 7:34 AM | Permalink

    Polite requests haven’t worked. Neither has public shaming. Seems the next options are buying up enough stock for a controling interest which in this case would be a petition signed by a sufficient number of AGU members and legal action by means the organizational constitution.

  2. Lucy Skywalker
    Posted Jul 14, 2012 at 7:40 AM | Permalink

    I would like to see another “rule”, that is, that reasonable communication about enforcement of policy, like yours, should be dealt with as a matter of course, and replied to appropriately.

    Silencing / ignoring the voice of reason that you and CA represent in this way, is a big factor in current corruption in Climate Science and reason for the general public’s growing skepticism and distrust of “climate scientists”.

  3. Craig Loehle
    Posted Jul 14, 2012 at 8:22 AM | Permalink

    It has been my experience with editors when a paper of mine has been rejected, that if I object that the reviewers completely misunderstood my paper or were rude and out of line (and I don’t do this as a matter of course), that I have never gotten a reply. Never. Shirking of responsibility by unpaid editors, who would have imagined it?

  4. j ferguson
    Posted Jul 14, 2012 at 8:38 AM | Permalink

    “… rude and out of line …” ?

    My daughter assures me that these are good symptoms that your paper was looked at by academics.

  5. Matt Skaggs
    Posted Jul 14, 2012 at 8:41 AM | Permalink

    This may just be an ivory tower thing, but one wonders if the real policy – publish a robust requirement but ignore it – actually keeps the porridge just right. After all, you would not want a prominent scientist to publish in a different journal just because you enforce your archiving policy and the other journal does not. Isn’t it the goal of a journal to have the most citations? I suppose you achieve that by publishing the most prominent papers, not the best papers.

  6. John Silver
    Posted Jul 14, 2012 at 9:43 AM | Permalink

    “O’Dowd never even acknowledged any of multiple emails (though I’m a member of AGU.)”

    What about paper letters or even registered paper letters?
    I have found that registered letters always impresses the receiver. Yes, here is cost, but there is also cost-benefit analysis.
    (I’m not being sarcastic)

  7. AntonyIndia
    Posted Jul 14, 2012 at 9:44 AM | Permalink

    Such scientific publications are no more than modern emperors with “new clothes” (= non). Within their specialized field people got used to these data-less fashions and don’t notice it any more but the wider (Internet!) public is not equally bedazzled. Authority (legacy) now depends on verifiable proof (data) not on reputations, dominating positions, connections or niche hide outs. The open 21st century will not tolerate this 20th century’s alchemy.

    • Duster
      Posted Jul 14, 2012 at 7:02 PM | Permalink

      It is a symptom of a growing legitimization of faction formation that is just as likely to bring down this civilization as it did the Byzantine Empire. Factions (Teams) discard the “civil” in civilization. In Byzantium the “teams” were quite literally the adherents of of chariot racing teams. The Smithsonian site has an interesting article on this “Blue versus Green”. A new dark age would certainly take care of any CO2 issues.

  8. theduke
    Posted Jul 14, 2012 at 9:58 AM | Permalink

    From the op-ed:

    The whole point of papers is to allow scientists to benefit from each others’ research; we could do this much more easily if AGU journals strongly encouraged authors to provide, as text files with some annotation, the information on which
    they based their conclusions.

  9. Posted Jul 14, 2012 at 10:18 AM | Permalink

    Several years ago a journal (in psychology) was taking a very long time to review a paper. I wrote to the editor. No reply. I wrote to the board overseeing the editor. They took action that addressed my complaint.

    • Tony Mach
      Posted Jul 15, 2012 at 2:23 PM | Permalink

      Seth Roberts is here? Cool! If you are interested what the spawns of Gallo are up to nowadays, you should look at the joint work of Ruscetti and Mikovits – smells like HTLV-III all over.

      Welcome and happy hunting here in Climate Science land!

  10. MarkB
    Posted Jul 14, 2012 at 12:50 PM | Permalink

    The writer actually says that the availability of results should be standard, although not required. How does that solve anything? The Usual Suspects would simply choose to not follow the standard, and there would be no recourse.

    Let’s go back to first principles. Every freshman in a Science 101 lab class learns that a Methods section must include everything needed to replicate the work described. In computational and statistical research, one cannot replicate work with a different data set or different computation or analytic tools. It is literally impossible.

    Therefore, no such work passes muster for college freshmen, much less professional scientists. IN order to get on with it and keep the publication gravy train running, everyone in the fields go along with this practice, but it does not hold up under scrutiny.

    When data sets were held in lab notebooks, this could be excused on pragmatic grounds. This is obviously no longer true. In order to replicate your work, I need your data and your code. Both should be archived and available without asking. Just shining the light of day on such work would do wonders for the relevant fields, if not for the careers of the inmates.

  11. dearieme
    Posted Jul 14, 2012 at 3:52 PM | Permalink

    The struggle against data denialism must continue.

  12. Coldish
    Posted Jul 16, 2012 at 8:43 AM | Permalink

    “My exchange with O’Dowd is mentioned by Climategate correspondents”. Steve, I’d like to read those emails. Is there a reference number? Thanks for all the sterling work. Coldish

  13. djbiggs
    Posted Jul 16, 2012 at 3:36 PM | Permalink

    I believe the message of open archived data may be getting through, or at least 50m is getting thrown in the general direction of the problem.

    Steve: different issue entirely. That’s about the price of academic articles. Not about data archives.

  14. djbiggs
    Posted Jul 17, 2012 at 2:59 AM | Permalink

    Perhaps you missed,

    “Such repositories are important because they contain all the experimental data that was collected during the experiment as well as its results.

    It is the raw data that is as important if not more so to other researchers as the results themselves in order to make full use of the information to help their own research. It is unclear at this stage whether publishers will allow access to the raw data or impose a charge for access to this information.

    “The Government, despite having made a bad choice, still has an opportunity through the detailed implementation of the new structure to ensure that researchers and taxpayers do not lose out completely,” Mr Friend told BBC News.”

    http://www.friendofopenaccess.org.uk/index.php/current-issues is a bit more substantive.

  15. fastfreddy101
    Posted Jul 17, 2012 at 3:49 PM | Permalink

    “European Commission backs calls for open access to scientific research”

    The European Commission, which controls one of the world’s largest science budgets, has backed calls for free access to publicly funded research in a move that could force a major change in the business model for publishers such as Reed Elsevier.

    “Taxpayers should not have to pay twice for scientific research and they need seamless access to raw data,” said Neelie Kroes, European Commission vice-president for digital agenda.

    The EC saidon Tuesday that open access will be a “general principle” applied to grants awarded through the €80bn Horizon 2020 programme for research and innovation.

    From 2014 all articles produced with funding from Horizon 2020 will have to be accessible and the goal is for 60% of European publicly funded research to be available by 2016.

    source: The Guardian

    Nice, but will the data be free for open access?

%d bloggers like this: