A couple of days ago, I posted up a copy of a letter that I sent to Science on archiving or non-archiving in connection with Osborn and Briffa 2006, Esper et al [2002] and Thompson et al 1989 (Dunde); 1997 (Guliya).
I received a reply from Science today, which they stated was not for “public posting”. I don’t understand the idea of official organizations purporting to deal with formal complaints on an off-the-record basis.
Since relevant information was provided in the email, I presume that it’s OK to transmit the sense of the email.
Science stated that Osborn and Briffa had complied with their policy by depositing their data at WDCP ( here by the date of publication) and that they had omitted this reference “by mistake”, which they were now correcting. I confirmed with WDCP that some data had been deposited with them prior to publication and that this data had been placed online prior to Feb 10, 2005. Science also sent me digital chronologies for 13 of 14 sites used in Esper et al 2002 which are now archived at climateaudit here – I don’t know why they wouldn’t archive this at their website).
As always, the devil in the details. What did we actually get? Well, the “data” archived by Briffa are 40-year smoothed versions of the 14 series used in their compilations, rather than the series themselves. Obviously if you want to check annual correlations to temperature (which was discussed in my letter), 40 year smoothed versions are no good. As summarized below, they didn’t provide any data citations for measurement data or any temperature versions.
The Esper package was 13 of 14 site chronologies – Mongolia inexplicably omitted. They provide Esper’s version of Polar Urals, which I’m going to post up in a few minutes. Most of these correspond to Briffa versions, but there are a couple of odd exceptions, which I’ll comment on later. I didn’t actually request these site chronologies, as my impression of Esper et al [2002] was that they did very large-scale RCS at a supra-site level (but I could be wrong as Esper et al is Mannian in the obscurity of their methodological description.) Science did not provide an operational definition of how trees were allocated between being “linear” and “nonlinear” – which is not, shall we say, a Linnaean typology.
As for Thompson’s Dunde and Guliya data, they point out that they previously provided Kilimanjaro data to me (which is interesting and something that I wanted, but doesn’t shed a great deal of light on the conundrum of inconsistent Dunde versions) and that the publications were prior to the establishment of a policy. This is interesting with respect to NSF – who said that they relied on journal data archiving policies as a standard. If Science had no policy, then what was NSF relying on in 1997 for Guliya – as NSF had obligations of its own.
Anyway, here’s a summary of the status of my requests. Some useful information, but frustrating. Do you suppose that Briffa archived smoothed versions just to be irritating? What’s their problem with providing exact data citations or providing measurement data? I will write back to them formally.
| Item | Author | Request | Status |
| 1 | Briffa | Digital versions of all 14 series as used in their final compilations; | 40-year smoothed versions of the 14 series archived at WDCP here |
| 2 | Briffa | For each of the tree ring sites analysed (both the 11 retained and Esper sites not used, including Gotland, Jaemtland, Mackenzie Mts and Zhaschiviersk), an exact data citation to a public archive (e.g. WDCP)for the data set used; or, in the alternative, an archive of the data set at the Science website. In cases, where the publicly archive dataset for a site is related to but different from the version used by Osborn and Briffa, please archive the data set as used. | Not provided |
| 3 | Briffa | Digital versions of the specific gridcell temperature series used in each of the reported temperature correlations together with version date. | Not provided |
| Esper | Site chronologies for 13 of 14 sites (Mongolia unaccountably missing). See http://www.climateaudit.org/data/Esper_RCS.crns | ||
| 4 | Esper | Exact data citations to a public archive for all datasets used, or, if such do not exist, an archive of the data set at the Science website. | Not provided |
| 5 | Esper | A clear and operational definition distinguishing “linear”‘? and “nonlinear”‘? trees, preferably with source code showing any differences in methodology. | Not provided |
| 6 | Thompson | A complete archive of both Dunde and Guliya ice cores, including both isotope and chemical data | Not provided. Science pointed out that they previously sent some data from Kilimanjaro; that these cores were published prior to the existence of a journal policy and they were “working” with Thompson on the matter. (My first request to them was in February 2005.) |
Climate Audit 
11 Comments
Data requested from six articles, one set of raw data provided.
Until the climatology “establishment,” if they have one, adresses these concerns I cannot view “climatology” as a science. It is an affront to science and is properly classified as propaganda.
The climatology establishment fears that if they practice real science instead of political science a large percentage of the annual $5 billion pumped into their coffers will be cut. It’s all about money and power. The science is clearly secondary.
I have to say that this stuff is delightful, and perfect as supplementary information to a submission to NAS, or to NSF if they ever choose to see if their data availability policies are upheld.
I think you have got your approach exactly right. A further formal letter pointing out what they have not provided, and making clear (1) that this is required by Science’s own policy (2) that this is essential information if the paper is to be replicated and shown to be reproducible (3) the necessary information for (2) is not provided ! It is of course salient to note that the cost of making such information available in an archive is trivial, and causes minimal disruption if archived at the time of paper submission.
yours
per
I have no idea why Science doesn’t read the riot act to these guys. If I were a deputy editor of Science, I’d just tell these guys – here’s our policy, it’s black-and-white, comply with it to the letter. I’m tired of writing stupid letters protecting you; I’m not going to spend any more time protecting you; I don’t want to waste time arguing about it. Either archive your data or we’ll retract the article. In the present environment, it’s easy for them to clean up this sort of situation. Hanson knows that I’m going to write back to him with a detailed reply making him look bad. Then he’s going to have to spend more time and still look bad. The crazy thing is that he thinks that I’m the problem, when it’s Briffa and Osborn and Esper that are the problem.
Six brix plus six brix mayx a doxing.
======================================
Science editors likely agree with the positions “these guys” take, and therefore are attempting to give them as much fudge room as possible. This tactic should backfire eventually, with public egg on all. Hopefully this is the type of conflict that will eventually result in a shakeup of the system.
Mark
One solution – strike Science from the approved list for government magazine subscriptions. Fiction is a personal expense.
Sir:
There may be some form of legal option available. Research publicly funded should be under some form of public records requirement in that a legal request to produce (such as Freedom of Information Act in USA) must be followed. Perhaps a legal request might get those editor’s attention.
YMMV
The problem of archiving is a simple one to solve. The scientist is not eligible to receive another grant until he has completed the archiving required from the last one.
Science editors likely agree with the positions “these guys” take, and therefore are attempting to give them as much fudge room as possible.