Ring-a-Round 2: Queens University Belfast v Doug Keenan

WUWT readers may recall that Queens University Belfast is being asked to provide tree ring data and so far has been refusing all but a small portion. Here is an update on that story first carried in WUWT.

Guest Post by John A

Following on from the last post on Doug Keenan’s struggle to get tree ring data from Queen’s University, Belfast, we have Mike Baillie from QUB to explain to Benny Peiser of CCNet:

Dear Benny,

although I am retired from basic dendrochronological work, I would like to correct a small part of the diatribe against Queen’s University, Belfast, that you carried on CCNet on the 15 August, namely the allegation that we are deliberately withholding data of climatic significance.

Your source, Mr Keenan, gives the impression that data from only one Irish oak site is available, namely Garryland Wood, Co Galway.  This is a site he used in an attempted correlation with temperature records.  He points out the problem of dealing with data from an individual site, and states that “Those problems could be at least partially addressed by considering the individual trees at the site, rather than the average for the site, and also by considering trees at other sites in the British Isles.  Doing so would presumably lead to additional increases in the correlation (that he found between Garryland tree rings and temperature records)”.

Now any fair minded reader would take it from that quotation that “the individual tree” data from Garryland is not available.  Also that same reader would take it that data from other “trees at other sites in the British Isles” are not available.  Presumably, if the data were available, Mr Keenan would have extended his analysis in the search for an even better correlation between tree-growth in the British Isles and temperature, either local or Hemispheric.

The point is, not only are the individual tree data (14 trees) available from Garryland Wood, but equivalent individual tree-ring data is available from twelve other modern oak sites in Ireland, namely Ardara, Baron’s Court, Breen Wood, Caledon, Cappoquin, Enniscorthy, Glen of the Downs, Killarney, Loch Doon, Rostrevor and Shane’s Castle.  Moreover, individual tree data is also available for seven English and Scottish sites originally sampled by myself and colleagues at Belfast.  Thus anyone wanting to undertake research on tree-rings from the British Isles with respect to climate variables simply has to go into the NOAA World Data Centre for Paleoclimatology and access the data laboriously assembled, measured, documented and presented by workers at Queen’s University Belfast.

These comments are necessary because Mr Keenan has stated on your web site that at “QUB researchers do not have the expertise to analyse the data themselves and they do not want to share their data with other researchers who do”. Personally I would like an apology from both Mssrs Keenan and Peiser. However, I don’t expect to see one.

Mike Baillie

EDITOR’S NOTE [Benny Peiser]: Mike – Thanks for your response to Doug Keenan’s account. Let me make just one correction, as far as my role as editor of CCNet is concerned. Contrary to your perception, Keenan did not publish his text on my website. He published it on a popular U.S. climate blog called “Watts Up With That.” <http://wattsupwiththat.com/2009/08/14/another-uk-climate-data-scandal-is-emerging/> I only forwarded the essay because I considered it to be in the public interest, particularly in light of the ongoing data withholding controversy surrounding CRU (see <http://www.nature.com/news/2009/090812/full/460787a.html> ). In short, just because I circulated Doug Keenan’s text does not mean that I support his views or claims. As the CCNet disclaimer states explicitly: “The opinions, beliefs and viewpoints expressed in the articles and texts and in other CCNet contributions do not necessarily reflect the opinions, beliefs and viewpoints of the editor.” I hope this clarification will reassure you. BJP

Well it appears that Doug Keenan is not apologizing or going away:

D.J. Keenan <doug.keenan@informath.org>

Following are some comments on the claims in the Response of Mike Baillie (CCNet, 18 August 2009).

The Response claims that “anyone wanting to undertake research on tree-rings from the British Isles with respect to climate variables simply has to go into the NOAA World Data Centre for Paleoclimatology and access the data laboriously assembled, measured, documented and presented by workers at Queen’s University Belfast”.

Only a tiny portion of the data from QUB is in the World Data Centre (i.e. ITRDB). For example, there is no data in the ITRDB for prior to AD 1500; yet measurements go back 7000 years–as Baillie’s own publications state.

QUB originally made the same claim, but has now admitted that most data is on disks that have not been uploaded. And the Assistant Information Commissioner has visited QUB, and confirmed that he saw much more data. That most of the data has not been uploaded and that QUB has been “deliberately withholding data of climatic significance” (Baillie’s phrase) is thus provable, acknowledged by QUB, and independently verified.

The Response also claims that my post “gives the impression that data from only one Irish oak site is available, namely Garryland Wood”. It further claims that my post implied “the individual tree data from Garryland is not available”. These claims are not based on my main post, but on the page, linked by my post, at <http://www.informath.org/apprise/a3900/b910.htm>

That page presents a short, simplified, theory how Ireland is uniquely affected by the North Atlantic Drift and deep water formation and how this links with global climate.  Briefly, if you had to pick one place in the world to study the climate, Ireland would seem to be it.

After presenting the theory to support that, the page gives a simple example, to illustrate that the theory works in practice. The example uses averages from one site in Ireland–Garryland Wood–and some basic mathematics–correlation and addition. This was done so that readers who are unfamiliar with the relevant science could judge the viability of the theory for themselves, at least to some extent. (The page was originally written for people who might not have any scientific training — staff at the Information Commissioner’s Office and the Aarhus Convention Secretariat, to support my requests for the data.)

After presenting the simplified example, the page notes that a proper analysis should consider individual trees, trees at other sites, and more sophisticated mathematics.  The claims of the Response are based on misrepresenting all this, as if the example comprised the only data and the only mathematics that were available. Those claims are thus baseless.

The Response additionally quotes from my post, “QUB researchers do not have the expertise to analyse the data themselves”, and says that Baillie wants an apology for that. If Baillie has the expertise, why did he never publish any research using it?  Moreover, I have had several discussions with Baillie over the years, and have a rough idea of his mathematical skills. The branch of statistics that seems most relevant for analyzing the data (multidimensional time series, probably nonlinear) is difficult and specialized: if Baillie can pass an introductory-level examination in the subject, I’ll pay a large sum. (Note: I would not pass either.)

To summarize, the untruthfulness in Baillie’s Response is so obvious that it seems unlikely that it was intended to be believed. Rather, this is perhaps just Baillie’s way of saying “go away”. Up until 2005, there would have been nothing that could be done.  In 2005, though, the UK Freedom of Information Act came into effect. I look forward to seeing the Act enforced for such important data.

Douglas J. Keenan

http://www.informath.org

Now all of this is really about principles – the question of ownership of scientific data and the principle of scientific replicability and analysis that can only happen if data and methodology are willingly shared.

Doug also noted to me that the blog at Nature also mentions this spat and Doug appears to think that Nature is being rather disparaging about him being  praised on Climate Audit. I can’t quite see the slight myself but then I’m not an academic trying to protect my hoard of data from hordes of unwashed mathematical analysts who “might find something wrong with it”.

I think more importantly that both Climate Audit and WUWT have both opened the way for many people to reanalyze what we’re been told by populist magazines like Nature or Science which cheerfully admits that they filter their received papers to those that are deemed “provocative” by junior editors. Its easy to see how a science magazine’s published output can be skewed to the belief system of those junior editors.