From: Tim Osborn To: "Phil Jones" ,"Keith Briffa" Subject: Fwd: Re: McIntyre-McKitrick and Mann-Bradley-Hughes Date: Fri, 07 Nov 2003 16:12:53 +0000 >From: "Sonja.B-C" >Date: Fri, 7 Nov 2003 15:58:06 +0000 >To: Steve McIntyre >Subject: Re: McIntyre-McKitrick and Mann-Bradley-Hughes >Cc: L.A.Love@hull.ac.uk, Tim Osborn , > Ross McKitrick >Priority: NORMAL >X-Mailer: Execmail for Win32 5.1.1 Build (10) > >Dear Steve >Please send your material for comment direct to Tim, Osborne.I >would like to publish the whole debate early next year, but >'respectful' comments in the meantime can only help and the CRU people >seem genuinely interested and have integrity. I have never heard of >such bad behaviour here as appears to have been the case between >Sallie and Soon and the rest..the US adversarial system and too many >egos?? >As you know ,the contact is Tim Osborn and I take >the liberty to forward this to him now. You seem to suggest that this >is welcome and are making make direct comments on his remarks to me >concerning your paper. > >We shall get the printed proof, as a single electronic file today, and >shall look through it early next week. I am sure you do not want to see >your paper again? I think that adding anymore now (the exchanges >between you and Mann/Bradley and perhaps now Tim as well) is premature >and we shall wait until the next issue. Mann is said to be writing >something, but he has not yet contacted me, though I just hang up on >that journalist Appell who keeps on ringing. I told him that I will >deal only directly with Mann. What cheek, after threatening me with >litigation...Just keep me in the loop. Thanks. > >Sonja >PS .By the way The Economist has taken up a previous paper from E&E >(Castles and Henderson, the social science critique of teh emission >scenarios), and teh Australian and UK Treasuries have become involved. >I have not seen it yet. As you know, I have always argued that the real >'driver' of teh IPCC deception, if that is the right word, has been on >teh social /technology forcing side, with focus of WG III. > >In London I heard two days ago that the WTO might make ratification of >Kyoto conditional for something Russia wants. The source was speaker >from the Deutsche Bank, a Justin Mundy, former advisor to the EU >Commission on EU-Russia coordination and once senior advisor to the >European Centre for Nature Conservation, he also worked for the World >Bank.) >Sonja > >On Fri, 7 Nov 2003 09:50:33 -0500 >Steve McIntyre wrote: > > > Dear Sonja, > > > > > > The interesting thing about their preliminary response, however, is > that it > > > > indicates that the difference in results might be fully explained by a > > > > simple error in not using many of the early tree-ring data. If > this is > > > > confirmed by their fuller response, then, even though there may be > some > > > > problems with the proxy data used by Mann et al., it implies that > these > > > > problems do not actually make a lot of difference to the results - > the main > > > > difference comes from omitting the early tree-ring data. A paper that > > > > identifies some problems with the proxy data used by Mann et al. would > > > > still be interesting, but if these problems made very little > difference to > > > > the results obtained, then it would be of rather minor importance. > > > > > > (1) IMHO the data issues rise above "some problems". When you're > doing a prospectus, audit or engineering-level feasibility study, there > is a concerted effort to eliminate every error. I have never seen such > sloppy data as MBH98. Perhaps from my business experience, I am used to > a more demanding approach to data integrity than the above comment > suggests about academic studies. Even the MBH response criticizes us for > failing to use obsolete data. How silly is that. Bradley has also said > that an "audit" should use original data and should not verify against > source data and says that I should know better. I think that my > experience with audits and engineering studies is more substantial than > Bradley's and this is an extraordinarily silly thing for him to > say. After the fact, one of the key mis-steps in the Bre-X fraud was > the engineering report in which ore reserves were calculated using false > data supplied to the consulting engineers by Bre-X, without any > verification being carried out by the engineers. > > > (2) There was not a "simple error" of simply not using many of the > early tree-ring data. The early tree-ring data in question are principal > components of North American tree ring sites and of Stahle/SWM (also > North American) tree ring sites . MBH98 states that they used > conventional principal components methods for temperature. They do not > explicitly say that they used conventional principal components methods > for tree ring regions, but, in the absence of disclosure otherwise, this > is certainly the most reasonable interpretation of the public disclosure > (leaving aside Mann's refusal to provide clarification in response to our > inquiries on methods.) A "conventional" principal component calculation > requires that there be no missing data. Accordingly this indicator became > unavailable in the earlier years using conventional principal component > calculations - it was not "left out". MBH now disclose for the very > first time that they used a "stepwise principal components approach", > although this is nowhere disclosed in MBH98 or in the SI thereto. They > have still not disclosed the rosters of principal components involved. If > this method is material to their results, as they now state, then it was > a material omission in their prior disclosure. It seems like a very > strange rebuttal for MBH to say: you're at fault because we made a > material non-disclosure on methodology in our papers. If I were in MBH's > shoes, I would be embarrassed at this non-disclosure and mitigating the > situation by making full disclosure now. . When you do a prospectus, you > have to sign an affidavit that there are no material omissions. I have > approached disclosure questions on the basis that prospectus-level > disclosure is the minimum level of public disclosure in this matter, > assuming that this level of disclosure would be exceeded. > > > > (3) I've redone calculations with a re-calculated US PC1 in and get > results similar to those in E&E, rather than the MBH response. This is > not a guarantee that I have fully replicated still undisclosed MBH > methodology. However, MBH disclosure of their methodology is very > inadequate and without full disclosure by MBH of their methods, it is > possible to be somewhat at cross-purposes. This defective disclosure is > entirely their responsibility. It should be remedied immediately through > FTP disclosure of their computer programs and full description of their > methodology. > > > > [snip] > > > > > > > > > >>It is quite obvious that if the opinion of these three people > from the > > > > >>UK University of East Anglia concerning publication of teh M&M paper > > > > >>had been sought and taken, there would not have been no publication. > > > > > > > > Then I suggest you read our commentary again, which does not state > this at all. > > > > > > Part 2 has been drafted and I would be delighted to obtain comments on > it from UEA/CRU. Indeed, I think that it would be very constructive, > since Part 2 is significantly more hard-edged than Part 1. Because we > have stated that we would post up a reply to the MBH response, we would > have to disclose something on our websites, but I'd be prepared to deal > with this. Intuitively, full, true and plain disclosure would be to state > that we have prepared a reply and submitted it to UEA/CRU for > comments. I think that the many data errors will be self-evident to > UEA/CRU; we have organized our materials to show this, as will be the > material non-disclosures on methodology by MBH. However, if they are > prepared to comment, this would have to be agreed on very quickly as we > are very close to finalizing our repy. > > > > Regards, > > Steve > >---------------------- >Dr.Sonja Boehmer-Christiansen >Reader,Department of Geography, >Editor, Energy & Environment >(Multi-science,www.multi-science.co.uk) >Faculty of Science >University of Hull >Hull HU6 7RX, UK >Tel: (0)1482 465349/6341/5385 >Fax: (0)1482 466340 >Sonja.B-C@hull.ac.uk Dr Timothy J Osborn Climatic Research Unit School of Environmental Sciences, University of East Anglia Norwich NR4 7TJ, UK e-mail: t.osborn@uea.ac.uk phone: +44 1603 592089 fax: +44 1603 507784 web: http://www.cru.uea.ac.uk/~timo/ sunclock: http://www.cru.uea.ac.uk/~timo/sunclock.htm