From: Tim Osborn To: Stefan Rahmstorf , Fortunat Joos Subject: Re: latest draft of 2000-year section text Date: Wed, 01 Mar 2006 16:59:37 +0000 Cc: Jonathan Overpeck ,Keith Briffa , cddhr@giss.nasa.gov,Eystein Jansen Hi again Stefan, At 14:55 01/03/2006, Stefan Rahmstorf wrote: What about saying something along the lines: >"At present, the extent of any such bias in specific reconstructions >is uncertain, although probably not as large as suggested by Von >Storch et al. (2004), whose work was affected by a calibration error >(Wahl, Ritson and Amman, 2006)." This sounds good and Keith is currently working your suggested wording into the paragraph in question. >p.s. Tim: Are you convinced the more recent papers by the VS group >use the correct calibration? In those curves that are intended to >show the pseudoproxies perform poorly even when calibrated >correctly, as long as you add a lot more noise, I wonder why the >pseudoproxies perform poorly even within the calibration interval, >where they now should be calibrated to properly reproduce the 20th C >warming trend, and they don't? I am not certain, of course. And yes, there is a link between the degree to which the trend over the calibration period is captured and the amplitude of long-term fluctuations in the reconstruction. That many of Burger's multitude of methods do not obtain the full warming trend, while Mann et al. do, is certainly a concern here. But it is also true (and I have myself analysed this one year before von Storch et al. was published - if only I'd realised the implications I could have had another Science paper! :-)) that correct implementation of a regression method, keeping the trend in, can still lead to a massive underestimation of that trend. So there's still more work to be done on this topic! Cheers Tim 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