date: Mon Dec 15 14:17:34 2008 from: Tom Melvin subject: Re: NERC proposal to: Martin Wilmking Martin, I would start by separating trees detrended with horizontal line from the others. Then try signal-free method. Should produce slight improvement but each chronology will have an "unknown" overall slope. Then try RCS method, but hopefuly with sufficient trees to use multiple RCS curves (for separate growth rate classes of tree) and see what resuls are like. Can group trees with a common signal when building RCS curves. Can you send me what data you have? I would only use it with your permission. Tom At 14:01 15/12/2008, you wrote: Tom, great if you could run the data! In the data I collected there is some pith offset (which we could "correct"), but since we are also using data collected by a colleague who worked at the same site and I dont know the offset, maybe it would be best to only test the traditional version? Martin PS I am ccing Dr. Jayendra Singh, who is currently working on the data Tom Melvin wrote: Martin, Yes. I am currently amending program but should be able to run the data through signal-free method, both curve-fitting and RCS. Hopefully programs will be sufficiently developed soon (with help and documentation) for you to be able to use them. Do you have any missing pith offset data for the sites. For "modern" chronologies it can make a significant difference to RCS. Tom At 13:20 15/12/2008, you wrote: Dear Tom, thanks for the copy! Dear Keith, get well soon. Martin btw, Tom, we have a data set from eastern Alaska, where we do have massive change in temp sensitivity (about 90% of sampled trees), if we detrend traditionally or with RCS. Would it be possible for you to test this data set with your signal free approach, since it seems as if the Canadian sites (Mackenzie Delta) close by are affected by standardization method? thanks martin Melvin wrote: Martin, Attached is a copy of the proposal that we submitted and an explanatory letter from Keith (who is sick with flu at the moment). Tom Dr. Tom Melvin Climatic Research Unit University of East Anglia Norwich, NR4 7TJ, U.K. Phone: +44-1603-593161 Fax: +44-1603-507784 -- Martin Wilmking, Ph.D. Working Group: Ecosystem Dynamic Institute for Botany and Landscape Ecology University Greifswald Grimmer Strasse 88 D - 17487 Greifswald, Germany Tel: +49 (0)3834-864095 Fax: +49 (0)3834-864114 [1]http://biogeo.botanik.uni-greifswald.de Dr. Tom Melvin Climatic Research Unit University of East Anglia Norwich, NR4 7TJ, U.K. Phone: +44-1603-593161 Fax: +44-1603-507784 -- Martin Wilmking, Ph.D. Working Group: Ecosystem Dynamic Institute for Botany and Landscape Ecology University Greifswald Grimmer Strasse 88 D - 17487 Greifswald, Germany Tel: +49 (0)3834-864095 Fax: +49 (0)3834-864114 [2]http://biogeo.botanik.uni-greifswald.de