date: Thu, 29 Jun 2000 11:56:44 -0400 from: Ed Cook subject: Re: ARSTAN to: Keith Briffa I accept your comments totally. I would hardly view the lack of a comparison with the RCS method as a fatal flaw. But it still should be done at some point. >I tried ringing you about this - but without luck. I have to go and collect >my daughter (Amy) now so will try again tomorrow. We need to talk about this. >The comparison is far from simple. I don't consider that applying the RCS >at the site level is valid because many site collections are made up of >near equal age populations. The correct test at the scale we are working is >therefore the large region - i.e. produce one RCS curve , based on all >trees/species in the region. You may argue that a separate curve for each >genus in each area is valid , and then again , each species - as shown in >the example Figure 2. Who knows which is correct? Surely the point of the >paper we have submitted is that we have offered one solution - EVEN IF IT >TURNS OUT TO BE REPRODUCIBLE by some other , even existing method . I do >intend to do a more detailed paper describing and comparing the >ABD approach , but that is for a tree-ring audience (and it is not clear >what needs to be done in the comparison anyway). Here we show that this >method does capture more long-timescale information than the >Huggershof and that is important in this climate context. To negate the >effort because the RCS might do it in some manifestation is surely unfair. >Keith >At 10:44 AM 6/29/00 -0400, you wrote: >>Hi Keith, >> >>You always are right to the point. I agree that the "dot curve" idea may be >>"pretty much a waste of programming effort". However, the problem of >>unfilled band years is tractable I believe. If it is, then it may not be >>such a waste of time afterall. Of course, it may not do anything better >>than your banding method. >> >>Tim sent me the banding ms. I will look at it. Without having reading the >>earlier version all that carefully, I would say that the biggest "problem" >>with your proposed method is the lack of direct comparison with the RCS >>method. Have you done that now? If not, why not? It really should be done, >>you know, to show that the banding method has any advantage over the RCS >>method. I think it does, but thinking is not showing. >> >>Cheers, >> >>Ed >> >>=========================== >>Dr. Edward R. Cook >>Tree-Ring Laboratory >>Lamont-Doherty Earth Observatory >>Palisades, New York 10964 >>Phone: 914-365-8618 >>Fax: 914-365-8152 >>Email: drdendro@ldeo.columbia.edu >>=========================== > >-- >Dr. Keith Briffa, Climatic Research Unit, University of East Anglia, >Norwich, NR4 7TJ, United Kingdom >Phone: +44-1603-592090 Fax: +44-1603-507784 =========================== Dr. Edward R. Cook Tree-Ring Laboratory Lamont-Doherty Earth Observatory Palisades, New York 10964 Phone: 914-365-8618 Fax: 914-365-8152 Email: drdendro@ldeo.columbia.edu ===========================