cc: <mann@virginia.edu>
date: Wed Sep  8 09:32:00 2004
from: Phil Jones <p.jones@uea.ac.uk>
subject: Re: your Rv.Geophys. paper
to: "Fritz Schott" <fschott@ifm.uni-kiel.de>

    Fritz,
       The data we used in most of the various figures are on the CRU web site. From the
    main page go to data, then to paleoclimate and then to the Jones and Mann paper. The NAO
    is in Fig 6b (based on the 1961-90 period and on Gibraltar and Reykjavik)
     An updated NAO plot is also here  (with the base period 1951-80 based on Gibraltar and
   Reykjavik)
   [1]http://www.cru.uea.ac.uk/~timo/projpages/nao_update.htm
      Now to your question.  The 'paleo NAO' series in Fig 6b all end early (Vinther in 1970,
   Cook
    in 1979 and Luterbacher in 1990). Of these three series, the Luterbacher et al. one is by
   far
    the most reliable as from the 1680s onwards it is almost always based on some pressure
   data
    (Paris from 1677-1713 and Uppsala from 1722, and loads of series from the 1760s). This
   series
    also agrees with the 'real' NAO best in Fig 6b. In the sentence you refer to in our text,
   we don't
    use the word 'trend' in the statistical sense. The highest positive values in the
   'observations' for
    30-year filtered data is in 1993 at 0.75 (numbers are in the later part of the file with
   Fig 6b). For
    Luterbacher et al it is 0.68 in their last year 1990. So they would likely get higher
   numbers a
    few years later if they had more years. If the last few years are added (4 of them, from
   2000/2001),
    the 'obs' line would start reducing, but the weights of the filter mean that 1993 would
   still be
    much the same value. We use Gaussian filters, so weights reduce away from the central
   value.
    They are not running means.
       The highest value in the Luterbacher series prior to the recent past was 0.72 centred
   on 1909.
    The value for the 'obs' then was 0.71. So, the maximum 30-year value (i.e the smooth line)
    was in the early 1990s (1992/3). This still seems unprecedented to me in the Luterbacher
    series which goes back to 1500.
       Recently the NAO has fallen below the peak values of the early 1990s, but the unusual
   peak
    in the series was in the early 1990s..
        I would recommend you get the data from our web site and not from Michaels. The Cook
    series there is his latest (from Jim Hurrell's book), but the series ends in 1979,so can't
   have
    the most recent late-1980s/early-1990s peak - and similarly for Bo Vinther's series.
      Hope this helps.
    Phil
   At 08:27 08/09/2004, Fritz Schott wrote:

     Dear Phil,
     I have to give an overview talk on Atlantic THC variability at the CLIVAR
     workshop here next week and would like to refer to your figure 6b on NAO
     variability and your quote (section 4.4.3, 1st para) "that the trend  in
     recent decades toward an enhanced positive phase of the NAO is relatively
     unusual, if not unprecedented" .
     It does not quite look that way when inspecting the Hurrell index where that
     1965-95 trend ended and the index fluctuates around normal now (I also do
     not see it in the (unfiltered) Cook index (that I got from Michaels Website,
     if I remember right).
     Could your statement and the corresponding end in fig 6b have to to with the
     fact that the figure is low-passed and thus not representing the last
     developments? Or do you uphold the quote? Your comments would be
     appreciated.
     Best regards
     Fritz
     trend in recent decades toward an enhanced positive phase of NAO relatively
     unusual, if not unprecedented, in recent centuries"
     trend in recent decades toward an enhanced positive phase of NAO relatively
     unusual, if not unprecedented, in recent centuries"
     can you pl. clarify a q

   Prof. Phil Jones
   Climatic Research Unit        Telephone +44 (0) 1603 592090
   School of Environmental Sciences    Fax +44 (0) 1603 507784
   University of East Anglia
   Norwich                          Email    p.jones@uea.ac.uk
   NR4 7TJ
   UK
   ----------------------------------------------------------------------------

