date: Fri Jan 16 10:27:37 2004
from: Phil Jones <p.jones@uea.ac.uk>
subject: Re: 1. Scottish temperature data  2. NAOI 
to: "Max Beran" <maxberan@oldboot.demon.co.uk>

    Dear Max,
        It was Marc Becker who was Project Officer from SEPA for the work on the Scottish
    and NI work. He would be pleased so thank for adding it all in. He will be happy for the
   data
    to be used as often as possible, so go ahead with using it. Paper was accepted earlier
    this week - email from IJC. I guess I'll get proofs in a month or so, so expect it to be
   out
    by the summer. Email later if you want to check page numbers etc, as they often send
    these before it comes out.
       As for the NAO, the paper is attached.  This discusses various indices as does the
    book - which is the thing to get on the subject. It is AGU's best selling Geophysical
    Monograph by the way.  Much of my take on the subject is in the paper, so if you've
    not got it it should make good reading. Important in my mind is the phenomenon itself,
    how we measure it and then how it impacts temperature and precip patterns. Much of what
    I see if that it does change but also its influence changes and the two are independent of
    each other. The influence doesn't change because data are poorer before 1950. They aren't.
    They may be before 1850, but the influence is clearly weaker between 1920-60 - just
    at the same time as there was a weaker influence of ENSO and less strong and maybe
    fewer El Ninos.
       So there is a lot of variability and just saying it is decadal variability doesn't mean
   that we
    can forget about it !  Also putting this in a box doesn't mean we know what is causing it.
    My view is that we will never know. Is the influence stronger now because of anthro
   effects -
    who knows, it is a possibility but no more than that.
       I also think that there is too much emphasis on the NAO  (and SOI) and other factors
    become more important at times.  There is some modelling work with volcanoes which
    says that the aerosols impact the stratosphere which then feeds back on the troposphere
    causing the NAO to be more positive in the winters after eruptions. Maybe the 3 large
    eruptions in recent decades and the lack of eruptions between 1920 and 1960 is a factor.
    Again who knows.
       I don't think of David Stephenson as the leading authority on the NAO in the UK. I
   don't
    think we have one - we all know bits.
        There seems nothing special about recent trends in the instrumental record. There does
    if the paleo stuff is used - but here you come back to changes in the influence.
       Finally, you can explain much more of CET, EWP if you develop an index based
    on Plymouth and Lerwick. The NAO is a fundamental mode of variability, but is it
    nothing more than a measure of westerly wind strength?
    Cheers
    Phil

   At 11:15 13/01/2004 +0000, you wrote:

     Dear Phil

     1.    Thanks again for sendng me the Scottish temperature index series. I used the
     post-1861, winter-season, mainland data to illustrate my talk to the Royal Met Soc's
     Scottish Group last Friday. I think the talk was appreciated though I might have been a
     bit heavy on the statistics as there were some heavy eyelids in those bits. I'll try to
     rectify when I repeat the talk to the North-East group in the Spring.

     Anyway, among the audience was a SEPA person; I think his name was Becker but I might
     have misremembered. He seemed "over the moon" (so perhaps his name was Beckham, not
     Becker)  that I had used the data which I guess he has had a hand in funding the
     compilation of. As you requested I gave you verbal and slide-embedded acknowledgment,
     and IJC-in-press as a reference.

     Having  done all the preparatory work for these talks, I'm now ready to put together
     Part 2 of my Weather article on Record Breaking (part 1 was in August 2002). This is to
     be on record breaking under non-stationary conditions. I'd like to use the same run of
     data for this article especially as there is a strong user interest in Scottish Winter
     data. Articles have appeared on the ski industry and on the longevity (or shortevity) of
     snow patches. I don't suppose for a moment my article will appear before your IJC peer
     review process but I'll check before submission. It certainly doesn't sound like SEPA
     are going to mind.

     2.    I wonder if I can steal half an hour of your precious consultancy time to ask you
     about another matter entirely. What is your take on the NAO? A couple of things I've
     done recently on floods, wind and temperature, show numerically just how much apparently
     secular change the NAOI can swallow. But I find myself rather at odds with others
     (perhaps with you too) on how to view this.

     The way I see it is that the NAO is a reflection of large-scale, long-term, almost
     certainly nonlinear interplay between atmospheric and oceanic dynamics, and even
     land-surface hydrological processes. I imagine water of varying salinities sloshing
     around in various compartments, building up to thresholds, spilling over, to
     neighbouring compartments horizontally and vertically, creating shifts in the average
     position of Atlantic/Arctic ocean circulation patterns and through them, on modes of the
     atmospheric general circulation. We don't have much of a clue about any of the details
     and perhaps never will, but we may assume that something like this is what is in place -
     needs to be in place to explain the dominant fraction of the spectral energy in the NAOI
     time series that is present at multi-decadal periods. The NAOI, as I see it, is an
     imperfect window into the NAO which happens to work reasonably successfully in winter
     when other more immediately energetic factors are relatively unimportant.

     So, after that lengthy preamble, when I see that NAOI hugely out-correlates and
     out-explains year number in a regression, I interpret this as evidence for multi-decadal
     background, natural fluctuations fooling us into thinking that there is more secular
     (and hence anthropogenic) control on weather than there really is. I am aware that
     others do not. In particular Stephenson, who I take to be the number-one authority on
     the subject in the UK, hardly mentions the long-term dynamics and talks as if the NAO
     was itself being pushed around by the same, even mightier forces, that are pushing
     around other climatic elements. One sees it in papers and IPCC report where it is
     supposed that global warming will affect NAOI to the same degree as it does temperature;
     also in the literature on extending the NAOI back in time on the basis of local
     variables. Then there are the many references to the recent strong positive mode but
     this is very unconvincing as anything special when you look at the entire time series
     with its zero trend. Anyway, aren't we supposed to be due for weaker meridional flows as
     the zonal contrast builds up?  Much of this strikes me as topsy-turvy when, in my
     version of the world, NAO (doubtless among other powerful prime movers) stands upstream
     of local weather manifestations such as storminess, CET or EWP.

     Would you mind emailing me an electronic version of your Geophysical Monograph 134
     article.

     Thanks

     Max

          -----Original Message-----
          From: Phil Jones [[1]mailto:p.jones@uea.ac.uk]
          Sent: 17 December 2003 14:34
          To: Max Beran
          Subject: RE: CRU half-degree temperature series v CET
           Dear Max,
               At this very moment I'm modifying a paper submitted to IJC about Scottish and
          N. Irish
           temps. The paper will go back to IJC (accepted subject to minor mods) and the
          report will
           go to SNIFFER, part of SEPA.
              What you want I'll attach. First the 3 monthly temp series in abs deg C *10,
          then the
           figures. These are those I altered re Armagh/NI in September, so haven't checked
          these
           are OK yet for resubmission. Finally the paper which I've just altered, which
          should make all
           clear.
               Data go to 2002. Met Office should be updating series at some time in future
          years.
               Anything more yet on CET ?
              Have a good Christmas and New Year break - I guess it's all break for you now !
           Cheers
           Phil

          At 14:15 17/12/2003 +0000, you wrote:

          Dear Phil
          I'm due to give a talk on record breaking to the Scottish Group of the Royal Met Soc
          in January and was looking around for a suitable Scottish data set to use as an
          example. I could use the TS2.0 data especially as Tim Mitchell seems to have
          softened his line on the legitimacy of using those data in time series mode.
          However I note from this earlier email that you have a Scottish equivalent to the
          CET which is presumably homogenised in the temporal sense so might prefer to use
          that rather than creating my own from TS2.0 with its foibles.
          Any chance you can point me to it - I will naturally acknowledge?
          Max Beran
          -----Original Message-----
          From: Phil Jones [[2]mailto:p.jones@uea.ac.uk]
          Sent: 01 May 2003 09:59
          To: Max Beran; david.parker@metoffice.com
          Cc: mark.new@geography.oxford.ac.uk; t.mitchell@uea.ac.uk
          Subject: RE: CRU half-degree temperature series v CET
           Max,
              David makes a good point re the NAO and the SST change around our coasts. Also
          the
           NAO influence is weaker in the 1920-1950 period.
              Tim Osborn has a paper in ASL (the RMS online journal) which looks at the
          circulation
           influence on CET and EWR. Email Tim if you can't find this on the RMS web site. He
          has
           a pdf version, I do also but I can't find it.
               In CRU we were given loads of papers by Manley's widow around 1980 but these
          are just
           papers. Durham have a diagram of Manley's given to them by Jean Grove about 5 years
          ago.
           I'm not sure whether his working papers will be in Cambridge - nor any idea who
          would
           know if they still exist. Dick Grove might be a starting point.
              I am doing some work on Scottish series. Over the 1901-2000 period the gradient
          is to
           slightly less warming than CET towards the NW, so the Scottish Islands warm less
          that
           CET, with Scottish lowland/upland sites warming only marginally less than CET.
          Period
           is different though.
               Also, as we've said before CRU05 and TS2.0 were not designed for climate change
          studies.
           Tim will be making this clear in the next version of Mitchell et al. (2003).
              I could send you monthly series for De Bilt and Armagh if you're interested in.
          The Dutch
           series goes back to the 1700s and Armagh to 1860. Both have had considerable work
          done
           on them in recent years.
           Cheers
           Phil
          At 22:46 30/04/03 +0100, Max Beran wrote:

          Dear David

          Thanks for those initial reactions.

          I'm aware (and have a copy) of your "official" paper but the internal Met Office
          report contains the sort of detail that is more a propos an unpicking exercise as
          this.

          Your interpretation of the CET/NAOI offset hadn't occurred to me though I'm not sure
          how convincing it is. It seems to rest on a rather too simple transport of warming
          from whatever phenomenon lies behind the dynamics of the North Atlantic pressure
          gradient through to central England via one mighty step whose only mediator is the
          ocean surface temperature.

          My knowledge of urban warming is perfunctory and largely gathered since getting into
          this exercise. I had been brought up on the notion that urban warming was a property
          of winter, of nighttime, and of major metropolises. I now gather that it is vastly
          more complicated, far reaching, and a wide variety of temporal patterns of warming
          can arise, seasonally and diurnally, depending on development type, location and
          baseline climatology.

          You express surprise about the steepness of the spatial gradient of trend. Is it the
          direction of the gradient or its magnitude that is surprising? I've just been
          reading a paper by Julian Mayes in Geog Journal, Vol 166, pp125-138 about SE to NW
          regional gradients - he shows a diagram (Figure 8c) that is entirly compatible with
          that sort of number, it could possibly accommodate a larger one.

          Do you have any inside info on Manley's precise procedures beyond what he reports in
          his QJ papers? In other words can you save me the fag of going over to Cambridge?

          I look forward to more.

          Max
          -----Original Message-----
          From: david.parker@metoffice.com [[3]mailto:david.parker@metoffice.com]
          Sent: 30 April 2003 11:11
          To: Max Beran
          Cc: Phil Jones; mark.new@geography.oxford.ac.uk; t.mitchell@uea.ac.uk;
          chris.folland@metoffice.com; richenda.connell@ukcip.org.uk;
          jim.briden@environmental-change.oxford.ac.uk
          Subject: Re: FW: CRU half-degree temperature series v CET
          Max
          I read your interesting attachment on CET and have some preliminary comments:
          The definitive "Parker et al" is in Internat J Climatol 12, 317-342 (1992).
          Urban warming may be greatest in summer because a) Retention of solar heat by
          building fabric is greater and b) it is, in general, less windy.
          We have made preliminary analyses of daily CET anomalies on windy vs calm days.
          Warming is not significantly greater on the latter, as would be the case if there
          were marked urban warming. However there is, as expected, considerable noise in the
          anomalies, making detection of small effects difficult by this method.
          A positive offset of 1951-2000 CET relative to a regression against NAO based on
          1901-50 is to be expected as the ocean has warmed.
          The trend difference exceeding 0.5 deg C / century between SE and NE England in
          TS2.0 seems surprisingly large.
          We will get back to you again in due course
          Regards
          David Parker

     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
     ----------------------------------------------------------------------------


   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
   ----------------------------------------------------------------------------

   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
   ----------------------------------------------------------------------------

