cc: David Easterling <David.Easterling@noaa.gov>, "Parker, David (Met Office)" <david.parker@metoffice.gov.uk>
date: Fri Apr 22 13:12:09 2005
from: Phil Jones <p.jones@uea.ac.uk>
subject: Re: Figures and Tables  in 3.2 and also 3.3
to: Kevin Trenberth <trenbert@ucar.edu>

    Kevin,
       A few thoughts. Fig 3.2.6 will bring out the recent land/ocean diverging trends. I'll
   bring
    that in to the revised 3.2 I'm currently doing.
        The possible new addition to 3.2.4 could replace what we currently have. I'm a little
    wary about have too many plots of this kind as the reviewers will want even more. It will
    likely be difficult to get an in situ plot like 3.6.10.
       David is working on new Tables 3.2.1 and 3.2.2 so the comments about digits are
    pertinent.  We should be able to decide on these in Beijing.
        One final comment on figures in this section is that the rest of Fig 3.2.4 needs
   updating
    and needs some of the other SST analyses added. We said add Ishii et al 2005. We'll
    need an NCDC one as well, and need to get GISS into 3.2.1.

        After Beijing we should have someone responsible for each Figure, so that updates
    can be done for the FOD, SOD etc. Similarly for Tables. If we all have a spreadsheet of
    responsibilities we'll know who to contact.
    Cheers
    Phil
   At 18:19 21/04/2005, Kevin Trenberth wrote:

     Phil some comments:
     Phil Jones wrote:

      Dave, David and Kevin,
           In addition to Kevin's point about where to break the map (30E is another
     possibility),
      I have now been through both sets of comments on 3.2.
        On the Figures, all the SH people have rightly asked for the seasonal maps (i.e
     .3.2.9 and 3.3.3)
      be labelled December to February, March to May etc, rather than the Winter, Spring
     there
      at the moment. Can you pass this on to Byron, Dave?
        None of the reviewers said anything about time latitude sections, only requests for
     more
      regional aspects.

     I did though.  Question is how to handle this.  I don't much like linear trends, so maps
     of linear trends are not sufficient.  I am thinking more and more that a big issue
     emerging is the increasing trend of land vs ocean.    We have a land vs ocean figure:
     Fig 3.2.6?  Is that the right one? We need to capture that well.  This relates to the
     max vs min and DTR issue, and also MSU issue.  How to bring out where the land vs ocean
     are separating: maybe that's in the 1979-2004 trend map?

     Figures 3.2.5a and now b have additional time series plots for the Tropics
      (20N-20S) and NH N of 20N and the SH S of 20S. Having the tropics answers one reviewer
      (Fu), but I feel, David, the extratropics are too like the whole hemispheres. Chris
     Folland
      wanted an Arctic plot. Doing this would mean we could get into Ch 4 - they had an NCEP
      temperature average for the Arctic in one of their answers to a CQ! So, can you ask
     John
      Kennedy to see if he can do an Arctic and an Antarctic plot. Do the latter from only
     1958
      onwards - there is some earlier data but it is sparse and very noisy. I guess once
     HadCRUT3
      gets used for the SOD (or maybe the FOD) we can get error bars on these non-hemispheric
      plots.

     Please also note Fig 3.6.10:   This one is based on remote sensing though and can in
     situ match it?  This one relates to SAM.
     Also for use in the hurricane section, can we include either the attached figure or a
     new version as a replacement for 3.2.4e.
     This is based on HADISST.

          I would say the two polar regions are poleward of 65 degrees.  Any views on this
     and
      dropping the two extratropic plots?
         As for the Tables (3.2.1 and 3.2.2) there were a number of comments. John Janzante
      wanted them laid out better. His comment is on p10 of the 66 page set of LA comments.
      Basically he's suggested omitting the significance level and use italics and bold to
      indicate 5 and 1% conf. levels. The ranges would go on the line below the trend
     estimate.
         This seems reasonable.  I still think they should be in a Table, but the plot Kevin
     used in
      3.4.1 was effective.
         Finally, there was one comment on the precision used (DK comments # 3.28 on pg 8 of
     33).
      I have some sympathy for one less significant figure for the shorter periods, but that
     three
      is needed for the 144 and 104 year periods, otherwise for these they all be similar.
     Need to
      consistent for all periods, so stick with three. David is going to redo these, so any
     views?

     I strongly agree with 1 less digit.  For longer period, units maybe should be per
     century?

      Cheers
      Phil
     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    [1]p.jones@uea.ac.uk
     NR4 7TJ
     UK
     ----------------------------------------------------------------------------


--
****************
Kevin E. Trenberth                              e-mail: [2]trenbert@ucar.edu
Climate Analysis Section, NCAR                  [3]www.cgd.ucar.edu/cas/
P. O. Box 3000,                                 (303) 497 1318
Boulder, CO 80307                               (303) 497 1333 (fax)

Street address: 1850 Table Mesa Drive, Boulder, CO  80303

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

