date: Tue Mar  2 08:52:19 2004
from: Phil Jones <p.jones@uea.ac.uk>
subject: Re: Max/min/dtr data at the surface
to: "Russell Vose" <Russell.Vose@noaa.gov>

    Russ,
       I do have a permanent job, but I do like to plan ahead ! I think all these papers will
    be useful for the next IPCC AR4 report.
       On the various Reanalyses, it is essential that a few people assess how good they
    are with datasets like GHCN and CRU for surface temperature. I keep encouraging the
    GPCC people to do something similar for precip.
       I'm hopeful of doing something with the SH island and Antarctic MSLP datasets I
    have and NCEP/ERA-40 data, but this is just an idea.  I hear so many people giving talks
    saying that NCEP/ERA-40 are observations - given up trying to correct people !
       Looking forward to seeing Dave's results.
    Cheers
    Phil
   At 15:21 01/03/2004 -0500, you wrote:

     Phil:
     You already have tenure I presume, so there's no need to be thinking about so many
     papers in advance.  Unless you enjoy it, which is a completely different story...
     Sometime in April NCEP plans to release the North American Regional Reanalysis, which
     covers the period 1979-2002 and has a very high spatial resolution (something like 12
     km).  It might also be interesting to do some comparisons of GHCN and CRU with this new
     product.
     I processed your new file today and have forwarded the files to Dave Wuertz (he'll be
     doing the first difference computations).
     Phil Jones wrote:

      Russ,
           All the stations in the file sent come in through CLIMAT or MCDW or via Antarctic
      contacts. So all should have normals in the file.  I would suspect that this file also
      contains some updated data in 2002 (maybe 2001) as well as 2003.
          Once I get a better version of the ERA-40 paper I'll get back to you. It seems as
     though
      they have a slightly different agenda than me - lots of comparisons with a second
     version
      of ERA-40 as well as the standard one. Trying to figure out what this second version is
     -
      it seems to be a model with just SST/Sea Ice input but no sondes/synops. The ECMWF
      people are using this to try and determine the effects of observations in some way. It
      surprisingly does better than the official one in some parts of the world.
         At the moment I see 3 papers
      1. The direct GHCN/CRU and NASA comparisons.
      2. ERA-40 with CRU monthly mean temps.
      3. ERA-40 max/mins with GHCN - need to get the ERA-40 data to do this.
      Cheers
      Phil
      Cheers
      Phil
     At 16:38 27/02/2004 -0500, you wrote:

     Phil:
     Sorry for not responding yesterday -- they actually closed NCDC because of the snow,
     which ironically never had any impact on travel at all.
     I got your data and will merge them into what I already have on Monday.
     I take it that these are mostly non-U.S. stations and that they all should have normals
     in the "allnorms61-90.dat" file?
     If you are unable to include the GHCN-ERA40 comparison in your paper due to space
     constraints, I'd be willing to help out on a separate paper to that effect.
     Phil Jones wrote:

      Russ,
         I'll send the update file with station data for 1991-2003 when I can find someone to
     tell
      me why my disk can't be attached. Hopefully later in the day I can send it. It would be
     good
      to update the series to include 2003. The file will be the same as the one you have, so
      it should just be a matter of rerunning some programs.
         I will over the next few months add some stations, but haven't done this yet. There
     seem
      to be about 40 stations that now send CLIMATs as a result of being GSN stations that
     I'm
      not using. I have back data for them but my files have nothing for them between 1991
      and 2000 so I omitted them from the update file.
         As for the ERA-40 work, mins do warm a little more than maxs, but they've not
     calculated
      averages yet. I'm just looking at maps and can see more reds on one compared to the
      other. There are problems with a few countries though as they didn't get synops before
     1967.
      Australia is blue on all maps for 1958-2001 as all surface data is too warm before 1967
      by 1 deg C at least.
         I am hoping to compare your new GHCN gridded max and min fields with ERA-40. There
      might be too much in the paper with just the means though. With NCEP as well, the ECMWF
      people have sent me loads of plots and loads of text - all of which I've still to take
     in.
         Good luck with the thesis defence in April. I'm sure it will be fine.
      Cheers
      Phil
     At 17:18 25/02/2004 -0500, you wrote:

     Phil:
     In reverse order...
     1. Almost all is well with me.  I am set to defend my dissertation on April 2.  I did
     the coursework, field exam, and comprehensive exams while I worked in Arizona, but the
     diss itself has been a nemesis ever since I arrived at NCDC.  It's turned out well, but
     I will not miss it.
     2. Believe it or not, I have finally found time to write up some of the comparison of
     the NCDC, CRU, and NASA datasets.  In the process, it struck me that all of our trends
     are through 2002, but data are now available for all of 2003.  Do you think we should
     update them?
     If so, then can you supply us with your station data for 2003 so we can run them through
     the first difference method?
     3. The schedule for the GHCN max/min dataset is probably the same.
     I say "probably" because I just found out yesterday that NCDC has finally received World
     Weather Records data for 67 countries (including much of South America and Africa), and
     I am told to expect the data from the Asian WMO region in early April.  Since we
     actually have much of these data in the building, it would be a little irresponsible of
     me not to add them to GHCN before I release the new max/min dataset.  If I really can
     get all the data by early April, then it shouldn't take more than a month or so to
     eliminate the duplicates and to finish the Q/C because all of the software is written
     (except for one new spatial Q/C check we may add).
     Are you hoping to compare the new GHCN max/min data with the ECMWF?
     Given the results you've described for mean temperature, such a comparison could be
     pretty interesting.  Do you know if the min in the ECMWF is warming much more than the
     max?
     Phil Jones wrote:

      Russ,
         Is the schedule for the GHCN max/min dataset still the same?
     The people at ECMWF
      have worked quicker than they told me  ! They have some very interesting results when
      comparing their 2m mean temperatures with those we have in CRU. Agreement is amazing
      and when it isn't they can easily find the reason from the data input. For example, for
      Australia they were not getting any Synops before the early 1970s. Before the key date
      the model was about 1 deg C out, afterwards it became almost the same as the observed
      surface temperature data. Agreement is so good over N. America after 1967 that the
      Kalnay and Cai result seems an artifact of the NCEP data. Before 1967 there were fewer
      Synops and agreement is not quite as good. I would suspect that as ERA-40 uses the
      same input as NCEP, that the Kalnay and Cai result (i.e. their differences in trends
     with
      real month observations) is almost certainly spurious.
         How's work going on the comparisons between the NCDC, CRU and NASA datasets?
      I still feel this would make a needed contribution and would be referenced by the IPCC.
      The work with ECMWF would also make an excellent paper.
         Hope all is well with you.
      Cheers
      Phil
     At 10:18 26/01/2004 -0500, you wrote:

     Phil:
     We have an "alpha" version of the new GHCN max/min dataset in gridded (5 x 5) form.  We
     figure the final version will be done in 3 months.
     The station data are almost finished; it's the adjustments that will take a little more
     time (we're using a new procedure to do our homogeneity adjustments, and the people
     working on it told me Friday they would finish it in about 3 months).
     If you want the alpha data, just let me know...
     Phil Jones wrote:

      Dave, Russ,
          ECMWF won't be in a position to start on this for a couple of months - they need to
      submit their big paper on ERA-40. So there isn't a rush. I can forward the grids when
      ready. Appreciate seeing a draft of the paper when ready.
      Cheers
      Phil
      PS to Russ - don't forget the other work on direct comparisons of the 2 surface
     datasets.
      I see the ECMWF and NCEP comparisons as separate.

     E-mail: Russell.Vose@noaa.gov

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

     --
     Russell S. Vose, Chief
     Climate Analysis Branch
     National Climatic Data Center
     151 Patton Avenue
     Asheville, North Carolina 28801
     Phone: (828) 271-4311
     Fax: (828) 271-4328
     E-mail: Russell.Vose@noaa.gov

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

     --
     Russell S. Vose, Chief
     Climate Analysis Branch
     National Climatic Data Center
     151 Patton Avenue
     Asheville, North Carolina 28801
     Phone: (828) 271-4311
     Fax: (828) 271-4328
     E-mail: Russell.Vose@noaa.gov

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

     --
     Russell S. Vose, Chief
     Climate Analysis Branch
     National Climatic Data Center
     151 Patton Avenue
     Asheville, North Carolina 28801
     Phone: (828) 271-4311
     Fax: (828) 271-4328
     E-mail: Russell.Vose@noaa.gov

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