date: Thu, 17 Jan 2008 12:24:12 +1300
from: David Thompson <davet@atmos.colostate.edu>
subject: Re: Some text - sent again!! 
to: Phil Jones <p.jones@uea.ac.uk>

   Hi Phil,

   Thanks much for the great comments and for catching a couple of errors. I've made a number
   of changes to the manuscript, and have also addressed some points specifically below. I
   plan on sending a revised version to the whole group for any final comments within the next
   few hours.

   An aside: I wanted to acknowledge John Kennedy's work on the SST data in the authorship on
   this paper, but I expect the list will be different for the volcano paper. If you have any
   concerns about the authorship list, please let me know.

   Some clarification and responses....

   1. I know what you are saying about going 'over the top' for Nature. But I'm hesitant to
   add the ranking of warm years mainly because the COWL filtering removes a component of the
   warming due to circulation changes, which may or may not be linked to anthropogenic
   forcing. I decided to submit to Nature mainly because a lot of IPCC folks strongly
   suggested the errors in the SST data should be disseminated to a wide audience. If the
   Nature editors decide the discontinuity isn't worth publishing, then I'll be just as happy
   with J Climate. I plan on analyzing the residual time series in more detail in the longer
   paper on the volcanoes.

   2. The COWL time series is dimensionless, so I think the correlation coefficients given in
   the table are probably more useful than the regression coefficients....

   3. I've clarified that smoothing the data will not give you the same result as filtering
   the COWL pattern. This is because the COWL pattern has the same temporal resolution as the
   original data. So our 'residual' time series has less high frequency noise, but it has not
   been low-pass filtered. This is key for identifying the step in 45.

   4. Fig. 4 looks a little cluttered with more % time series on it, but if the paper survives
   review I could perhaps add a new time series with lots of different countries on it... I
   really like the slide from Scott Woodruff, but couldn't find a refereed reference to add
   that information. Perhaps we can add the anecdote about the shortsightedness of the US
   Maritime Commission as 'personal communication' in the revised version.

   5. The TENSO time series I'm using is similar to a lagged and low-pass filtered ENSO index.

   6. The COWL time series is linearly related to the NAM, but also to any other pattern that
   changes the large scale NH circulation (ie, the PNA, etc).

   Thanks again,

   Dave

   On Jan 16, 2008, at 3:57 AM, Phil Jones wrote:

      Dave,
         We've had problems with our email system. Sending again, so you may get this twice.
      Cheers
      Phil
      Dave,
         Rather than go through the doc file, I'll make a few points directly by email.
      1. I'll reckon you'll have to go over the top to get Nature to send this out for
     review.
      One way of doing this would be to add in some quick analyses of the
      residual global mean series. for recent years. Only a few sentences.
      Basically to show that years like 2005 and others in the period 2002-2007 are
      after extraction warmer than 1998. Maybe also over 1997/8 to 2007 show
      the trend. I know this is somewhat silly, but there is a lot of rubbish on
      web sites about global warming stopping. Maybe just rank the top ten
      years in the residual series. This might give it more appeal, but not detract
      from the main 1945 message.
      2. The variability increase as you move back in time is mostly sampling
      and inevitable. Probably needs saying - could a foot/endnote. I don't
      think this would be any different if you'd used HadCRUT3v or CRUTEM3v,
      but I'd like to believe it would be!!
      3. You mention regressions for ENSO and COWL. Can you give these
      regression coefficients? Or are they sort of pre-determined from theory?
      I read this late last night and couldn't quite follow.
      Some less important points.
      1. With COWL, you could essentially smooth the series and get much the
      same result i.e. if you extracted just COWL then smoothed and compared
      with raw smoothed they would look much the same.
      2. The companion paper on volcanoes ought to try and set up a measure for
      the drops - so it is all repeatable. I reckon there are one or two smaller NH
      volcanoes you might pick up.
      3. With Fig4 you probably also need to show a plot of the overall SST count
      per year. % US/UK is OK, but this hides much reduced numbers in the 1940s.
      Perhaps you could show a plot of %US, %UK and %others in three colours
      in a cumulative way - soo all adds to 100%.
      4. The bucket/engine room argument is the wrong way round. The change
      was to UK ships in Aug 45.
      5. Could also add that the UK/US are digitizing more WW2 UK (RN) ships. I take
      it you're aware of the fate of the US WW2 navy ships. Scott Woodruff sent me
      the attached - he blacked out the name of who signed this!
      6. We are the Climatic Research Unit and you could give our web site
      where all these series are.
      7. T subscript E isn't defined. It is the whole tropics from surface to TOA
      to get the 255K. It's not defined as I can see why it is E. E for equator?
      8. I presume the regressions are also based on 1950-2006 just like
      the correlations in the Table. Also these are monthly or appear to be from the
      plots. Are they similar for annual data?
      9. With the ENSO series in the middle of Fig1. You could show how this
      compares with a suitably smoothed SOI. It looks as though it will agree
      well.
      I'm attaching an article where I sort of did this with regression against the SOI.
      10. Does COWL look like a high-freq NAM?  COWL seems to lose
      variability before 1880.
      Cheers
      Phil
     At 03:05 14/01/2008, you wrote:

     Phil,
     You are more than welcome to show any slides... hopefully the work
     will be well into the review process by then... in fact, I'd be happy
     to grind out similar analyses for a longer paleo record for you, if
     you like.
     As for the papers: as I mentioned I've shown the step in 45 work to a
     number of folks closely involved in the last IPCC, and all were
     adamant we should submit that particular result to Nature. So I
     figured I'd give that venue a shot, saving the volcano results for a
     slightly longer J Climate paper.
     Anyway, I've attached a first draft of the 'step in 45' paper
     (figures as pdf; text as pdf and doc). I figured I'd run the text by
     folks individually once before I send a draft to the group. If all
     goes well I hope to try and submit something in the next couple weeks...
     Please feel free to offer any comments that come to mind, both
     general and specific.
     Thanks again for collaborating with me on this... it's fun....
     -Dave
     ￼￼￼
     On Jan 12, 2008, at 12:08 AM, Phil Jones wrote:

      Dave,
          Sounds fine. Will read through drafts when they come.
      A question. I'm planning to go to a meeting Henry Diaz is
     organising in
      Tahiti in early April. It's sort of Henry's swansong, so keen to
     go. It's
      also the only chance I'll get to go there !!
      I have to write an abstract - it's about paleo and ENSO. Is it OK
      to use a couple of the plots on ENSO extraction - from the global T
      records?
      Cheers
      Phil
     At 02:32 11/01/2008, you wrote:

     Hi Phil,
     A quick update:
     After mulling it over, I've decided to see if the 'step in 45' and
     volcanic results will have more impact if split into two papers.
     My feeling was that the 'step in 45' was getting lost in the
     volcanic results.
     The 'step in 45' paper will be relatively short and punchy; the
     volcano paper will be a bit longer. I've nearly got a first draft
     of both papers done. I expect to be able to send you a full first
     draft of the 'step in 45' paper sometime next week. I expect to
     have a full draft of the volcano paper ready shortly after
     submitting the 'step in 45' paper.
     Thanks again for collaborating with me on the papers...
     -Dave
     On Dec 14, 2007, at 6:37 AM, Phil Jones wrote:

      John,
         How the volcanoes stuff is done is in this paper.
      Jones, P.D., Moberg, A., Osborn, T.J. and Briffa, K.R., 2003:
     Surface climate responses to explosive volcanic eruptions seen in
     long European temperature records and mid-to-high latitude tree- ring density around the
     Northern Hemisphere, In (A. Robock and C.
     Oppenheimer, Eds.) Volcanism and the Earths Atmosphere.
     American Geophysical Union, Washington D.C. 239-254.
      Someone in the HC may have a copy or the library does. I do, but
     when we got the pdfs from
      AGU they were enormous (and still are) so can't email.
      This shows the individual eruptions for the 'raw' data rezeroed
     to the 5 year average
      of the period before the eruption.
      Dave is doing this for 10 years before, then using the
     residuals, then taking the trend out.
      I'm surprised it makes such a difference - especially with the
     trend approach. The
      residuals look fine.
      Dave has to look at the individual eruptions. There is quite a
     lot of inter-eruption
      variability.
      Cheers
      Phil
     At 16:57 13/12/2007, John Kennedy wrote:

     Dear Dave and Phil,
     In answer to Phil's question concerning other jumps in the mix,
     I think
     this is the only jump that can be identified based on the country
     metadata, where we can also be fairly sure there was a step
     change in
     the measurement method. The drop in US obs in the 1960s occurs
     when many
     more countries are contributing to the mix and so we're still
     not sure
     how this change maps onto measurement method, which is what really
     causes the biases.
     I had some difficulty understanding how Figure 3 was created -
     is the
     diagram just a straight average of the anomalies before and
     after the
     January of each volcano? Or has there been some zeroing?
     I'd be interested in seeing the individual volcanoes that go
     into the
     composite. Dave, if you have no objections to sending me the raw
     and
     residual time series, I can experiment with these things myself.
     As for the journal, I agree with Phil. David Parker also
     suggested BAMS,
     based on its having a wider readership, but JGR or J Clim sound
     fine to
     me.
     Best regards,
     John
     On Wed, 2007-12-12 at 16:31 +1100, David Thompson wrote:
     > Dear Phil and John,
     >
     >
     > Please find attached some draft text for the results section
     of the
     > paper.
     >
     >
     > At this stage, I want to make sure everyone is comfortable
     with the
     > main results section. I will send the other sections in a future
     > email.
     >
     >
     > Hopefully the text is straightforward to follow even though
     you don't
     > have a copy of the analysis details (and the figure captions
     are still
     > under construction). Please don't worry about editing - I'm
     mainly
     > interested in getting your general impressions.
     >
     >
     > (Phil: I've already iterated with John on the 'dip in 45'
     text; but
     > I'm curious to know what you think.)
     >
     >
     > One general question: I am planning on submitting the paper to J
     > Climate. Does this seem appropriate to both of you? I don't
     want the
     > 'dip in 45' text to be buried in a long paper, so I've been
     toying
     > with other venues... if you have any thoughts on the best
     journal,
     > please let me know. It will impact how the final writing evolves.
     >
     >
     > I will be on travel next week, then the holidays will be upon
     us. So
     > no rush. I'll be back working on the paper shortly after the
     New Year.
     > And my hope is to submit the paper in January.
     >
     >
     > Thanks again ...
     > Dave
     >
     --
     John Kennedy  Climate Monitoring and Research Scientist
     Met Office Hadley Centre  FitzRoy Road   Exeter   EX1 3PB
     Tel: +44 (0)1392 885105   Fax: +44 (0)1392 885681
     E-mail: [1]john.kennedy@metoffice.gov.uk   [2]http://www.metoffice.gov.uk
     Global climate data sets are available from [3]http://www.hadobs.org

     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    [4]p.jones@uea.ac.uk
     NR4 7TJ
     UK

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

     --------------------------------------------------------------------
     --------------------------------------------------------------------
     David W. J. Thompson
     [5]www.atmos.colostate.edu/~davet
     Dept of Atmospheric Science
     Colorado State University
     Fort Collins, CO 80523
     USA
     Phone: 970-491-3338
     Fax: 970-491-8449

     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    [6]p.jones@uea.ac.uk
     NR4 7TJ
     UK

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

     --------------------------------------------------------------------
     --------------------------------------------------------------------
     David W. J. Thompson
     [7]www.atmos.colostate.edu/~davet
     Dept of Atmospheric Science
     Colorado State University
     Fort Collins, CO 80523
     USA
     Phone: 970-491-3338
     Fax: 970-491-8449
     Phil,
     You are more than welcome to show any slides... hopefully the work will be well into the
     review process by then... in fact, I'd be happy to grind out similar analyses for a
     longer paleo record for you, if you like.
     As for the papers: as I mentioned I've shown the step in 45 work to a number of folks
     closely involved in the last IPCC, and all were adamant we should submit that particular
     result to Nature. So I figured I'd give that venue a shot, saving the volcano results
     for a slightly longer J Climate paper.
     Anyway, I've attached a first draft of the 'step in 45' paper (figures as pdf; text as
     pdf and doc). I figured I'd run the text by folks individually once before I send a
     draft to the group. If all goes well I hope to try and submit something in the next
     couple weeks...
     Please feel free to offer any comments that come to mind, both general and specific.
     Thanks again for collaborating with me on this... it's fun....
     -Dave
     On Jan 12, 2008, at 12:08 AM, Phil Jones wrote:

      Dave,
          Sounds fine. Will read through drafts when they come.
      A question. I'm planning to go to a meeting Henry Diaz is organising in
      Tahiti in early April. It's sort of Henry's swansong, so keen to go. It's
      also the only chance I'll get to go there !!
      I have to write an abstract - it's about paleo and ENSO. Is it OK
      to use a couple of the plots on ENSO extraction - from the global T
      records?
      Cheers
      Phil
     At 02:32 11/01/2008, you wrote:

     Hi Phil,
     A quick update:
     After mulling it over, I've decided to see if the 'step in 45' and volcanic results will
     have more impact if split into two papers. My feeling was that the 'step in 45' was
     getting lost in the volcanic results.
     The 'step in 45' paper will be relatively short and punchy; the volcano paper will be a
     bit longer. I've nearly got a first draft of both papers done. I expect to be able to
     send you a full first draft of the 'step in 45' paper sometime next week. I expect to
     have a full draft of the volcano paper ready shortly after submitting the 'step in 45'
     paper.
     Thanks again for collaborating with me on the papers...
     -Dave
     On Dec 14, 2007, at 6:37 AM, Phil Jones wrote:

      John,
         How the volcanoes stuff is done is in this paper.
      Jones, P.D., Moberg, A., Osborn, T.J. and Briffa, K.R., 2003: Surface climate responses
     to explosive volcanic eruptions seen in long European temperature records and
     mid-to-high latitude tree-ring density around the Northern Hemisphere, In (A. Robock and
     C. Oppenheimer, Eds.) Volcanism and the Earths Atmosphere. American Geophysical Union,
     Washington D.C. 239-254.
      Someone in the HC may have a copy or the library does. I do, but when we got the pdfs
     from
      AGU they were enormous (and still are) so can't email.
      This shows the individual eruptions for the 'raw' data rezeroed to the 5 year average
      of the period before the eruption.
      Dave is doing this for 10 years before, then using the residuals, then taking the trend
     out.
      I'm surprised it makes such a difference - especially with the trend approach. The
      residuals look fine.
      Dave has to look at the individual eruptions. There is quite a lot of inter-eruption
      variability.
      Cheers
      Phil
     At 16:57 13/12/2007, John Kennedy wrote:

     Dear Dave and Phil,
     In answer to Phil's question concerning other jumps in the mix, I think
     this is the only jump that can be identified based on the country
     metadata, where we can also be fairly sure there was a step change in
     the measurement method. The drop in US obs in the 1960s occurs when many
     more countries are contributing to the mix and so we're still not sure
     how this change maps onto measurement method, which is what really
     causes the biases.
     I had some difficulty understanding how Figure 3 was created - is the
     diagram just a straight average of the anomalies before and after the
     January of each volcano? Or has there been some zeroing?
     I'd be interested in seeing the individual volcanoes that go into the
     composite. Dave, if you have no objections to sending me the raw and
     residual time series, I can experiment with these things myself.
     As for the journal, I agree with Phil. David Parker also suggested BAMS,
     based on its having a wider readership, but JGR or J Clim sound fine to
     me.
     Best regards,
     John
     On Wed, 2007-12-12 at 16:31 +1100, David Thompson wrote:
     > Dear Phil and John,
     >
     >
     > Please find attached some draft text for the results section of the
     > paper.
     >
     >
     > At this stage, I want to make sure everyone is comfortable with the
     > main results section. I will send the other sections in a future
     > email.
     >
     >
     > Hopefully the text is straightforward to follow even though you don't
     > have a copy of the analysis details (and the figure captions are still
     > under construction). Please don't worry about editing - I'm mainly
     > interested in getting your general impressions.
     >
     >
     > (Phil: I've already iterated with John on the 'dip in 45' text; but
     > I'm curious to know what you think.)
     >
     >
     > One general question: I am planning on submitting the paper to J
     > Climate. Does this seem appropriate to both of you? I don't want the
     > 'dip in 45' text to be buried in a long paper, so I've been toying
     > with other venues... if you have any thoughts on the best journal,
     > please let me know. It will impact how the final writing evolves.
     >
     >
     > I will be on travel next week, then the holidays will be upon us. So
     > no rush. I'll be back working on the paper shortly after the New Year.
     > And my hope is to submit the paper in January.
     >
     >
     > Thanks again ...
     > Dave
     >
     --
     John Kennedy  Climate Monitoring and Research Scientist
     Met Office Hadley Centre  FitzRoy Road   Exeter   EX1 3PB
     Tel: +44 (0)1392 885105   Fax: +44 (0)1392 885681
     E-mail: [8]john.kennedy@metoffice.gov.uk   [9]http://www.metoffice.gov.uk
     Global climate data sets are available from [10]http://www.hadobs.org

     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    [11]p.jones@uea.ac.uk
     NR4 7TJ
     UK
     ----------------------------------------------------------------------------


     --------------------------------------------------------------------
     --------------------------------------------------------------------
     David W. J. Thompson
     [12]www.atmos.colostate.edu/~davet
     Dept of Atmospheric Science
     Colorado State University
     Fort Collins, CO 80523
     USA
     Phone: 970-491-3338
     Fax: 970-491-8449

     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    [13]p.jones@uea.ac.uk
     NR4 7TJ
     UK
     ----------------------------------------------------------------------------


     --------------------------------------------------------------------
     --------------------------------------------------------------------
     David W. J. Thompson
     [14]www.atmos.colostate.edu/~davet
     Dept of Atmospheric Science
     Colorado State University
     Fort Collins, CO 80523
     USA
     Phone: 970-491-3338
     Fax: 970-491-8449

     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    [15]p.jones@uea.ac.uk
     NR4 7TJ
     UK
     ----------------------------------------------------------------------------

     <us-ww2-merchant-logbooks.ppt><Jones_ENSO_1990.pdf>

   --------------------------------------------------------------------
   --------------------------------------------------------------------
   David W. J. Thompson
   www.atmos.colostate.edu/~davet
   Dept of Atmospheric Science
   Colorado State University
   Fort Collins, CO 80523
   USA
   Phone: 970-491-3338
   Fax: 970-491-8449

