date: Mon Nov 28 15:32:03 2005
from: Keith Briffa <k.briffa@uea.ac.uk>
subject: Re: Fw: 2005JC003188 Decision Letter
to: Sandy Tudhope <sandy.tudhope@ed.ac.uk>

   Hi Sandy
   I look forward to more drink and discussion - but in the meantime hope everything is well
   with you. My feeling as you know , is that the pre-selection of "useful" coral records does
   negate some of the power in the verification - and we kow of course about the
   uniformatarianism assumption . However this is a first attempt , and still suggests
   unprecedented warming (in the context of these quite short data) and I suppose we could
   refer to "isotopically warmer" and put the caveats in - but the work still deserves
   publication. Let's see what Rob comes up with as a revision and response - but I am
   confident we can get this (and it deserves to be) published.
   Cheers
   Keith
   At 06:01 22/11/2005, you wrote:

     Hi Keith,
     Sorry I won't get to share a beer with you in Bern.  Next time I'm back I'll come down
     to UEA ... or you could come up to Edinburgh.
     I've taken a quick look at the reviews.  Some are easy to deal with, but I imagine that
     diagnosing the causes of temperature change in HadCM3 will depend on how much time
     Philip and Simon can spare to do the work.
     Mike Evans is of course right that the empirical approach hides complexities of changing
     relationships between SST and SSS as a function of frequency, and it is hard to know how
     stable even these frequency-dependent relationships will be through time.
     Cheers,
     Sandy
     Keith Briffa wrote:

     Rob
     we will digest and talk in Bern
     Keith
     At 15:54 21/11/2005, Rob Wilson wrote:

     Greetings,
     reviews are in.
     Not accepted in its current form.
     However, I had a quick look through - I think all the comments can be addressed.
     see you in Bern
     Rob
     ----- Original Message -----
     From: <[1]mailto:jgr-oceans@agu.org>jgr-oceans@agu.org
     To: <[2]mailto:rob.wilson@ed.ac.uk>rob.wilson@ed.ac.uk
     Cc: <[3]mailto:rob.dendro@virgin.net>rob.dendro@virgin.net
     Sent: Monday, November 21, 2005 3:44 PM
     Subject: 2005JC003188 Decision Letter
     Dear Dr. Wilson:
     Thank you for submitting your manuscript "250-years of reconstructed and modeled
     tropical temperatures" [Paper #2005JC003188]. I have now received 3 reviews of your
     manuscript which I enclose for your reference.
     Based on the comments of the reviewrs and the associate editors, I findyour manuscript
     unsuitable for publication in JGR-Oceans in its present form. The reviewers do find the
     contribution valuable and ingteresting and make a large number of constructive critical
     remarks and suggestions. Additional concerns were raised by the associate editor about
     the reconstruction technique which is quite inadequately described.
     From lines 21-23 on p.6, the reconstruction is a simple averaging of normalized coral
     records available for the target period. If so, then there are no need to tune anything,
     why then there are calibration period? Further how verification statistics, like CE or
     RE can be "extended back to 1644" (p.8, line 2 and further, Table S1), if the
     instrumental dataset HadISST starts in 1870?
     While clarifying this, the authors also need to provide explicit
     formula for all statistics they use that are less standard than mean, correlation
     coefficients, and standard deviations. Even using RE and CE without definition will make
     this paper unreadable for JGR-Oceans audience, not to mention more exotic statistics
     (PM,ST,LIN).
     Another basic issue not discussed is what kind of consistency can be expected, to begin
     with, between paleoreconstructions and coupled GCM runs. The authors take the mean of
     the coupled GCM ensemble and compare it on annual basis with their paleoreconstruction.
     Earth climate has a lot of noise, on annual basis. It is not obvios why anything but
     very long-term trends (and perhaps imulse responses to volcanic eruptions) should be
     consistent. And considering that all trends correlate, it is necessary to ensure that
     the trend signals are produced for the same reasons or are due to the same mechanisms.
     A reconsideration of your manuscript will require detailed responses to each of the
     reviewers comments and appropriately highlighted revisions which are necessary and
     mandatory.
     Please submit your revised manuscript by December 31, 2005.  If you do not plan to
     submit a revision, or if you cannot do so in the time allotted, I would be grateful if
     you could let me know as soon as possible.
     Please review the Important Links to JGR Information attached below before uploading
     your revised manuscript.
     When you are ready to submit your revision, please use the link below.
     <<[4]http://jgr-oceans-submit.agu.org/cgi-bin/main.plex?el=A1D1BjvY3A2CcrO2I7A9KGXg2FZaf
     NJvsZyA2JF0mAZ>[5]http://jgr-oceans-submit.agu.org/cgi-bin/main.plex?el=A1D1BjvY3A2CcrO2
     I7A9KGXg2FZafNJvsZyA2JF0mAZ>
     I thank you for submitting your best work to Journal of Geophysical Research - Oceans.
     Sincerely,
     Raghu Murtugudde
     Editor, Journal of Geophysical Research - Oceans
     --------------------IMPORTANT PUBLICATION INFORMATION---------------------
     To ensure prompt publication:
     1. Follow file format guidelines
     2. Provide a color option
     3. Combine figure parts or provide separate captions
     4. Provide copyright permissions for reprinted figures and tables
     5. Sign and send copyright transfer agreement
     6. A formal estimate will be sent to you a few weeks after acceptance.
     For information on all of the above items, see Tools for Authors at
     <[6]http://www.agu.org/pubs/inf4aus.html>[7]http://www.agu.org/pubs/inf4aus.html. If you
     have any questions, reply
     to this e-mail.
     A manuscript tracking tool is available for you to to track the status of your article
     after acceptance:
     <[8]http://www.agu.org/cgi-bin/ms_status/ms_status.cgi>http://www.agu.org/cgi-bin/ms_sta
     tus/ms_status.cgi
     Adobe Acrobat Reader is available, free, on the internet at the following URL:
     <[9]http://www.adobe.com/prodindex/acrobat/readstep.html>[10]http://www.adobe.com/prodin
     dex/acrobat/readstep.html
     ************************************END*************************************
     Reviewer Comments
     Associate Editor(Comments):
     All reviewers consider this paper interesting and valuable, and made a large number of
     constructive critical remarks and suggestions. However, neither reviewer commented about
     the reconstruction technique, but that is precisely where I find myself completely
     confused by the manuscript.
     From lines 21-23 on p.6, the reconstruction is a simple averaging of normalized coral
     records available for the target period. If so, then there are no need to tune anything,
     why then there are calibration period? Further how verification statistics, like CE or
     RE can be "extended back to 1644" (p.8, line 2 and further, Table S1), if the
     instrumental dataset HadISST starts in 1870?
     While clarifying this, the authors also need to provide explicit
     formula for all statistics they use that are less standard than
     mean, correlation coefficients, and standard deviations. Even using RE and CE without
     definition will make this paper unreadable for JGR-Oceans audience, not to mention more
     exotic statistics (PM,ST,LIN).
     Another basic issue not discussed by the authors is what kind of consistency can be
     expected, to begin with, between paleoreconstructions and coupled GCM runs. The authors
     take the mean of the coupled GCM ensemble and compare it on annual basis  with their
     paleoreconstruction. Earth climate has a lot of noise, on annual basis. Why anything but
     very long-term trends (and perhaps imulse responses to volcanic eruptions) should be
     consistent.
     My recommendation is to request a major revision, addressing these issues and all
     comments by the reviewers. Detailed descriptiopn how comments were addressed is also
     necessary.
     Reviewer #1(Comments):
     This is a careful, detailed and useful paper that shows how information from massive
     corals can be used to develop reconstructions of average tropical temperatures and thus
     contribute to a more realistic (ie less biased to extra-tropical latitudes) history of
     past temperature variations.  It is also informative to see the comparisons with long
     model runs.  I consider the paper suitable for publication in JGR after revision.
     Specific comments include:
     · The reliability of the reconstruction declines markedly prior to 1850.  The authors
     need to emphasize even more than they do that this is a "first go" to illustrate what is
     possible.  I am concerned that the full series maybe used by people unfamiliar with the
     problems as a robust reconstruction.
     · The Isdale et al (1998) coral fluorescence series has been superseded by the series
     published by Hendy et al (2003) who also noted dating errors in the original; also
     Isdale et al (1998) noted that a long-term trend in the Havannah Island coral record may
     not be real;  I am not sure how much this would affect the outcomes of the present
     study;  these problems should, at least, be noted.
     · I would prefer the names of the coral sites be written in full in the text, rather
     than "RUS, PAL and MTA" etc
     · It would be useful to include a Table giving the names & locations of the series used
     and their record length & showing which series were used in the various nestings.  This
     will help the reader follow the "nested" procedures.
     · Why were the series averaged together to form the various reconstructions?  Why were
     multiple regressions not used?  Would this have made any difference to the outcomes?  I
     think, at least, there should be a bit more justification of the averaging approach.
     · I think the work of Evans in various papers, although reconstructing patterns, should
     be highlighted more, as several of the conclusions are similar.
     I have also made a number of comments and suggested changes on the attached manuscript.
     Additional references:
     Evans, M.N., A. Kaplan & M.A. Cane, 1998. Optimal sites for coral-based reconstruction
     of global sea surface temperature. Paleoceanograhy, 13: 502-516.
     Evans, M.N., A. Kaplan & M.A. Cane, 2000. Intercomparison of coral oxygen isotope data
     and historical sea surface temperature (SST): potential for coral-based SST field
     reconstruction. Paleoceanography, 15: 551-563
     Hendy, E.J., M.K. Gagan & J.M. Lough, 2003. Chronological control of coral records using
     luminescent lines and evidence for non-stationary ENSO teleconnections in northeast
     Australia. The Holocene 13: 187.199.
     Reviewer #2(Comments):
     Please see attached file.
     Reviewer #3(Comments):
     In the manuscript:" 250-years of reconstructed and modeled tropical temperatures" from
     Wilson, Tudhope, Brohan, Briffa, Osborn and Tett, the authors present a new
     reconstruction tropical sea surface temperature derived from coral proxies and compared
     with model output.
     Such a reconstruction is of prime importance because tropical climatic variability plays
     a key role in global climate and remains poorly known. It could be publish after some
     minor precisions.
     The aim of the paper is clearly exposed. The strategy used to respond is well explained.
     However some details need to be revised.
     1- For assessing the calibration you do not indicate precisely the SST time series
     compared with pooling coral records: do you consider local data or an average over a
     specific area? The explanation given p 7 line 10-12 is not clear for me.
     2- You select the coral data by comparing the records with local SST. We could expect
     that Sr/Ca would be the best records, because it is commonly regarded as a pure SST
     proxy. Instead Sr/Ca, which is supposed to respond to one forcing, the best correlation
     is obtained with δ18O known to depend upon two factors. What do you mean p 5 line 12,
     by "not always"? Tudhope being among the authors we could suppose a comment or a
     discussion about this curious result.
     3- The trend recorded from the early 19th century is estimated about 0.7, 1{degree
     sign}C. The start of the temperature increase occurs very early and the amplitude seems
     overestimated for tropical latitude. There are no comments in the discussion.

     --
     Professor Keith Briffa,
     Climatic Research Unit
     University of East Anglia
     Norwich, NR4 7TJ, U.K.
     Phone: +44-1603-593909
     Fax: +44-1603-507784
     [11]http://www.cru.uea.ac.uk/cru/people/briffa/

     --
     _____________________________________________________________________
     Professor Sandy Tudhope,   AND: Sandy Tudhope,
     School of GeoSciences,          Ocean Sciences Building Rm521,
     Grant Institute,                Box 355351,University of Washington,
     Edinburgh University,           Seattle, WA 98195-5351,U.S.A.
     West Mains Road,
     Edinburgh EH9 3JW,U.K.          PACKAGES AND MAIL REQUIRING SIGNATURE:
                                    Marine Sciences Building Room G,
     Tel: +44 131 650 8508           University of Washington,
          +44 131 650 4842 (sec.)   Seattle, WA 98195-7940,U.S.A.
     Fax: +44 131 668 3184
     e-mail: sandy.tudhope@ed.ac.uk  Tel:  +1 206 221 5630 (work)
                                          +1 206 612 3243 (mobile)
                                     Fax: +1 206 685 3351
                                     e-mail: <sandyt@u.washington.edu>

   --
   Professor Keith Briffa,
   Climatic Research Unit
   University of East Anglia
   Norwich, NR4 7TJ, U.K.

   Phone: +44-1603-593909
   Fax: +44-1603-507784
   [12]http://www.cru.uea.ac.uk/cru/people/briffa/

