date: Wed Jun 18 09:39:51 2008
from: Keith Briffa <k.briffa@uea.ac.uk>
subject: Re: Review of your Holocene submission
to: "Nathan Malcomb" <nmalcomb@gmail.com>

   Nathan
   this is a difficult one - because I suspect you may be able to improve the standardisation
   to some extent but not by generally available approaches - my best suggestion would be that
   you might simply note the potential lack of lower frequency info and still present the
   results in a sense - as high-pass interpreted data. It turns out the main referee is citing
   information (it is not me or anyone here by the way) that is not that generally appreciated
   by many yet as the "recent bias" effect is in a paper (by Tom and me here) that this
   referee had access to . The chronologies really need standardising with a "signal-free"
   approach. HOWEVER, THERE IS CLEARLY NO SIMPLE FIX HERE. It is not impossible that if you
   submitted this elsewhere you might not meet the same objections  - but would that be
   satisfactory in your own mind anyway? Perhaps you could try a revision and then reconsider?
   sorry that I can not be more positive.
   best wishes
   Keith
   At 16:27 17/06/2008, Nathan Malcomb wrote:

     Dear Dr. Briffa,
     Thank you for sendng the first review of the paper.  The reviewer noted several
     undeniable flaws with our mass balance reconstructions, the most serious of which being
     the short calibration periods used in modeling.  This problem stems from the heavy use
     of Fritz Schweingruber's 1983 chronologies.  While many of the review's other
     suggestions can be addressed, it would not be possible for us to update these
     chronologies or omit them entirely.  Given these limitations, is this paper salvageable
     for re-submittall to the Holocene?  Can you recommend another journal or letter ?  Thank
     you.
     Nathan Malcomb
     On Fri, May 16, 2008 at 7:23 AM, Keith Briffa <[1]k.briffa@uea.ac.uk> wrote:

          Dear Dr Malcomb ( and Greg)
          I am really sorry for the delay in the review of this paper. The truth is that it is
          not my fault though. I received one review some weeks ago , but I have been trying
          for some time to get the other one. I have sent reminders but having received your
          latest request for news , via John Matthews , I sent another urgent request which
          has succeeded only in the reviewer informing me that he can not get the review done
          and asking me to find another reviewer. I am copying the initial review that I have
          below - but you will see that it is not favorable . You may care to consider the
          points and perhaps respond as to your likely course of action . In its present form
          and on the basis of this review , we would be unable to accept the paper for
          publication. I can if you wish send it out to another reviewer (perhaps you may even
          like to suggest one?). However we would be bound to respect the original reviewer's
          opinion and would require significant reworking of the paper even so. What course of
          action would you consider appropriate at this time ?
          We have now moved to an electronic submission system and you may wish to consider
          resubmitting a revised manuscript through that route. I apologise again that you
          have had to wait for this response, but I have to stress that it is getting
          increasingly difficult to find reviewers who are prepared to dedicate the time
          needed to undertake careful reviews. I look forward to hearing response best wishes
          Keith
          Paper by Malcomb / Wiles
          Tree-ring based mass balance estimates along the Northwestern Cordillera of North
          America
          Conclusion:
          The paper presents a tree-ring based reconstruction of the glacier mass balance
          evolution in Northwestern North America for the last few centuries. However, the
          dendrochronological methods used are inadequate for centennial reconstructions.
          Moreover, the calibration periods are shortened due to optimize the calibration
          results. This ends up in very short calibration periods. There are no statistical
          tests for the stability of the glacier tree ring relationships on which the
          calibration is based for the reconstruction period. Even some important references
          are not used.
          The paper should not be published with these major problems.
          Some details:
          Line 30: glacier-climate relation: statement that glacier size changes only can
          display at least multi-decadal fluctuations is too general directly observed length
          change records of some glaciers even document inter-annual reactions
          Line 47 and following: even European references on tree-ring based glacier
          mass-balance reconstructions should be used e.g. Raper et al., who discuss the
          important point of changing meteorological conditions and their influence on the
          reconstruction results
          Line 15: the calibration periods are extremely short shortened to optimize the
          calibration results. This in combination with multiple predictors selected out of a
          set of data due to good statistical results leads to high correlations for the
          calibration periods. However, the authors mentions that this calibration is not
          stable for recent years. Additionally there is no test of these reconstructions for
          the past. The authors could at least test the stability of the relationships between
          the tree-ring chronologies used for the reconstructions for the reconstruction
          period (e.g. running window correlations).
          Line 109: It is well known that individual standardisation of tree-ring series does
          not allow reconstructions for frequency bands greater than the mean segment length
          of the single series. Additionally this standardization produces even end effects
          that might be at least partly responsible for the a divergence problem in recent
          years. Because the study aims on a multi-century long mass balance reconstruction
          other standardisation methods should be used. The missing long term trend (e.g. in
          comparison the Northern Hemisphere temperatures as displayed in Fig. 3) of the mass
          balances reconstructions are probably a direct standardisation artefact
          Line 116 and following: the paper limits the time period for the calibration of the
          tree-ring mass-balance relationship by a simple statement on "model instability" in
          the recent years. It is too simple to say, that tree rings lost their climate
          sensitivity (high frequency or low frequency band ?) in recent years and therefore
          to exclude the last approx. 20 years of observation.
          General:
          -         instead of "instrumental mass balances" the term "glaciological mass
          balances" should be used, because glacier mass balances are not measured by using an
          instrument
          -         additionally and to avoid confusions, tree-ring based mass balances data
          are "reconstructed mass balances"
          -         the paper compares the mass balance reconstructions with the individual
          and regional glacier history (e.g. fig. 3) therefore the known history of the
          glaciers used should be shortly mentioned: know LIA glacier maximum, history since
          this max, distance of retreat
          -         The reconstructed mass balances are not really compared with local and
          regional glacier history. Some reconstructions show an overall positive mass balance
          trend for the last c. 200 years is this in accordance with "real" glacier evolution?
          Figures and tables:
          -         Figure 2 and 3: instead of characters the names of the glaciers itself
          should be displayed in the figures, figure 3: what kind of data is displayed should
          be indicated
          -         Table 1: more information on the tree-ring sites and chronologies used is
          necessary: number of trees in the chronology, mean segment length, inter-correlation
          of the series, altitude of site, distance of the site to local tree line
          --
          Professor Keith Briffa,
          Climatic Research Unit
          University of East Anglia
          Norwich, NR4 7TJ, U.K.
          Phone: +44-1603-593909
          Fax: +44-1603-507784
          [2]http://www.cru.uea.ac.uk/cru/people/briffa/

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

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

