date: Thu Apr 24 15:42:36 2003
from: Keith Briffa <k.briffaatXYZxyz.ac.uk>
subject: Re: Can you provide a brief comment if not full review?
to: Keith Alverson <keith.alversonatXYZxyzes.unibe.ch>
Sorry , I do have a recollection of quickly skimming this paper before and rapidly dumping
it on a pile labelled "probably not worth the effort of giving a thorough review".
Basically , I think the paper has little if anything to recommend publication. It adds
little , except confusion , to the science.
The main problem is a lack of focus and clear experimental design. The reducing sensitivity
of tree growth to temperature forcing is unarguably a difficult and complex problem because
the phenomenon is largely dependent on what trees/areas/variables/processing methods are
used to make the comparison with "temperature". The temperature variable is itself a
potentially ill-defined ( compromise in effect) choice . Briffa et al have published a
specific manifestation of this phenomenon - based on one highly selective set of data , for
which they describe the local and regional associations with one optimum "summer average "
for density data and another for ring width. The present paper , by not adequately defining
the rules upon which they based their regionalisation of the tree-ring data or the basis
for specifying a particular temperature season(s) to be used in the comparisons , serves to
confuse a number of potential factors that contribute to the possible time-dependence in
the correlations they describe. The overriding criticism is that they examine the regional
tree ring series correlations with only the one ( Northern Hemisphere annual mean )
temperature series. It is therefore not possible to know to what extent the results
represent a shift in the association between that Hemisphere mean series and the regional
climates of the areas represented by their regional chronologies. There are other problems
(such as time-dependent changes in the structure of these chronologies, non-comparability
in the simple correlations because of the different lengths of period - it would be better
to calculate significance levels over a moving window compatible in lengh to the short
recent period(s) and test whether the reduced values are significant in the context of the
longer records uncertainty estimates ) . Work has been published that documents how
temperatures averaged over different areas of the Hemisphere correlate with the Northern
Hemisphere mean and the associations are subtlety time dependent and time-scale dependent
and seasonally dependent . The association between Northern Sweden and The Hemisphere in
summer is especially weak and one would not presumably base a reconstruction of the latter
on only one Tornetresk series anyway as shown in 4. The paper does not offer much because
it needs to be very much reworked after considerable work - and the conclusions are pretty
much hand waving anyway.
I do not know whether this is sufficient but it does give my "overall" opinion.
At 02:24 PM 4/23/03 +0200, you wrote:
I hope you have recovered from your back surgery well. I am writing with
regard to the sonenchkin paper submitted for a special issue of paleo3 that
Olga Solomina and I are editing that I sent you asking for a review some
The timeline for the issue is rapidly drawing to a close so I absolutely
must send this back to the author with his reviews before the end of April.
The paper deals with the recent decoupling of temperature and tree ring
indicese in high latitude eurasia that you have pointed out in previous
publications, so I feel it is rather key to have your thoughts. The other
reviewer has provided a very thorough set of suggestions, so I don't really
need a thorough review, but I would very much appreciate it if you could
have a quick read of the paper and let me know your general thoughts, in
particular if there are any glaring errors in it! If you cannot find the
time, please also let me know so I can find another option.
Thanks in advance.
on 02/18/2003 11:56 AM, Keith Briffa,cru (Climatic Research Unit) at
> Unfortunately, I am forced to be away from the office for some weeks at least
> during February and early March, having surgery on my back and undertaking a
> period of recuperation. If you are contacting me regarding outstanding review
> requests or queries regarding the status of manuscripts submitted to The
> Holocene, please note that I am dealing with these during my absence and I
> will contact you directly. If your request is of a different nature, I will
> try to respond in due course, but you may prefer to try one of my colleagues
> (see below).
> Questions with regard to our current research proposals should be directed to:
> Tim Osborn (firstname.lastname@example.org) - SOAP or RAPID;
> Phil Jones (email@example.com) - HOLSMEER, ALPIMP.
> Keith Briffa
PAGES International Project Office
B�renplatz 2, 3011 Bern, Switzerland
Tel (office): +41 31 312 31 33
Tel (direct): +41 31 312 31 54
Mobile: (+41) 079 641 9220
Fax: +41 31 312 31 68
Professor Keith Briffa,
Climatic Research Unit
University of East Anglia
Norwich, NR4 7TJ, U.K.