To: Tim Osborn <firstname.lastname@example.org>, Scott Rutherford <email@example.com>
Subject: Re: RegEM manuscript
Date: Mon, 06 Jan 2003 09:13:24 -0500
Cc: firstname.lastname@example.org,Phil Jones <email@example.com>, Ray Bradley <firstname.lastname@example.org>,email@example.com, firstname.lastname@example.org
Thanks very much Tim,
Your comments are extremely helpful.
I'm open to eliminating the comparison w/ Esper et al --but lets see if there is a
consensus of the group as to what to do here. We're anxiously awaiting comments from the
p.s. Scott can be reached at either U.Va or U.RI email equally well (I believe the former
is forwarded to the latter)..
At 12:16 PM 1/6/2003 +0000, Tim Osborn wrote:
Dear Scott and Mike,
Over the Christmas break I (finally!) had time to read the RegEM manuscript in detail.
Phil had already read and annotated a copy - so I've added my annotations to that and
will mail it to you today. Mike asked for comments to go to Scott, so please tell me
which address I should use (Rhode Island or Virginia?).
I spoke to Keith and he has partly read it too, and will provide separate comments soon.
Overall, I think the paper is a very nice piece of work and I'm pleased to be involved
with it. The results regarding robustness with respect to proxy data, method, region
and season are definitely good to publish.
Among the many comments annotated on the manuscript, a few are repeated here so that all
authors may respond if they wish:
(1) Given the overwhelming number of values in the Tables, I suggest halving them by
dropping all the CE values (keeping just RE values). As the paper points out, getting
the verification period mean right is rewarded by RE but not by CE. Since we are
interested in changes in the mean, I don't think that's a problem. CE is fine in
addition, but dropping it would provide benefits of reducing manuscript size - and
especially the size of the tables.
(2) The "mixed-hybrid" approach sounds dubious to me - more justification/explanation of
why it is needed (and hence why it captures more variance than the simpler splitting
into high- and low-frequency components method).
(3) It is not clear to me that the paragraph and figure on the comparison with Esper et
al. are either correct or necessary. They also are problematic because it would appear
that we (Briffa & Osborn) were contradicting our earlier paper when in fact we aren't.
The paper is already long and to remove these parts would therefore be helpful anyway.
The comparison with Esper et al. is important - but much better dealt with in a separate
paper where it could be developed in more detail and with more room to explain the
approach and its implications.
(4) I still hope to write up some more detailed comparisons of the reconstructions using
just the MXD data but different methods and will let Mike/Scott know my plans on this
Happy new year to you all.
Dr Timothy J Osborn | phone: +44 1603 592089
Senior Research Associate | fax: +44 1603 507784
Climatic Research Unit | e-mail: email@example.com
School of Environmental Sciences | web-site:
University of East Anglia __________| http://www.cru.uea.ac.uk/~timo/
Norwich NR4 7TJ | sunclock:
UK | http://www.cru.uea.ac.uk/~timo/sunclock.htm
Professor Michael E. Mann
Department of Environmental Sciences, Clark Hall
University of Virginia
Charlottesville, VA 22903
e-mail: firstname.lastname@example.org Phone: (434) 924-7770 FAX: (434) 982-2137