date: Thu, 01 May 2003 10:27:12 +0100
from: Tim Osborn <t.osborn@uea.ac.uk>
subject: Re: Mike Mann's review
to: Phil Jones <p.jones@uea.ac.uk>,k.briffa@uea.ac.uk

<x-flowed>
Phil & Keith,

here are my suggestions for how to deal with Mike's comments - though we 
might want to wait for other review before doing anything.

(1) Valid point but easy to deal with by adding something like... "without 
more widespread data we don't know how general these changes are, but if 
they are applicable to other parts of the world then there are two 
principal implications..."

(2) Don't get into discussion right at the beginning on which 
reconstructions are results are applicable to; instead change meaning of 
our first sentence by "...Briffa and Osborn (2002) noted that IF 
reconstructions of annual temperature trends (for parts of the last 
millennium) ARE based on predictors that are strongly influenced by summer 
conditions, THEN THEY tacitly...."  Then expand the discussion section (see 
point 8 below) to mention some particular reconstructions and perhaps that 
the potential bias is less for those that use more non-summer-sensitive 
proxies.

(3) Worth doing the composites as they're easy to do - (if the composite 
does look like the NH-mean, then we can call it yet another NH 
reconstruction!).

(4) Some discussion of forcing vs. internally generated influences on 
seasonal differences could go in the discussion section (as I suggested 
prior to submission!)

(5) Do you have these N. American series Phil?  Are they long enough to 
tell us more than is already in the NH network?

(6) Just cite Shindell et al. as an example where we say that seasonal 
differences are a good diagnostic for testing model performance, for model 
runs of the last millennium.

(7) Not sure whether we want to detrend or not, since the trend itself is 
part of the signal we're after reconstructing.  But certainly we could 
discuss that fig. 3 is based on 20th century variations and the 
seasonal-annual relationships may well differ in other centuries which 
don't have such strong anthropogenic forcing.  This is, of course, the 
whole point of the paper, that seasonal-annual relationships may not be 
stationary!

(8) In the discussion section I think we can expand things to make the 
caveat that not all proxies are summer sensitive and therefore some 
reconstructions may not be biased by seasonality changes so much (e.g., 
Mann et al., 1998).  But then go on to note that this probably doesn't 
mitigate the bias early on when the corals and the TexMex tree-rings aren't 
available (e.g., Mann et al., 1999), and therefore are results are valid to 
that - and that early bit is the crucial bit when claiming that 20th 
century temperatures are warmest in the millennium!

(9) This will all have been covered by the above.

What do you think?

Cheers

Tim

Dr Timothy J Osborn                 | phone:    +44 1603 592089
Senior Research Associate           | fax:      +44 1603 507784
Climatic Research Unit              | e-mail:   t.osborn@uea.ac.uk
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

</x-flowed>
