cc: wg1-ar4-ch06@joss.ucar.edu
date: Mon, 19 Dec 2005 01:47:16 +0100
from: Stefan Rahmstorf <rahmstorf@ozean-klima.de>
subject: [Wg1-ar4-ch06]  the sports-implement-that-must-not-be-mentioned
to: Jonathan Overpeck <jto@u.arizona.edu>

<x-flowed>
Dear chapter 6 colleagues,

I'd like to come back to a point Susan made in the plenary: the 
millennium reconstruction curves should be normalised for the 20th 
Century (or parts thereof, e.g. mid 20th Century). I strongly agree with 
Susan here, for the reasons below:
- normal people (policy makers) think of climate change relative to 
recent experience, not in some abstract absolute terms
- those data do in fact not give absolute temperatures, but those 
relative to the recent instrumental period - they are in fact normalised 
for this (a very valid point that Gabi made in the plenary)
- the main conclusion we draw is that recent warmth is unprecedented - 
if you want to see this, you must show the curves relative to the recent 
times
- it makes no sense if the proxy curves diverge strongly in the 20th 
Century, since this is only period where we really know what the climate 
was like.

These arguments are slightly different for the model results, hence in 
the current draft, the proxy data are normalised for recent climate, and 
the model results for something else. For the models maybe we need to 
debate this - the main reason for doing it differently for the models 
seems to be to show the correction for the Von Storch result. When the 
AR4 comes out, this will be in the published literature, so no need to 
make our whole graph look different just to cater for one bad model run. 
I think the models would be better shown normalised the same way as the 
proxy data; this would also help avoid possible confusion. But I very 
strongly feel that the simplified joint plot shown in the plenary, with 
the envelopes of both models and proxies, definitely must be normalised 
for the 20th Century.

Cheers, Stefan

_______________________________________________
Wg1-ar4-ch06 mailing list
Wg1-ar4-ch06@joss.ucar.edu
http://www.joss.ucar.edu/mailman/listinfo/wg1-ar4-ch06
</x-flowed>
