cc: wg1-ar4-ch06@joss.ucar.edu
date: Mon, 10 Jan 2005 11:19:55 -0500
from: David Rind <drind@giss.nasa.gov>
subject: Re: [Wg1-ar4-ch06] comments to 6.3.2.1 (mainly for Keith)
to: Stefan Rahmstorf <rahmstorf@pik-potsdam.de>

   Hi Stefan,

   Thanks for your comments. As to what I believe, I think that both the forcing and the
   response are too poorly known to make any definitive comment about climate sensitivity from
   this time period, although there have been plenty of people who have tried. That's
   basically the conclusion I drew in the climate sensitivity section, 5.8. (which includes a
   listing of the various references that have interpreted climate sensitivity by choosing to
   believe that they knew either the forcing or the response).

   With respect to the question at hand, your comment that the uncertainty for the LIA does
   not bear on the question of the "Medieval Warm Period": if it is the response which is at
   issue for the LIA, then it is equally at issue for the "MWP". As suggested above, I'm
   equally skeptical of the response as of the forcing, for it suggests a very low tropical
   sensitivity relative to that in the extratropics. There is now ample evidence in
   paleoclimate that what has started out as a view of small tropical response (LGM, Tertiary
   climates in general) is now being seen more and more as an underestimate of the tropical
   response. Granted, these are equilibrium climates, and it is possible the extratropical
   responses seen over the last 1000 years have more to do with atmospheric wave propagation
   changes (although what drove them?) than radiative forcing. Nevertheless, the scarcity of
   tropical data, and the questions associated with attempts to reconstruct them from
   extratropical variability, leave room for a lot of doubt on this score.

   David

   At 5:00 PM +0100 1/10/05, Stefan Rahmstorf wrote:

     Hi David,
     well, yes and no.

     Well, yes and no. If the mismatch between suggested forcing, model sensitivity, and
     suggested response for the LIA suggests the forcing is overestimated (in particular the
     solar forcing), then it makes an earlier warm period less likely, with little
     implication for future warming.

     If it suggests climate sensitivity is really much lower, then it says nothing about the
     earlier warm period (could still have been driven by solar forcing), but suggests future
     warming is overestimated.

     Yes, IF. But it doesn't suggest that. Or do you think any of those results could
     possibly be interpreted as meaning that the climate sensitivity lies outside the 1.5-4.5
     range? My argument is: the forcing for the past millennium is too uncertain for
     estimating climate sensitivity; this is why it has not been used for that, and this is
     why none of these studies affect our present estimates of climate sensitivity, which are
     based on other things rather than those proxy data. Many people don't know this; I even
     know one paleoclimatologist in Germany who thinks that our projections for the future
     are based on the proxy data for the last millennium, and if he doesn't believe in the
     Mann-curve it means he also doesn't have to believe in future warming.
     A case in point is the GKSS paper which appeared in Science, which produces far greater
     variability than all the proxy data, with a model that has a perfectly medium climate
     sensitivity somewhere near 3 C, while with nearly the same climate sensitivity, you can
     also be very close to the low-variability extreme, i.e. the Mann curve (e.g. the Bauer
     et al study). All just within forcing uncertainty. This is why all of this does not
     constrain or affect climate sensitivity in any way. (For this reason I think that the
     Osborn&Briffa statement in the Science commentary about the GKSS study is incorrect in
     stating that the larger variability implied if the Storch result is correct would imply
     a greater climate sensitivity - it just doesn't.)
     Stefan

     If however it implies the reconstructions are underestimating past climate changes, then
     it suggests the earlier warm period may well have been warmer than indicated (driven by
     variability, if nothing else) while suggesting future climate changes will be large.

     This is the essence of the problem.
     David

     --
     Stefan Rahmstorf
     www.ozean-klima.de
     www.realclimate.org

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