Friday: bright sun again in contrast to my somewhat jaded feeling: I think I can concentrate for another half day though. Up early to check out and get to posters at 8 to put mine up.
The complete disconnect between the publicly visible debate and the scientific debate is very striking. No one here is saying, will there be GW, only, how much will it be, and I didn't see anyone with a sensitivity below 1.5 oC (and the author of that noted that it was probably low). And even starting with that example is misleading: the vast bulk of the conference is about things invisible to the outside: atmospheric teleconnections (e.g. my poster...), model inprovements, etc etc etc etc. The reason is easy to understand: discussing the technical details is tricky, requires technical knowledge and access to the literature; whereas wurbling about what Richard Lindzen said n years ago requires (perhaps even demands...) very little. And of course the things that the press depts direct into the outside world are biased towards the sexy stuff.
Palaeo session, which should be good. Whaqt would be intesting would be some detailed debate about the discrepancies between the various recons. But we don't get that, sadly.
Medal-winning Goosse: ECBILT model: EMIC, 100 sims, last 2kyr. Ensemble mean v similar to MBH (later shows also sim to Moberg, if you scale Moberg down. Someone Qs this: to which Goosse replies (what the RC post on Moberg pointed out) that most of the Moberg wavelet stuff is non-dimensional; it needs to be scaled at the end, and is, to ?1850-1980?, so there is ambiguity in the scaling)). Now, we assume that the ensemble mean averages out the noise and so represents the response to the forcing. He then points out the obvious: that individual members can differ strongly from the ensemble mean. But the true climate of the earth was like one of these members: so we have to be careful about over-interpreting regional effects in the reconstructions: they may just be natural var. He then selects the member most like a reconstruction (either fit to MBH hemispheric mean, or a European only, or) and then that effectively gives you a sort of model interpolation of the proxies.
Allan: 5x5 degree MSLP reconstructions back to 1850 (www.hadobs.org). He seems to lack Antarctic data... I must mail him.
American/Pacific proxies during the "MCA" (ie, Mediaeval climate anomaly, a term he prefers since calling it the MWP is odd, because the period usually called the MWP was cold and dry over there...). Then deducing some ENSO-type EOF'd tropical pacific SST pattern and using that to force a GCM, but possibly that was stretching things a bit.
Edwards: W N America/Canada: otoh he has a MWP that is +4 oC (from something to do with C13). Whew! But when q'd on this he appears to be unconfident of the scaling, though he insists on the shape.
And then off to the posters, and mine in particular. And thats about it...