Thursday, December 16, 2010

More AGU

OK, here are some highlights and lowlights from the first two days...

On Monday lunchtime, John Holdren gave a party political broadcast for the Democrat Party. "Never before has a President spoken so much..." I was waiting for "...to so little effect" but in fact the continuation was "...about Science and Technology". To be fair, he did acknowledge that talk wasn't actually that useful in itself, and also gave a laundry list of how everything is getting better under Obama. Which may even be true, I don't presume to judge. Later that day Julia Slingo in her "Frontiers of Geophysics" lecture gave some insight into the direction the UKMO was taking towards hazard prediction, mostly on weather-related time scales of course.

Of course a fair proportion of the posters are pretty pointless, due to people having to submit one to justify the trip. Can't blame them for that, but it does make it hard to find the handful of interesting ones. Steve Schwartz was trying to pretend that all the CO2 will simply vanish into the ocean in a few decades if we stop emitting, but there was someone already there debunking him as I passed by. The talks are a mixed bag and seem a pretty incestuous affair. Particular raspberries are due to the organisers responsible for offering Gardar Johannesson an invited talk to present LLNL's vague plans and tentative initial steps towards investigating parametric uncertainty in a single GCM - there was little hint towards the uncomfortable fact that they are about 5 years behind several other groups around the world in this respect - in fact, as well as the Hadley Centre and CPDN work with HadCM3, and us in Japan with MIROC, there are even two groups already doing this sort of thing with the very same model in the USA (Ben Sanderson at NCAR, Charles Jackson at U. Texas)! I'm sure it is no coincidence that one of the organisers of that session was from the same lab. On top of that, Nychka and Sain spun out a single piece of collaborative work into two invited talks in adjacent co-sponsored sessions. While I was listening to a sequence of talks that leant heavily on the truth-centred paradigm for the interpretation of climate models (ugh), Jules went to a session pointing toward policy relevance of climate science which seemed to be largely based on Roe and Baker's pdf for climate sensitivity (sigh). Well, I suppose it gives us plenty to do over the next few years, but the lack of progress or penetration of new ideas is a bit depressing to witness. On a more positive note, Steve Easterbrook gave an interesting overview of his anthropological investigations into the software engineering practices of several climate research labs (see his blog for more). I suspect RIGC may be a little less well organised than the places he visited, but hopefully not too far behind. So far, I think the dinner at Slanted Door has been the highlight (especially the rack of lamb), but there is more to come...

5 comments:

Anonymous said...

I saw the Steve Schwartz poster and considered stopping... and then decided I had better things to do with my time. But I am still curious - what is it that convinces some people that their simple model is superior to the work of a whole field worth of experts? (this seems apropos: http://xkcd.com/793/)

The carbon cycle seems to be a favorite target where people get really, really off base because they apply their intuition about standard decay functions and ignore the physical reality underlying the process. Steve Schwartz, Essenhigh, Mark Jacobson...

-M

Magnus Westerstrand said...

:( Could it have have any thing to do with scientists living in small groups just focusing on results and not thinking about how to apply it to the "real" world? (TC and so on)

James Annan said...

I think it's just a facet of the natural human condition - in the case of scientists and academics, the expression "gone emeritus" seems to fit, but probably many of us know older people (not all, but some) who have become increasingly confident and dogmatic in their opinions and virtually impossible to discuss things with. With Schwartz in particular, his work has some pedigree with his previous wrong attempt at working out the time scale of the physical system. It seems that he hasn't learnt from the experience...

skanky said...
This comment has been removed by the author.
skanky said...

Professor Slingo's talk:

http://www.agu.org/meetings/fm10/lectures/lecture_videos/U15A.shtml