Welcome to European Tribune. It's gone a bit quiet around here these days, but it's still going.
I haven't met him in person, but I've read a paper of his and a couple of things he's written elsewhere. And the most polite thing I can say about them is that they are underwhelming.

That doesn't prevent him from being a nice guy otherwise, or even competent in his own branch of physics, but in any discussion of global warming, he's a major red flag.

I don't have any thoughts about stochastic vs. deterministic models in the current context, because I haven't read up sufficiently on precisely how those terms are used in the context of climate modelling. But assuming that they mean roughly what they usually mean in statistical physics, I would suspect that any model of geoclimate would have elements that could be called deterministic and elements that could be called stochastic (and elements that could be called deterministic but chaotic).

Some parts of GCM can be constructed ab initio from elements that I would call deterministic (such as the zeroth-order effect of a solar and GHG forcing), some parts will use approximations that can be either deterministic or what I would call stochastic (cloud formation and precipitation will probably fall into the latter category), parts will be mainly empirical (some feedback mechanisms) and parts will be WAGs.

Overall, the current models are deterministic and what I would call quasichaotic. And the results they produce are pretty convincing.

- Jake

Friends come and go. Enemies accumulate.

by JakeS (JangoSierra 'at' gmail 'dot' com) on Thu Jul 31st, 2008 at 10:22:19 AM EST
[ Parent ]
And the results they produce are pretty convincing.

Do you mean the precise results or the overall picture?
by Colman (colman at eurotrib.com) on Thu Jul 31st, 2008 at 10:28:48 AM EST
[ Parent ]
I mean that if you calibrate four of the best models for the 19th cent. and then run them for the 20th cent. with all appropriate climate inputs (solar forcing, volcanoes, GHG emissions and so on and so forth) and you extract the mean temperature for each year, you can't tell them from the observed ground-station mean temperature unless you know which years the el-Ninõ happened in real history (because it's one of those stochastic and/or quasichaotic phenomena that current models can't predict to the year). They cover the same range, have the same magnitude of chaotic (or stochastic) fluctuations, the same trend and pretty much the same 30-year running averages.

Eigil Kaas of the Niels Bohr Institute in Copenhagen has a slideshow where he does this, but I can't seem to find it on his homepage.

Now, we can argue about whether global mean temperature is a meaningful checksum to use, of course, but that happens to be the most often used.

- Jake

Friends come and go. Enemies accumulate.

by JakeS (JangoSierra 'at' gmail 'dot' com) on Thu Jul 31st, 2008 at 10:51:56 AM EST
[ Parent ]


Top Diaries

Latest Data Omicron Variant

by Oui - Nov 30

Sweden almost has a new PM

by fjallstrom - Nov 26

Ten Days That Redefined PM Johnson

by Oui - Nov 14

Salman's Saudi Tiger Squad

by Oui - Nov 7

Occasional Series