cc: wg1-ar4-ch06@joss.ucar.edu date: Wed, 12 Jan 2005 16:04:23 +0100 from: Stefan Rahmstorf subject: Re: [Wg1-ar4-ch06] Bullet points for the executive summary for to: David Rind Dear David, now we're talking science; that's good. I appreciate that for atmospheric dynamics, the effects you are talking about are the first-order effects. How well any attempts to parameterise these effects work has to be judged by the results, e.g., the comparison of a parameterisation with data and with results from more comprehensive models. How well, for example, a model like CLIMBER2 performs in this respect is documented in the published literature. The results for large-scale features (e.g., zonal averages and the like), which we are aiming for, lie generally within the range of GCM simulations. And this is not the result (as is a common prejudice amongst the GCM friends) of some illegitimate "tuning" practise - this is all discussed in detail in the papers. If a scientist like Peter Stone didn't succeed, I have all the more respect for my Russian colleagues who were more successful in tackling this problem (it also took them most of their career). The next question then is to what accuracy you need to resolve a specific process to study a particular problem. If you want to study how a tsunami spreads, you can obviously forget about all the atmospheric dynamics, as it plays no role. This is not a climate problem of course, but I would argue that there are also many climate problems in which atmospheric dynamics is not the prime mechanism, and for which you don't need to resolve atmospheric dynamics except that you need to get some very basic feedbacks and responses (e.g., how Hadley cells shift and change in strength), which a parameterised version can give you to an acceptable level of accuracy. If that was not the case, EMICs would be useless, of course. Cheers, Stefan -- 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