cc: ottobli@ucar.edu date: Sun, 16 Jan 2005 12:31:02 -0500 from: David Rind subject: Re: [Wg1-ar4-ch06] Emailing: IPCC6.4.2.3PMIP.doc to: stefan@pik-potsdam.de, "W.R Peltier" Dick, The only caveat I can add to Stefan's comment is that what actually happens in coupled models does depend on the model - not all models get reduced poleward heat transport. In particular, several coupled Atmosphere-Ocean GCMs actually have strengthened NADW production, due to the removal of the freshwater input from the higher latitudes that is occurring today. People at WHOI have questioned how much poleward heat transport actually did decrease, as the isotope evidence shows indications of enhanced intermediate water production. Nevertheless, Chris Charles, comparing the isotope signature from simulations with and without a coupled mixed layer ocean model (with prescribed, current day ocean heat transports) found that using current day transports produced a high latitude isotope signature that was too light compared with the observed value, indicating that at least with that model's overall climate sensitivity,it was probably an overestimate. However, that same model produced much more appropriate tropical isotope signatures by using current day transports than when the model was run with the CLIMAP SSTs (including the poleward ocean heat transport, and allowing the model to run to equilibrium with LGM forcings, produced colder tropical temperatures). David At 6:15 PM +0100 1/16/05, Stefan Rahmstorf wrote: >Dear all, > >Dick, your section looks good, except for a few small things. I >think it's not quite right to say: > >>Although mixed layer ocean coupled models may be employed to make >>predictions of the equilibrium LGM distribution of SSTs >> >since if you don't get the ocean circulation and heat transport >right, you also get big errors in the SST distribution. In the first >(and most-cited) LGM simulation with a coupled model, Ganopolski et >al. (Nature 1998) compared the coupled version to one with a "mixed >layer ocean", and found that northern hemisphere mean LGM cooling >was enhanced by 50% when you use the coupled model. The reason is >simple and plausible: the "mixed layer ocean" uses prescribed heat >transport, which brings a lot of heat to the northern Atlantic, >preventing the expansion of sea ice. When you use a coupled model, >the warm surface currents in the North Atlantic retreat to the south >(as they should in glacial climate) with a corresponding expansion >in sea ice, and enhanced ice-albedo feedback cooling. > >Also, I think you are right that the question whether the AMOC >strengthens or weakens depends on changes in the water cycle, but I >don't think one can conclude that differences in those can be >attributed to the acceleration techniques - I think there could be >any number of other reasons for this. > >Also, I think it is worth mentioning that the strength is only one >parameter - perhaps more important to climate than the strength of >the AMOC is the latitude up to which the heat transport reaches, as >we analyse in detail in a series of papers (starting with my Nature >1994 paper which looks specifically at this in an idealised model). >There is plenty of evidence from sediment data that such latitude >changes did occur and are important. > >Cheers, Stefan > >_______________________________________________ >Wg1-ar4-ch06 mailing list >Wg1-ar4-ch06@joss.ucar.edu >http://www.joss.ucar.edu/mailman/listinfo/wg1-ar4-ch06 _______________________________________________ Wg1-ar4-ch06 mailing list Wg1-ar4-ch06@joss.ucar.edu http://www.joss.ucar.edu/mailman/listinfo/wg1-ar4-ch06