Wednesday, February 15, 2017

What Climate Models Have Gotten Right

An interesting video by Peter Sinclair:


David in Cal said...

How many different climate models are there? The IPCC looks at dozens. These models change at each new IPCC study. Various IPCC studies may have included over 100 diffrerent climate models. In addition, there are climate models that are not a part of the IPCC.

It's difficult to say how accurate "the climate models" have been, when there are so many different ones.


DocRichard46 said...

David, they use a wide range of models, and do repeated runs, over and over. And as the film shows, the models make predictions, and the observations confirm those predictions. Not 100%, because nothing is perfect, but they are good enough to tell us that we need to stop adding GHG load in the atmosphere.

David in Cal said...

Doc, yes there's no doubt that the planet is warming and GHGs are contributing to the warming.

David Appell said...

David, I don't know any more about the video than what you saw. But I would expect that every global climate model reproduces what's described in the video. Certainly the dozen or so big ones, like are run by GISS, the Hadley Centre, etc.

David Appell said...

By the way, I don't think it's right to take all climate models equally. Perhaps the 10 or so big global models can be compared, but many other models are trying different things -- different representations of physical processes -- different parametrizations, etc -- that they aren't designed to be compared.

When I've talked to modelers I've been struck by how little attention they pay to the projected warming by 2100 or up until today or the things that policy makers and the public want to know.

Instead they see models as experiments -- what happens if a particular important parametrization is changed, or if some input parameter is changed, or if the initial conditions are varied? How does it perform over the Quaternary? What if the representation of clouds is changed? Etc.

Different models do their physics in different ways. Comparing them all to one another isn't necessarily useful.

DocRichard46 said...

What impresses me is that some of the most complex models will spontaneously generate their own varying ocean currents, although they are not in synch with the real currents.