Hello fp;ls,
Chris Hope replied to Steven B.:
> At 07:29 AM 7/24/01 -0600, you wrote:
> >Here is an article about NASA's new super computer. It suggest that
issues
> >about climate change will be resolved through the use of models. I post
this
> >to raise the EE issue of the use of models in environmental science. I've
> >always been dubious about the use of models by ecologists in place of
actual
> >evidence. I think that the use of models in ecology is similar to the use
of
> >correlation in economics. Both are substitutes for "real" science in the
> >sense that they rely on a certain type of faith. Anyway, things have been
> >quite, so I thought I'd stir the pot.
> >
>
>http://www.enn.com/news/enn-stories/2001/07/07232001/supercomputer_44384.as
p
> >?site=email
Chris:
>
> If you read the scientists rather than the reporter, they are much more
> sober in their assessments. A factor of ten increase in computing power
> will undoubtedly help, but the uncertainty in the basic science will still
> make exact climate prediction impossible.
Ray here:
Imho, there is no such thing as "exact" predictions of climate or anything
else. All analysis depends on the validity of the fundamental data
including relationships and uncertainty, etc necessitate consideration of
probabilites.
Chris:
>
> My hope is that the scientists will use some of the increased power to run
> 'ensemble' forecasts with slightly different initial conditions, as they
do
> in weather forecasting, to see which predictions are robust, and which are
> fragile.
Ray here:
I agree. It seems to me that the *use* of computers is not fundamentally
different from the *use* of the old hand-cranked Freiden calculators. Their
significance is that the power of the computer tool permits us to manage
more complexity, manage a much larger set of data, look at more
alternatives, test for sensitivities, etc. in a phenominally more
informative manner and with great facility.
But it is still "GOGO" = garbage in, garbage out.
Ray
>
> Chris
>
>
>
> Chris Hope, Judge Institute of Management Studies,
> University of Cambridge, Cambridge CB2 1AG, UK.
> Voice: +44 1223 338194. Fax: +44 1223 339701
> e-mail: [log in to unmask]
|