GCMs-misuse!

rjw9y rjw9y at envsci.evsc.virginia.edu.
Sat Dec 4 17:23:45 EST 1993


i have been reviewing the scientific jounals for cliamte change
articles and reviewing them for a fellowship over the last 2 years.  My
concerns are that GCM output is often used inappropriately.  It is a
real problem when one uses GCM output to estimate the "potential impact
of global warming" on any system especially when a)precipitation
simulations are used b) results are interpolated to a finer spatial
grid scale.  Most probably know why those are problems however if not
feel free to respond.  

This observation is not what prompted me to write.  Instead, I recently
had the oportunity to listen to S. Schneider talk about GCMs.  He said
that because GCMs were parameterized after vast amounts of CO2 were
emitted so they cant be expected to correctly simulate climate to the
degree which would satisfy critics like myself.  An article sent to
Nature and Science recently demonstrating the inaccuracy of GCMs to
capture the variation in global temperature (it involved Eigenvector
analysis) because one or more reviewers commented that it was unfair to
evaluate GCMs in that manner.  

I dont get it.  There are a myriad of articles (especially in Dr.
Schneider's jounal-Climatic change and even in Nature) were authors use
temp and pcp output for 2xCO2 runs and interpolate to a smaller scale
in order to relate, for example, how fish communities in lakes of the
Canadian Maritime Provinces will be affected in "the future".

If modellers dont want the output compared to observational data how
come they dont govern the use of their model output?  Moreover, how in
the world do articles like these get past review?  

I'm disillusioned by the obvious inconsistancies displayed here.  Where
has scientific objectivity gone?  beleive it or not it seems that Steve
Schneider's mentality of not bothering to seperate beliefs and bias
from scientific work (not an exact quote, i admit but not far) has
become widespread.  

I feel the scientific literature has revealed that while GCMs have
their usefulness and are "state of the art", they are woefully
inadequate to predict either qualitative or quantitative climate change
estimates yet.
This is especially true in light of the extreme climate variations
exhibited in proxy climate data which has been published over the last
5 years or so.

Bob Wood
e-mail rjw9y at virginia.edu----------- i am a rational person interested
in rational responces not a political discussion, please. 



More information about the Ag-forst mailing list