Posted on 09/08/2006 8:11:43 AM PDT by cogitator
U.S. scientists are predicting we will soon enjoy precision climate modeling, offering detail and scale only imagined a few years ago. John Drake and David Erickson at the Oak Ridge National Laboratory's computer science and mathematics division say it's possible for scientists to create meteorological models that take into account such things as the complete carbon cycle, terrestrial biology, El Ninos and hundreds of other factors.
The goal is to provide what scientists call a fully integrated Earth system model that can be simulated every 15 minutes for centuries.
"Before, we had to make compromises that ultimately limited the resolution and scope of our models and subsequent predictions," said Drake. "Now, using what we have learned and with computing power exceeding 50 teraflops, we can make our models far more sophisticated."
The upgraded model will incorporate the most recent atmospheric, oceanic and glacial ice data as well as improved chemistry, biology and physics, Drake said.
The research also involves scientists at the National Center for Atmospheric Research, NASA, Duke University, Georgia Tech and various national labs to develop a climate end station.
Recent work by Erickson and colleagues appears in the journal Geophysical Research Letters.
I thought you would find this particularly interesting.
** ping **
No, that would be the engineers. (smirk)
I recall Steven Schneider and others of his ilk at NCAR say the same thing 30 years ago when they got their brand new CRAY.
Models are only as good as their algorithms and their data. Both continue to get better, but are still woefully insufficient, in my view, to support predictive analyses.
Models and algorithms are always improving and getting much better at proving manmade global warming. In "science", it always helps to have a goal in mind, if that is what you want to prove.
So.....weathermen can be wrong more frequently?
Wonder if this will support the global warming theory.
Actually, I'm a mechanical engineer and I took a bunch of fluid dynamics and what I know is that there are classes of equations that don't result in deterministic solutions. That is, if you tip the initial conditions ever so slightly the result changes drastically. They may have 50 terraflops, and I'm sure its worth a lot of insight, but the reality is that there will still be a huge "political" element to these simulations.
So will it rain tomorrow or not?
Exactly. These "Models" are used for nothing but propaganda. Can they start one 100 years ago and predict the present accurately? I predict not.
That would be a good test. Probably they could go back only 20 or 30 years when they have better data to start from and find out how well it works. They still probably won't get tomorrow's weather right anyway.
Yes.
Kristina Abernathy - call your agent.
Don't hold your breath. There is a LONG way to go before something like this comes to fruition.
LOL!
I thought you'd get a kick out of that.
You were right.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.