Posted on 09/08/2006 8:11:43 AM PDT by cogitator
Projected global warming is about water vapor (CO2 only provides some warming, the bulk is water vapor). But water vapor is distributed very unevenly unlike CO2 which has slow, distributed sources and sinks. Water vapor has very fast sources (evaporation) and sinks (rainfall). To figure out the overall warming, you have to accurately figure out the water vapor at as many points on the earth as possible at a fairly fast rate (e.g. their 15 minutes).
That will never happen because chaos effects can never be modeled adequately any practical possible computer. But the issue is not whether they predict the weather but whether they model it accurately. For example, do the climate models contain tropical waves and storms? Most current models do not, which makes them poor predictors of climate, never mind weather.
The prediction of tomorrow's weather is not necessary, but the prediction of variation that matches reality is. What is not needed for climate modeling is accurate timing of weather features. What is needed is accurate depiction of weather features and increasing time and space resolution as these people are doing is a step in that direction.
Thanks for the ping. Hopefully I clarified things a little. The realclimate weather prediction discussion was very useful, I'll have to post my thoughts there soon.
BTTT
"So.....weathermen can be wrong more frequently"
Either that or wrong with more decimal points.
I have heard that just saying the weather tomorrow will be the same as today is as or more acurate than many production weather models...
Doesn't take as many teraflops to calculate, either
Yes; I found this interesting because FReeper palmer and I have speculated that in a decade or less, the models will be so much improved, resolution-wise, that the dilemma of uncertainty around global warming will pretty much disappear. This article indicates that the modelers are pushing to get there.
That will never happen because chaos effects can never be modeled adequately any practical possible computer. But the issue is not whether they predict the weather but whether they model it accurately. For example, do the climate models contain tropical waves and storms? Most current models do not, which makes them poor predictors of climate, never mind weather.
Well, my issue is not a weather report, but "Global Warming" -- especially the "evidence" for a human cause. The usual "evidence" for human causation are the results of such modeling. If a model cannot predict the present based on the past, how can it be trusted to predict the future?
I'd be careful with that. A model can always be jiggered to predict the present from the past. It's predicting the future (like this year's lack of hurricanes) that proves that climate can be predicted. I haven't seen that happen yet.
Of course. But all of that is why I can't believe a word Algore says.
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.