From the 2001 IPCC Third Assessment Report: "The climate system is a coupled non-linear chaotic system, and therefore the long-term prediction of future climate states is not possible." All weather models fail after a few days. Claim of accuracy at 100 years is laughable!
Note the immediately proceeding line from the report:
Rather the focus must be upon the prediction of the probability distribution of the system's future possible states by the generation of ensembles of model solutions.
I need some help from a math whiz here. I'm not a slouch, but this is making me feel a bit retarded. If you cannot predict a future state of a system based on the current state (indeed, the authors say it is 'impossible'), then how do you generate probability distributions for future states? The idea is we have a deterministic system exhibiting random behavior (epistemic only?), meaning some outcome measure is random with respect to the starting conditions.
How do you develop probability distributions if this is true? By their nature, those distributions give you the objective likelihood for a particular state to obtain relative to other possible states. It seems that the only possibility is to make the system non-chaotic and something subject to experimental research, as you. . .cause it not to represent the climate any longer. And I guess this would also involve a huge amount of simplification, like leaving out entire inputs (e.g. cloud feedback) from the model, which we know their models have been doing, despite acknowledging cloud feedback as the biggest source of greenhouse effect (their reason is that cloud feedback is 'natural').
[ + ] BikeLock
[ - ] BikeLock 4 points 8 monthsAug 24, 2023 10:44:40 ago (+4/-0)
Now please give me a doctorate in wind economics and solar theory, so I can flease, errr educate the goyi-general public.
[ + ] lord_nougat
[ - ] lord_nougat 3 points 8 monthsAug 24, 2023 14:05:20 ago (+3/-0)
[ + ] Ex_hack
[ - ] Ex_hack 3 points 8 monthsAug 24, 2023 10:52:52 ago (+3/-0)
[ + ] rabidR04CH
[ - ] rabidR04CH 2 points 8 monthsAug 24, 2023 16:46:29 ago (+2/-0)
[ + ] CHIRO
[ - ] CHIRO 2 points 8 monthsAug 24, 2023 16:01:40 ago (+2/-0)*
I need some help from a math whiz here. I'm not a slouch, but this is making me feel a bit retarded. If you cannot predict a future state of a system based on the current state (indeed, the authors say it is 'impossible'), then how do you generate probability distributions for future states? The idea is we have a deterministic system exhibiting random behavior (epistemic only?), meaning some outcome measure is random with respect to the starting conditions.
How do you develop probability distributions if this is true? By their nature, those distributions give you the objective likelihood for a particular state to obtain relative to other possible states. It seems that the only possibility is to make the system non-chaotic and something subject to experimental research, as you. . .cause it not to represent the climate any longer. And I guess this would also involve a huge amount of simplification, like leaving out entire inputs (e.g. cloud feedback) from the model, which we know their models have been doing, despite acknowledging cloud feedback as the biggest source of greenhouse effect (their reason is that cloud feedback is 'natural').
[ + ] UncleDoug
[ - ] UncleDoug 1 point 8 monthsAug 24, 2023 15:02:19 ago (+1/-0)
[ + ] whitemail
[ - ] whitemail 1 point 8 monthsAug 24, 2023 14:22:56 ago (+1/-0)
[ + ]deleted
[ - ] deleted 1 point 8 monthsAug 24, 2023 11:33:31 ago (+1/-0)
[ + ] DukeofRaul
[ - ] DukeofRaul 1 point 8 monthsAug 24, 2023 10:43:40 ago (+1/-0)