"Has anybody yet come up with a model of greenhouse warming that includes cloud formation? yes or no?"
Actually, formal answer to your question is "yes". For example,
the most massive network computational experiment,
climateprediction.net ,
does have the cloud component in their model. So the greenies will be quick to point this out.
However, your question is ill-posed. The right question to ask is: to what extent the cloud formation parameterization can be verified, and a correct range selected for prediction experiments? The answer to this question is: it is not possible.
Under-educated "post-normal" scientists and their groupies have difficulties to comprehend seriousness of this question. They could point you to Computational Fluid Dynamic (CFD), where the whole climate modelling belongs to, formally. However, there is one "minor" difficulty. The problems in question contain turbulent processes at extrememly high Reynolds numbers. The problem of turbulence is so far beyond human capabilities to be solved in details, so some fields has to be approximated with so-called "parametrizations". The CFD works (sometimes with only up to 30 or 50% errors) for certain problems because correct paramterizations can be established by running real-time real experiments, where huge detailed information can be collected, and later turned into simplified parametrizations. The main condition for the success of this approach is the ability to run experiments longer than the characteristic time of the problem. Then the derived parametrizations can be extrapolated into similar flow geometries, and calculations give us more or less accurate picture of processes. Unfortunately, this cannot be true for models of climate. Past climate is known only to very little details, all fluctuating variables have been convoluted into averages, and inverse problem cannot be resolved backwards. So, there is not enough information to reconstruct glaciations and cloud history and derive parametrizations from the past. For current instrumental era, there is still not enough spatio-temporal information in the first place, plus the data has to be monitored for longer than the typical time of climate changes, which is what, 10,000 years? So, the parametrizations cannot be established using scientific methods. What is left is a guess, which can give you almost any answer you wish for. That is the real problem in predicting climate turns.
- Ali |
