“In short it looks like less than 4% of the science, the climate change part, is doing about 55% of the modeling done in the whole of science.”
From Investor’s Business Daily:
“In short it looks like less than 4% of the science, the climate change part, is doing about 55% of the modeling done in the whole of science.”
From Investor’s Business Daily:
“… the modeler said “That can’t be right it doesn’t fit the model.” ”
Unfortunately, Slof, there are so many who say the same thing and have no clue why it makes no sense.
This story was related to be by a friend. Some years back I worked on the Yucca Mountain project in Nevada. One item of interest was heat conductivity so, of course they built a model from core measurements (actually very few core measurements). The friend of mine who worked for Sandia National Labs said he was sitting in a meeting with the modeler, some managers and a a fellow who was actually making measurements on some new data. When he reported his results the modeler said, “That can’t be right it doesn’t fit the model.”
SMS is correct. Back in the mid to late 80’s the military spent a huge amount of money to try and model atmospheric turbulence and scintillation to attempt to predict its effect on high energy lasers. If there could be a predictive model, it would have greatly simplified the use of laser weapons. Over several years both modeling and lab tests were developed. The results showed the models and the lab tests correlated for only the first 2 minutes or so, showing the models were technically sound but then chaos quickly set in and everything fell apart. You cannot discern order from chaos. This was the seed that greatly accelerated the development of adaptive optics where the atmospheric phenomena are measured and corrected for in real time; a very expensive solution but a solution. Most all new astronomical telescopes and high energy laser weapons use adaptive optics to correct for atmospheric distortion. The atmosphere’s behavior is the climate and can not be predicted, it is chaotic by definition.
Models, any model is useless unless it can be validated. So far not a single ‘climate model’ has been able to predict anything. Data sets are fudged, models are tweaked to try to get some predetermined result, then when the results don’t match the real world, the results are hidden and lied about and on to the next round of fake models.
You cannot model a chaotic system. All the climate models currently being used to determine the effects of adding a small trace amount of gas to an existing trace amount of gas add up to nothing more than a waste of the taxpayers money.
I have used numerous computer models in my work – from simulated free fall of an object to estimate coefficient of air friction to simulated titrations of multi-basic metal cations to simulated trajectories of ions in the analyzer of a mass spectrometer to analysis of the viscous behavior of mixtures of gases to multi-variate least-square regression of mass spectrometer data to evaluate compositions of natural gas mixtures.
I have NEVER assumed that the models exactly represented the reality, let alone were reliable at predicting unanticipated phenomona.
In a small, closed system where the mathematical model is shaped by the known physics models can be useful for interpolating data, but should only be used for extrapolations in short ranges, and with extreme caution as the uncertainty grows with the distance of the point of interest from the database of experience with which the model is conditioned.
Even a tyro’s knowledge of fluid dynamics should be sufficient to cause one to cringe at the phrase ‘global circulation model’.