Replies

Yeah, and the computer models have actual measurments fed in the front end, and use equations for fluid dynamics, heat transfer, etc. which are all based on observations. The models are more extensive than simple curve fitting. My point stands.

BTW, there's nothing wrong with computer models per se: if someone ever makes a computer model based on the only last 150 years of data and general physical principles that correctly retrodicts the little ice age, the subsequent warming, and the warm period before the little ice age, I'll be interested in seeing what it predicts.

Yeah, and the computer models have actual measurments fed in the front end, and use equations for fluid dynamics, heat transfer, etc. which are all based on observations. The models are more extensive than simple curve fitting. My point stands.

The models are BS read this link The global warmers set the models to suppport their hypothesis. You can make a computer model to predict anything you want and cover it with all sorts of buzzwords, but it's still BS.

if someone ever makes a computer model based on the only last 150 years of data and general physical principles that correctly retrodicts the little ice age, the subsequent warming, and the warm period before the little ice age, I'll be interested in seeing what it predicts.

Old news.

Climate model shows dual cause [of 20th century temperature trends]

And the models have improved since this was published:

By Looking Back, Scientists See a Bright Future for Climate Change Forecasting

"Chen and his colleagues report in the April 15 issue of the journal Nature that an improved climate model, known as LDEO5, for the first time predicted every major change in the temperature of the tropical Pacific Ocean over the past 150 years with up to two years of advance notice."

There is one problem with your request: the models use data that constitutes the boundary condition/starting point. The less accurate the starting point data, the less accurate the model will follow what actually happened. The further you go back in history, the data become less accurate, and therefore add more noise to the model reconstruction. So what you're really asking for is a model that makes accurate predictions based on bad data, and nobody really expects that to work.