History and epistemology of models: Meteorology (1946-1963) as a case study (Q5944755)

The paper deals with the modeling in meteorology. The meteorology was a descriptive science for a long time, with an empirical tradition. It followed a dynamical meteorology and a so-called synoptic tradition. Data were collected for a long period and forecastings were given. The more complicated modeling is connected with computers, science of chaos, dynamical systems and numerical instability! Richardson's model is described, it is given by 7 PDE. Four of them are based on Newton's law and Boyle's law, the others are related to mass-density and to temperature. The PDE-system was numerically solved by means of difference methods. A 6-hour-forecast needed 6 weaks! After the 1940s some new models were initiated, e.g. by J. v. Neumann and J. Charney. Important problems during World War II forced the development of meteorology. A hierarchy of atmospheric models of increasing complexity were studied, the restrictions were given by the computers capacity. Eventually, the paper refers to the results of Lorenz (Lorenz attractor), who pointed out that very complicated behaviour could arise from very simple systems. A big number of famous scientists working in this topic is mentioned in the present, very nice paper.