Abstract
A large set of hourly meteorological data from seven towers deployed in Montesina Valley (Cordoba, Spain) is studied in relation to the prevailing synoptic situation. The complete collection of daily synoptic maps for 1985-90 has been classified in seven basic weather patterns. First-order transition probabilities and weatherpattern persistences have been calculated for each of the patterns. The behavior of the local valley wind field is described as a function of the synoptic patterns. The work reported here, based on observations of the characteristic time and space patterns of flow in the valley under different synoptic conditions, has made it possible to obtain a set of empirical rules and regression functions to produce forecasts of the local wind field as a function of the synoptic situation and the hour of the day to be used in the emergency plans. The result of this work is currently being implemented in an atmospheric dispersion module of an expert system that will be used as a tool to predict the evolution of accidental and routine hazardous emissions to the atmosphere in complex terrain such as valleys and coastal areas.