Bayesian models of the PM10 atmospheric urban pollution

In this paper we illustrate a forecast method of atmospheric pollution critical events caused by particulate matter (specifically PM10) based upon the application of Bayesian networks. These Bayesian networks model the temporal series of the pollutant during the day and the influence that meteorological parameters have upon them. Each network has received some data (coming from historical records or meteorological forecasts) and used them to calculate its own forecast. Typical inputs of the networks have been the pollutant concentration at a certain hour and the meteorological parameters at the further hours of the day. The output provided by the networks is the estimate of the probability of reaching a certain pollutant level in the various hours of the day. The results we have obtained in the prediction of the concentration of pollutant in the medium term are satisfactory and this approach can be profitably used to foresee critical episodes.