AIR-QUALITY PREDICTIONS - DISPERSION MODEL VS BOX-JENKINS STOCHASTIC-MODELS - AN IMPLEMENTATION AND COMPARISON FOR ATHENS, GREECE

This paper presents a comparison between the air quality predictions o btained by the use of a dispersion model and univariate Box-Jenkins st ochastic models. These two different approaches are implemented in ord er to predict the hourly mean values of concentration for CO and SO2 i n three different locations of the wider Athens area. The air pollutan ts predictions are compared with the observed values of concentrations , from the existing air quality monitoring network. In this way certai n criteria are calculated and the predicting capability of the two app roaches is evaluated for the specific case of the urban area of Athens , Greece. Finally the implementation of the two approaches in practice is considered and the advantages/disadvantages of both approaches are determined and discussed. (C) 1998 Elsevier Science Ltd. All rights r eserved.