REAL-TIME PREDICTION OF SO2 CONCENTRATION IN THE VENETIAN LAGOON AREA

The paper describes a certain number of stochastic predictors of daily and hourly SO2 concentration in the Venetian Lagoon area. More precisely, the SO2 concentration in each inhabited subarea of the region is represented by the Dosage Area Product (DAP), namely the integral of the daily dosage over each subarea. From each stochastic model a real time predictor is derived, namely a recursive relationship allowing to forecast, at the beginning of each hour or day, the DAP levels in future periods. Moreover, a comparison is made between the performance of predictors using only recorded SO2 data and predictors using also meteorological data such as wind directions and speeds, in order to evaluate the inclusion of such meteorological information. Possible effects of unknown fluctuations of the emission on the quality of hourly DAP forecast are also discussed. © 1979.