CITY AIR POLLUTION NUMERICAL PREDICTION SYSTEM AND ITS APPLICATION

The City Air Pollution Numerical Prediction System (CAPPS) consists of MM5 (or MM4) and the multi-scale grid-box model based on the integral of the atmospheric advective dispersion equation. The system has been used for predicting the Pollution Potential Index (PPI) and Air Pollution Index (API) in the urban areas without inventory data of sources. In the field of urban air pollution numerical prediction, CAPPS has overcome the difficulty brought by the inherent indefiniteness in the urban inventory of emissions.In this paper, CAPPS is applied to the air guality prediction experiments in Beijing and Shanghai and the results from these experiments are verified and compared with those from several photochemical air pollution models cited from publications. The comparison shows that the prediction accuracy of CAPPS is as good as most sophisticated photochemical models according to the correlation coefficients between the forecasted and the observed. Furthermore, the averaged prediction accuracy of pollution level has reached 61.5% in Beijing and Shanghai.