Air pollution modelling in urban environment based on a priori and reconstructed data

This paper presents preliminary results of the effectiveness analysis of an air quality forecasting system for the city of Novosibirsk with replenishment of the missing information on emission sources by solving an inverse problem with urban monitoring network data. In solving the inverse problem, a priori information about the location and mode of the sources is used. To simulate concentration distributions, the WRF-Chem model is used, and a simplified model of chemical transport is applied to solving the inverse problem. These models are offline coupled in a hybrid forecast system in order to improve the initial information about the spatial distribution of emission intensity and air quality forecast, respectively. The results of numerical experiments and their analysis are presented. The influence of an urban parameterization on the results of the forecast is shown.