A Semi-Lagrangian Spectral Algorithm for Pollutant Transport and Diffusion Studies

Abstract

A semi-Lagrangian model was developed based on the atmospheric transport-diffusion equation, employing the semi-Lagrangian technique incorporated with the spectral method. The numerical algorithms of the semi-Lagrangian technique and the spectral method for solving the transport-diffusion equation, subjected to boundary conditions, are described. The model simulated the transport and diffusion of pollutants from an elevated point source. The accuracy of the algorithms was revealed by numerical tests compared with analytical solutions in cases of simple profiles of the mean wind and the eddy diffusivity. To demonstrate the feasibility of the algorithms for studying air pollution in the atmospheric boundary layer, the model, using realistic profiles of the mean wind and the eddy diffusivity, simulated the distributions of pollutant concentrations under various atmospheric stabilities. As will be shown, the algorithms are capable of producing excellent results, making them attractive and promising for regional studies of pollutant transport and diffusion in the atmosphere.