A contribution to the problem of modelling the dispersion of pollutants in a time-varying, height-structured atmospheric boundary layer

Abstract

The numerical solutions of the time dependent advection-diffusion equation reported in this paper suggest that, for moderately tall stacks (effective height H, ⩽ 200m), the computed ground level concentration, at a down-wind of about 10 km, can be treated approximately as a “quasi-steady” function of time, when the input parameters (emission rate, wind velocity and turbulent diffusivity) change on time scales larger than about an hour. As the lateral, crosswind distribution of a steady-state plume can be assumed as Gaussian, if the meteorological parameters are horizontally homogeneous and the air stratification is not too stable (the wind rotation with height is not too marked, at and below the source height), the above result implies that the known non-Gaussian, analytic solutions of the two dimensional K-equation (Huang, 1979; Demuth, 1978) can be used to predict the g.l.c. also under time-varying conditions, as long as the input parameters change “slowly” over an hour and the vertical profiles are well approximated by power-law functions.