Stochastic Modeling of Dispersion in the Convective Boundary Layer

Abstract

Dispersion in the convective boundary layer (CBL) is a complex process caused by the vertical inhomogeneity, skewness, and large time scales of the turbulence. The inhomogeneity is most significant near the surface. and capping inversion where the large-scale updrafts and downdrafts are blocked and small-scale turbulence is important (Hunt et al., 1988). Owing to these complications, standard statistical or eddy-diffusion models cannot give realistic descriptions of vertical dispersion in the CBL. However, a Langevin or stochastic model for the Lagrangian velocity of passive particles is able to do so. One of the first Langevin efforts was by Baerentsen and Berkowicz (1984) who modeled vertical dispersion by treating the particle motion in updrafts and downdrafts separately. Although their model was somewhat complicated and required four time scales, it produced results in good agreement with the laboratory experiments of Willis and Deardorff (1976, 1978, 1981). More recently, Sawford and Guest (1987) used a generalized stochastic model with a non-Gaussian random forcing and a single time scale to simulate the dispersion; they also found good agreement between their results and the laboratory data.

The National Center for Atmospheric Research is sponsored by the National Science Foundation.