The simulation of particle motion in the atmosphere by a numerical random‐walk model

Abstract

Abstract Of all the models of turbulent motion that are based on a random‐walk technique, the simplest are ones which simulate individual particle trajectories. The accuracies of two such models are assessed by applying them to low‐level diffusion in the atmosphere. It is found that one based on a Markov‐Chain principle predicts particle concentrations over short and medium ranges which are consistent with observation. Calculations are extended to include dispersion in unstable conditions, and by postulating the dependence of several important turbulence parameters on z/L , it is possible to derive a physically realistic model.