A novel approach to atmospheric dispersion modelling: The Puff-Particle Model

Abstract

The Puff–Particle Model (PPM) uses a new approach to problems of three-dimensional atmospheric dispersion from micro- to mesoscale. The pollutant particles are grouped in clusters treated as Gaussian puffs, which are dispersed making use of the concept of relative diffusion. The centre of mass of each puff is moved along a stochastic trajectory. This trajectory is derived from particle trajectories given by a Lagrangian stochastic-dispersion model. In this way, the PPM retains the advantages of traditional puff models and those of particle models, and is able to take into account the correct probability density function of the stochastic velocity components. The effect of meandering (caused by turbulent eddies larger than the puff but not resolved by the flow field) is simulated by the puff centre trajectories, yielding a complete description of dispersion. The PPM is validated using measurements from three tracer experiments in Copenhagen, Lillestrøm and Kincaid.