Dispersion modelling of a tall stack plume in the spanish mediterranean coast by a particle model

Abstract

A Lagrangian particle model has been used to simulate the dispersion of a tall stack plume of a power plant located in a complex coastal site at the Spanish Mediterranean coast, under summer meteorological conditions: land and sea breezes and thermal low effects. These are responsible for a particular behavior of plume (rotations greater than 90°). The model is based on the numerical solution of Langevin's equation (Sawford, 1984; Thomson, 1984, 1987; de Baas et al., 1986) by following the trajectories of many particles. The displacement of these particles is governed by meteorological parameters resulting from Eulerian wind data adjusted by an objective analysis model based on variational calculus. The adjusted values should satisfy continuity as a strong constraint (Sherman, 1978; Mathur and Peters, 1990). The model allows to simulate the atmospheric dispersion both in homogeneous and nonhomogeneous turbulence according to de Baas et al. (1986) and Zannetti (1990) schemes. The numerical results obtained by the dispersion model are compared with experimental data from a measurement campaign developed at the surroundings of Castellon power plant. The model is applied to the problem of predicting the ground level concentration (GLC) (3 m, above ground level) of the SO 2 emitted by the power plant. Model behavior was evaluated through several statistical indices: relative mean bias, normalized mean square error and the cumulative frequency distribution of the point-by-point ratio between observed and predicted concentrations. Both models were developed at the Instituto de Tecnología y Modelización Ambiental (ITEMA) of the Universidad Politécnica de Cataluña (UPC).