Author(s)

Cecile Honore and Robert Vautard

Abstract

We aim at estimating the number of independent variables governing the evolution of a deterministic model of photochemical air pollution in an urban area. The urban boundary layer is modeled by one or more boxes, in which the concentrations of the chemical species we take into account are assumed to be homogeneous and driven by dynamical, physical and chemical processes. First, a theoretical approach, inspired by dynamical system theory, is used: we study the correlation exponent related to the trajectories of the concentrations computed by the model. This exponent ranges at very low values and provides us an estimate of the number of independent variables. A second, more straightforward way to proceed is to parameterize some of the production-loss chemical terms as a function of a few variables. To achieve this parameterization

Keywords