Extension of an operational short-range dispersion model for applications in an urban environment

Abstract

When operationally modelling dispersion of air pollutants over a city, "conventional" dispersion models are often used with modifications in some of the parameters, such as the roughness length. However, the presence of a roughness sublayer (RS) is usually neglected, although its turbulence structure is different from that in the surface layer. Therefore, in the present work a roughness sublayer for urban applications is introduced. The principle of the modification has already been tested by simulating two tracer experiments using a Lagrangian particle model, and it was shown that the introduction of an RS clearly improves the model performance for both experiments. In this paper, it is shown that the same holds for the introduction of the RS in the multi-source/multi-receptor Gaussian OML model. Its performance is investigated for the city of Zurich in the year 1990, when a detailed emission inventory and observations at 29 stations are available. It is concluded that the introduction of the RS increases the physical significance of the model in urban environments and thus the credibility of its predictions.