Advanced Operational Air Quality Forecasting Models for Urban and Regional Environments in Europe: Madrid Application

Abstract

During the last decades an increased interest on Air Quality Models has been found. Historically, air quality models have been developed in the 60’ and 70’ based on a very simplified solution of the Eulerian transport equation. The simplified differential transport equation had an analytical solution and therefore the results were obtained rapidly. Nowadays, these type of models (Gaussian) are applied with computer times very much reduced in comparison to the 60’ and 70’ operational models. These models have become quite complex by parameterizing new and more sophisticated problems such as buildings, very complex terrain, etc. However, because of the extraordinary advance of the computer power, during the 80’ a new generation of three-dimensional models started to be developed. During the 70’ the mathematical basis for solving the Navier-Stokes equation on the air fluid were established by developing sophisticated numerical techniques which conducted to have the first possible approaches for having a full three dimensional solution of the air dynamics. At the end of the 80’ decade the first models for mesoscale areas started to appear (MM5, MEMO, RAMS, etc.). (Pielke [5]; Flassak and Moussiopoulos [1]; Grell et al. [3]) These models offer a detailed diagnostic and prognostic patter of the wind, temperature and humidity -in addition of the pressure, vorticity, turbulent fluxes, etc.- in a three-dimensional mesoscale domain. The Gaussian approach was appropriate for point or lineal sources in a very local domain (5-10 km) but the rigid meteorological conditions and the difficulties of having on-line information to be incorporated to the model made almost impossible to apply these models for regional domains (20-500 km).