Improvement of an urban turbulence parametrization for meteorological operational forecast and air quality modeling

During the last century, urban pollution has increased with the growth of cities. Urban air quality has become a high priority as it is directly linked to concerns such as human exposure and health. The present work is dedicated to urban air quality modeling with focus on urban meteorology. The main goal is to improve meteorological and air quality simulations in urban areas. Based on measurements and numerical air quality simulations, Chapter 3 describes the meteorological situation and tests an emission inventory for an ozone pollution episode in Mexico City (2 March 1997), in order to determine the principal factors to be accounted in an air quality study. The Mexico City case shows the great influence of meteorological conditions and pollutant emissions on air quality. A thorough understanding of the phenomena governing the meteorological conditions, like small scale convergence, and an accurate emission inventory validated by VOC measurements in the city, are sensitive elements in an air quality study. In Chapter 4, the presence of an Urban Heat Island (UHI) is underlined over the city of Basel (Switzerland) with the BUBBLE measurements. Further on, the ability of aLMo (the operational weather prediction model of MeteoSwiss) to reproduce the effect of a city on the boundary layer atmospheric flow fields is investigated. Results show that aLMo is not able to reproduce the UHI and hence that its surface scheme based on the Monin-Obukhov Similarity Theory (MOST) is not adapted for the urban areas. Therefore, the Buildings Effects Parametrization (BEP) which has been developed by Martilli et al.(2002) especially for urban areas, is implemented in aLMo. Results show that alMo is now able to reproduce the main behavior of the urban boundary layer as the UHI and BEP has hence a real enhancement potential for aLMo. Chapter 5 shows the sensitivity of aLMo with respect to its vertical resolution. In order to limit the sensitivity of aLMo to the grid resolution, BEP is modified. The mesoscale model furnishes the upper boundary conditions for the inner calculation of BEP. That is, BEP recalculates independently the vertical profiles of wind, temperature and energy based on the surface fluxes of momentum, heat and turbulent kinetic energy. The results obtained show a decrease in the sensitivity to the resolution and a better agreement with the measurements. Furthermore, the modified version of BEP gives additional meteorological fields (temperature, wind and TKE) in the urban canopy. The computed temperature in the urban canopy shows a good agreement with measurements in a Basel street canyon. This work shows that it is not necessary to have a high resolution for taking into account modification of the atmospheric flow fields induced by a city.