Atmospheric Environment 37 (2003) 4217–4231 On the impact of urban surface exchange parameterisations on air quality simulations: the Athens case Alberto Martilli a, *, Yves-Alain Roulet a , Martin Junier a , Frank Kirchner a , Mathias W. Rotach b , Alain Clappier a a Air and Soil Pollution Laboratory, Swiss Federal Institute of Technology, Lausanne (EPFL), 1015-Lausanne, Switzerland b Swiss Federal Institute of Technology, Zurich (ETHZ), Institute for Atmospheric and Climate Science, 8057-Zurich, Switzerland Received 30 January 2003; accepted 7 July 2003 Abstract Most of the standard mesoscale models represent the dynamic and thermodynamic surface exchanges in urban areas with the same technique used for rural areas (based on Monin–Obukhov similarity theory and a surface energy budget). However it has been shown that this technique is not able to fully capture the structure of the turbulent layer above a city. Aim of this study is to evaluate the importance for meteorological and air quality simulations, of properly capture the dynamic and thermodynamic surface exchanges in urban areas. Two sets of simulations were performed over the city of Athens (Greece): a first using a mesoscale model with a detailed urban surface exchange parameterisation (able to reproduce the surface exchanges better than the traditional method), and a second with the traditional approach. Meteorological outputs are passed to a Eulerian photochemical model (the photochemical model is run offline). Comparison with measurements shows better agreement for the simulation with the detailed parameterisation. The differences between the simulations concern, mainly, wind speed (maximum difference of 0.5–1 m s 1 ), night-time temperatures (2–3  C), turbulence intensity (2 m 2 s 2 ) and heat fluxes (0.15 K m s 1 ) over the urban area, urban nocturnal land breeze intensity, timing and extension of sea breeze. These differences modify the pollutant distribution (e.g. for ozone maximum differences are of the order of 30 ppb). Differences between the simulations are also found in AOT60 values (inside and outside the city) and in O 3 chemical regimes. r 2003 Elsevier Ltd. All rights reserved. Keywords: Urban boundary layer; Urban surface exchange parameterisation; Mesoscale models; Photochemical models; Coastal air pollution; Sea breezes 1. Introduction Nowadays, air quality models are widely used to evaluate air pollution abatement strategies (Andersson- Skold and Holmberg, 2000; Metcalfe et al., 2002; Palacios et al., 2002; and many others) and new model output (e.g. accumulated exposure over a threshold, chemical regimes, etc.) are required to estimate the impact of air pollution on human health, vegetation growth and historical buildings preservation. To have this information, the precision in the spatial distribution, as well as the time evolution of pollutant concentrations is crucial. In this context, the modification of the planetary boundary layer (PBL) structure, induced by the presence of an urban area, can play a very important role (Bornstein et al., 1993; Oke, 1995; Martilli, 2002). In the standard version of most of the state-of-the-art mesoscale models, dynamic and thermodynamic surface exchanges in cities are represented using the same formulations adopted in rural areas (mainly based on ARTICLE IN PRESS AE International – Europe *Corresponding author. Department of Earth and Ocean Science, University of British Columbia, 6339 Stores Road, Vancouver, BC, Canada V6T 1Z4. Tel.: +1-604-8220531; fax: +1-604-8226088. E-mail address: amartilli@eos.ubc.ca (A. Martilli). 1352-2310/$ - see front matter r 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S1352-2310(03)00564-8