A computational fluid dynamic modelling approach to assess the representativeness of urban monitoring stations.

Air quality measurements of urban monitoring stations have a limited spatial representativeness due to the complexity of urban meteorology and emissions distribution. In this work, a methodology based on a set of computational fluid dynamics simulations based on Reynolds-Averaged Navier-Stokes equations (RANS-CFD) for different meteorological conditions covering several months is developed in order to analyse the spatial representativeness of urban monitoring stations and to complement their measured concentrations. The methodology has been applied to two urban areas nearby air quality traffic-oriented stations in Pamplona and Madrid (Spain) to analyse nitrogen oxides concentrations. The computed maps of pollutant concentrations around each station show strong spatial variability being very difficult to comply with the European legislation concerning the spatial representativeness of traffic-oriented air quality stations.