Data assimilation methods for urban air quality at the local scale

Abstract

Despite their wide applications in many fields of environmental sciences, data assimilation methods are still poorly adopted for the numerical simulation of pollutant dispersion at the local urban scale. In this study, we compare three data assimilation methods to evaluate the air quality at the local urban scale. The data assimilation methods used here are the Bias Ajustment Techniques, the Best Linear Unbiased Estimator, and the Source Apportionment Least Square method. We assess their performances on air pollution simulations in Lyon for the year 2008, focusing on ground-level NO2 hourly concentrations. The results indicate that the three methods improve air quality estimates and that their performances are similar. This study shows that data assimilation is a promising tool to ameliorate air quality simulations at the urban scale.