7A.4 Modeling NO2 concentrations in the urban area of Berlin/Brandenburg with WRF-Chem: model evaluation and sensitivity to traffic emissions

Kuik, Friderike, and Tim Butler, Institute for Advanced Sustainability Studies Potsdam, Germany

Despite extensive regulations, air pollution remains a challenge in Europe, especially in urban areas. For studying the impact of traffic emission scenarios on NO2 concentrations in the Berlin-Brandenburg urban area (Germany), WRF-Chem is set up and evaluated against observations. The model setup includes nested domains with horizontal resolutions of 15km, 3km, and 1km, a single-layer UCM and anthropogenic emissions from TNO-MACC III, which are downscaled to a resolution of 1kmx1km using proxy data. Compared to observations, urban background NO2 concentrations are underestimated at daytime in the summer, but simulated reasonably well in the winter. Underestimations are most pronounced during weekdays when traffic, and thus traffic emissions, peak. Furthermore, spectral decomposition of modeled and observed NO2 time series shows that the largest part of the model error can be attributed to the diurnal component. This suggests that both the diurnal cycle of emissions as well as problems with mixing, particularly in the summer, might cause this model bias. As a consequence, a sensitivity study with increased traffic emissions during the rush hours in the city center is done. Additionally, a statistical downscaling method for estimating NO2 concentrations at the roadside, using WRF-Chem results, is presented.