Verification of simple Wind Field Models for Air Pollution Simulation over the City of Trier

Air quality represents a key objective of urban governance, particularly in historically grown cities exhibiting complex topography. For this purpose, municipal authorities or engineering offices working on their behalf frequently utilise atmospheric models to assess the urban climate or the dispersion of pollutants. Due to the limited financial resources available, these tasks are unable to utilise research-grade high-resolution models such as PALM or WRF-LES, which require significant computing power for long-term simulations. Instead, the focus is either on the simulation of specific, limited periods under designated synoptic conditions. This approach is employed, for instance, in the context of urban climate assessment. Or simplified or analytical models are utilised, for example for the assessment of pollution dispersion . The latter are deemed to be "bad" in terms of the fact that they do not consider a significant number of processes included in a high-resolution, coupled physical model.  The central question is how "bad" the results are when compared to the actual wind field. 

The measurement of the wind profile above an city over a prolonged period is challenging. Nevertheless, their feasibility has increased significantly in recent years due to the advent of commercially available compact wind Lidar systems. In the present study, Lidar measurements obtained above the city centre of Trier, Germany, are utilised as ground truth, and are compared with simulations conducted using the two wind-filed models employed by the German regulatory pollutant dispersion model, AUSTAL. One of these is TALdia3, an analytical wind field model, and the other is METRAS-PCL, a prognostic wind field model derived from Uni Hamburg's mesocale model METRAS. In order to assess the resulting uncertainties in pollutant dispersion prognoses, hypothetical reference sources are simulated and the results are analysed.