First detailed air pollution analyses by assimilating UAV observations with EURAD-IM

In atmospheric science, Unmanned Aerial Vehicles (UAV) are relatively new technologies that started to be used recently for the assessment of atmospheric composition, bringing many opportunities to improve air monitoring. Within the MesSBAR (automatisierte luftgestützte MESsung der SchadstoffBelastung in der erdnahen Atmosphäre in urbanen Räumen/Automated airborne measurement of pollution levels in the near-ground atmosphere in urban areas) project, new drones carrying trace gas and aerosol instruments have been developed to measure near-surface vertical profiles of atmospheric pollutants with high temporal resolution while being flexible, inexpensive, and able to perform measurements close to the emission sources.

The use and benefit of the assimilation of such high-frequency observations in a regional chemical transport model have not been studied yet. However, it presents a possible promising opportunity to improve air quality forecasting as in particular, it supports to receive a better representation of the pollutants in the planetary boundary layer.

In this work, we evaluate the impact of the assimilation of UAV observations on the analysis and forecast of traces gases and aerosols. The observations used resulted from a series of drone measurements carried out close to a motorway in Wesseling, Germany, from 21 to 23 September 2021 as part of the MesSBAR project. We perform high-resolution analyses (1 km x 1 km spatially and ~20 s temporally) assimilating UAV profiles using the 4D-Var data assimilation technique in the EURopean Air pollution Dispersion - Inverse Model (EURAD-IM). The results are compared in the first place to the operational EURAD-IM forecast without assimilation to evaluate the impact of the UAV observations on the analysis. Then, the analysis is compared to ground-based observations measured during the campaign and to other independent data to evaluate the analysis accuracy. The improvement in the analysis obtained by UAV observations with respect to emissions factor optimization is assessed and discussed.