Emissions of the city of Augsburg determined using the mass balance method

Air quality models require emission data of air pollutants with high temporal and spatial resolution. The highly resolved emissions are calculated by complex emission models and may thus be subject to considerable uncertainties. To assess the uncertainty of the emission model calculations, emissions of the city of Augsburg in southern Germany were determined experimentally on several days in October 1998 using the technique of mass balance. Two algorithms were used to estimate the CO and NOx emissions from the data measured by two research aircraft, two tethered balloons, and an airship on the lateral surface enclosing the area of investigation. The two algorithms provided results which agree within the calculated uncertainties. The differences can be attributed mainly to the different interpolation methods applied to the layer between the lowest flight level and the ground. The measurements show that on the different weekdays both CO and NOx emissions varied by about a factor of two which is much larger than the variability predicted by the emission model. Higher than predicted CO emissions were observed on Saturday, 10 October 1998, whereas NOx emissions on 10 October were substantially smaller than predicted by the emission model.