- The purpose of my thesis is to describe a method which allows for the determination of pollutant emissions within a particular area under investigation through ambient air measurements. The concentration distributions of sources of emission on the windward and lee side were defined through vertical and horizontal profile measurements. On the basis of the concentration differences from these results and in combination with meteorological parameters, mass flows of the examined air pollutants were calculated and a mass balance was established. This is how the emissions within the area under investigation could be determined meteorologically and model calculations of the emissions, conducted simultaneously, were validated. This validation is important as emission models are used more and more frequently in the field of air pollution control. On the one hand, modelled emissions are included as input data into the more and more common dispersion models and on the other hand, they are needed for emission reduction scenarios. As the political and financial consequences of using these emission data are significant, it is inevitable to verify the accuracy of the calculated emissions.
This thesis uses the results of two experiments to describe the balancing method. In the first project, the balancing of all emitted air pollutants was conducted for the first time for a whole city. Concentration measurements had been carried out for several months in and around Augsburg. The vertical and horizontal distribution of air pollutants and of meteorological parameters was determined on the city's windward and lee side by means of three tethered balloon measurement systems, a remote controlled airship, two aircrafts and further measurement equipment. For the same time period, an emission inventory was established for the city in a high temporal, spatial and substantial resolution. The measured and balanced emissions were then compared with the modelled results, and thus validated the emission model. The tethered balloon and airship measurements were carried out or supported by the IFK. The big advantages of conducting measurements with a tethered balloon or an airship have shown themselves in this experiment as well. What the EVA project has also shown, however, is that a city like Augsburg with a population of about 255,000 has a horizontally very broad exhaust gas plume, which is to be measured as completely as possible. For such a big city, tethered balloon and airship measurements were not enough to gather sufficient measurements for the determination of an exhaust gas plume of 9 km. What has proven itself for the determination of the complete plume, however, is the combination of these two measurement systems with aircraft measurements. The comparison of measured NOx emissions with modelled ones gave equal results for most data which were suitable for an analysis, although the modelled values were constantly higher than the measured ones. But the instabilities for both the measurements and the calculation methods and emission models have partially been significant.
In the second project, extensive measurements and analyses were carried out along a motorway. The comparison of the NOx emission factors determined through vertical profile measurements with the NOx emission factors determined through an emission model revealed significant deviations. The deviations for heavy traffic were particularly important, with the measured emission factors being 20 % to almost 100 % higher than the modelled emission factors based on the Manual for Emission Factors (Handbuch für Emissionsfaktoren, HBEFA, Version 1.2). Following these results, the emission factors for heavy traffic have been readjusted upwards in the current version (2.1). The comparison of the VOC emission factors gave a very high consistency for the BTX components - benzene, toluene and xylene - as sufficient measurement data have been available for the emission models there. Other VOC components partly showed significant deviations between the measured and the modelled emission factors, with the measured factors lying principally higher than the modelled ones. The deviations for some components were six to 20 times higher! The reason for this is the fact that the data for the VOC emission factors available for the emission models have not been sufficient. All particle emission factors, which were determined in the BAB II project, were about the same scale as the reference values. The values for the active particle surface, determined in the BAB II project, were slightly higher than the reference values. For all other calculated and compared particle parameters, the emission factors determined in the BAB II project stayed within the maximum and minimum reference values.
The comparison of the emission factors of soot particles gave a very high consistency both in the direct comparison of the measurements with the results from the emission models of the BAB II project and in their comparison with reference values.