1Leibniz Institute for Tropospheric Research (IfT), Permoserstrasse 15, 04318 Leipzig, Germany
2Interdisciplinary Environmental Research Centre (IÖZ), TU Bergakademie Freiberg, 09599 Freiberg, Germany
3Helmholtz Zentrum München (HMGU), German Research Center for Environment Health, Institute of Epidemiology, 85758 Neuherberg/Munich, Germany
4University of Augsburg, Center for Science and Environment, 86159 Augsburg, Germany
Abstract. Aerosol particle number size distributions (size range 0.003–10 μm) in the urban atmosphere of Augsburg (Germany) were examined with respect to the governing anthropogenic sources and meteorological factors. The two-year average particle number concentration between November 2004 and November 2006 was 12 200 cm−3, i.e. similar to previous observations in other European cities. A seasonal analysis yielded twice the total particle number concentrations in winter as compared to summer as consequence of more frequent inversion situations and enhanced particulate emissions. The diurnal variations of particle number were shaped by a remarkable maximum in the morning during the peak traffic hours. After a mid-day decrease along with the onset of vertical mixing, an evening concentration maximum could frequently be observed, suggesting a re-stratification of the urban atmosphere. Overall, the mixed layer height turned out to be the most influential meteorological parameter on the particle size distribution. Its influence was even greater than that of the geographical origin of the prevailing synoptic-scale air mass.
Size distributions below 0.8 μm were also measured downstream of a thermodenuder (temperature: 300 °C), allowing to retrieve the volume concentration of non-volatile compounds. The balance of particle number upstream and downstream of the thermodenuder suggests that practically all particles >12 nm contain a non-volatile core while additional nucleation of particles smaller than 6 nm could be observed after the thermodenuder as an interfering artifact of the method. The good correlation between the non-volatile volume concentration and an independent measurement of the aerosol absorption coefficient (R2=0.9) suggests a close correspondence of the refractory and light-absorbing particle fractions. Using the "summation method", an average diameter ratio of particles before and after volatilisation could be determined as a function of particle size. The results indicated that particles >60 nm contain a significantly higher fraction of non-volatile compounds, most likely black carbon, than particles <60 nm. The results are relevant for future health-related studies in that they explore the size distribution and time-dependent behaviour of the refractory component of the urban aerosol over an extended time period.