Long term measurements of submicrometer urban aerosols: statistical analysis for correlations with meteorological conditions and trace gases

Long-term measurements (over 4 years) of particle number size distributions (submicrometer particles, 3– 800 nm in diameter), trace gases (NO, NO2, and O3), and meteorological parameters (global radiation, wind speed and direction, atmospheric pressure, etc.) were taken in a moderately polluted site in the city of Leipzig (Germany). The resulting complex data set was analyzed with respect to seasonal, weekly, and diurnal variation of the submicrometer aerosol. Car traffic produced a peak in the number size distribution at around 20 nm particle diameter during morning rush hour on weekdays. A second peak at 10–15 nm particle diameter occurred around noon during summer, confirmed by high correlation between concentration of particles less than 20 nm and the global radiation. This new-particle formation at noon was correlated with the amount of global radiation. A high concentration of accumulation mode particles (between 100 and 800 nm), which are associated with large particle-surface area, might prevent this formation. Such high particle concentration in the ultrafine region (particles smaller than 20 nm in diameter) was not detected in the particle mass, and thus, particle mass concentration is not suitable for determining the diurnal patterns of particles. In summer, statistical time series analysis showed a cyclic pattern of ultrafine particles with a period of one day and confirmed the correlation with global radiation. Principal component analysis (PCA) revealed a strong correlation between the particle concentration for 20–800 nm particles and the NOand NO2-concentrations, indicating the influence of combustion processes on this broad size range, in particular during winter. In addition, PCA also revealed that particle concentration depended on meteorological conditions such as wind speed and wind direction, although the dependence differed with particle size class. Correspondence to: B. Wehner ([email protected])