| Abstrakt: |
All available polycyclic aromatic hydrocarbon (PAH) concentration data in ambient air obtained over the past 10 years in Germany were evaluated to clarify whether it is justified to use benzo(a)pyrene (BaP) as a marker compound for the total PAH exposure. The data basis comprises annual mean concentrations from 1990 to 1998 supplied by the emission protection authorities of the federal states with additional information on the region, year and site of measurement. The data are very heterogeneous with respect to sample size, the number of individual PAHs analyzed, place of origin and year. Nine of 25 individual compounds with sufficient sample size (74 < N < 518) were submitted to an explorative descriptive analysis and multiple regression analysis. Six compounds were selected for factor analysis to evaluate whether associations can be related to one or several background determinants. The concentration data of BaA, BeP, BaP, DBA, BghiP, COR, BbF, BkF and INP are not normally distributed and demonstrate great variability. Location, East-West differences persisting after the German reunion and time characterize the distributive patterns, e.g., from 1991 to 1997, a significant decrease in BaP could be determined based on the data from North Rhine-Westphalia (1991, N = 51, median 1.6 ng/m³; 1997, N = 45, median 0.7 ng/m³; P ≤ 0.0001). The BaP concentrations in the new federal states in the East (N = 80, median 2.2 ng/m³) are about twice as high compared to the ones in the old federal states in the West (N = 438, median 1.1 ng/m³; P ≤ 0.0001). BaP measurements taken at sites characterized by traffic (1.8 ng/m³), industrial (1.3 ng/m³) or urban environment (1.3 ng/m³) rank significantly higher compared to measurements taken at rural sites (0.92 ng/m³; P ≤ 0.0001). Factor analysis of the intercorrelation matrix of BaA, BeP, BaP, DBA, BghiP and COR yields one factor to explain 92% of the variability of the six individual... [ABSTRACT FROM AUTHOR] |
