Generally, it is very difficult to distinguish between soil and BTB06584? fly ash contribution. However, resuspension of soil particles during wintertime, particularly when the ground is frozen, seems to be less important process than combustion. The second factor (F2) explained from 13 to 30% of the variance and was probably associated with industrial pollution sources because Cu, Pb, Cr, Mn, Fe, and As are known to have industrial origin. What is interesting, these metals can represent different types of pollution sources. Table 3VARIMEX normalized rotated factor loadings for a factor analysis on Brzezina PM10 data set. Loadings for which the absolute value is greater than 0.700 are indicated in italic.Daily contributions of each source to ambient PM10 in Brzezina were estimated using FA-MLRA methodology [7].

Figure 3 shows an example of time series of the contributions of the identified sources for the summer period in 2009 and winter in 2011. On the basis of daily contributions, the average mass contributions of each source (SC��source contribution) were calculated and the results are summarized in Table 3 (the last raw). It can be seen in Table 3 that the average mass contribution of distant industrial emission sources was 12% and 15% in the summer 2009, 2010 and 6% and 20% in the winter 2010, 2011. Background dust contributed with 61% or 31% of total PM10 mass during the summer campaigns. However, nonidentified sources contributed with about 55% in the summer 2010.

It was probably due to high concentrations of sulphates, nitrates, and a certain amount of organic carbon which are characteristics secondary pollutants and are formed, especially in the summer, when solar radiation and the temperature are high. Secondary pollutants are usually indicative for a long distance transport. However, these pollutants were not measured.Figure 3Daily source contributions to PM10 obtained by FA-MLRA and measured by gravimetry for summer 2009 and winter 2011.Local combustion sources contributed with 79% in winter 2010 and 55% in winter 2011, reflecting serious local problem associated with PM10 air pollution. Generally good compatibility was found between Dacomitinib the measured PM10 concentrations by gravimetry and estimated by FA_MLRA (Figure 3). The squared correlation coefficient, R2, was 0.80 and 0.72 for summer and winter data, respectively. Now, the question is what industrial sources could contribute to total PM10 mass. Trajectory cluster analysis, conditional probability functions (CPF), and potential source contribution functions (PSCF) have been successfully used to identify transport paths and source areas [21�C23].