Particulate matter source apportionment in a village situated in industrial region of Central Europe

Abstract

The bilinear receptor model positive matrix factorization (PMF) was used to apportion particulate matter with an aerodynamic diameter of 1–10 μm (PM1–10) sources in a village, Březno, situated in an industrial region of northern Bohemia in Central Europe. The receptor model analyzed the data sets of 90- and 60-min integrations of PM1–10 mass concentrations and elemental composition for 27 elements. The 14-day sampling campaigns were conducted in the village in summer 2008 and winter 2010. Also, to ensure seasonal and regional representativeness of the data sets recorded in the village, the spatial-temporal variability of the 24-hr PM10 and PM1–10 within 2008–2010 in winter and summer across the multiple sites was evaluated. There were statistically significant interseasonal differences of the 24-hr PM data, but not intrasummer or intrawinter differences of the 24-hr PM1–10 data across the multiple sites. PMF resolved seven sources of PM1–10. They were high-temperature coal combustion; combustion in local heating boilers; marine aerosol; mineral dust; primary biological/wood burning; road dust, car brakes; and gypsum. The main summer factors were assigned to mineral dust (38.2%) and primary biological/wood burning (33.1%). In winter, combustion factors dominated (80%) contribution to PM1–10. The conditional probability function (CPF) helped to identified local sources of PM1–10. The source of marine aerosol from the North Sea and English Channel was indicated by the Hybrid Single Particle Lagrangian Integrated Trajectory Model (HYSPLIT). Implications: This is the first application of PMF to highly time/size resolved PM data in Czech Republic. The coarse aerosol fraction, PM1–10, was chosen with regard to industrial character of the region, sampling site near the coal strip mine and coal power stations. Contrary to expectation, source apportionment did not show dominance of emissions from the coal strip mine. The results will enable local authorities and state bodies responsible for air quality assessment to focus on sources most responsible for air pollution in this industrial region. Supplemental Materials: Supplemental materials are available for this paper. Go to the publisher's online edition of the Journal of the Air & Waste Management Association for (1) details of measurement campaigns; (2) CPF for each of the sources contributing to PM1–10; (3) factors contribution to PM1–10 resolved by PMF; (4) diurnal pattern of road dust, car brake factor in summer and winter; (5) trajectories during the marine aerosol episode in winter 2010; and (6) temporal temperature, concentration, and wind speed relationships during the summer 2008 campaign and winter 2010 campaign.