Using a wide-range particle spectrometer (WPS), an environmental management system (EMS), KC-120H middle volume sampler, a 850 professional ion chromatography analyzer, and heat/light carbon analyzer (DRI2001A), we observed the number concentration of aerosols with sizes ranging from 10 nm to 10 μm, gas concentrations, and concentrations of PM2.5, water-soluble ions, OC, and EC in a Lin'an atmospheric background station from January 9 to 31, 2015. The positive matrix factorization (PMF) model was applied for source apportionment, and the size distribution and diurnal variations of emission sources were analyzed based on the meteorological data. The average aerosol concentration was 5062 cm-3·nm-1 and PM2.5 mass concentration was 123.6 μg·m-3. The average concentrations of NO3-, SO42-, and NH4+, the main water-soluble ions in PM2.5 were 19.2, 15.4, and 10.8 μg·m-3, which accounted for 37.9%, 30.4%, and 21.4% of total water-soluble ions, respectively. Theaverage concentrations of OC and EC were 24.4 μg·m-3 and 6.6 μg·m-3. Secondary aerosol formation, coal combustion, motor vehicle emissions, dust, andbiomass burning were the main sources of PM2.5 in Lin'an during winter with contributions of 42.3%, 21.4%, 17.1%, 8.7%, and 10.6%, respectively. Different sources had different aerosol number concentration distributions. The aerosol number concentration spectra of secondary sources, vehicle emissions, dust, and biomass burning followed unimodal-type distributions with peaks at 120, 50, 100, and 90 nm. Coal particle number concentration was a bimodal distribution which exhibited peak values at 25 nm and 100 nm (19842 cm-3·nm-1 and 18372 cm-3·nm-1, respectively). The spectra of surface concentrations of secondary sources, coal combustion, motor vehicle emissions, dust, and biomass burning followed a three-peak distribution. The peaks were at 650, 210, 160, 180, and 575 nm. The diurnal variations of particle number concentrations influenced by diurnal variations in the boundary layer and human activities were consistent with the variations in surface concentrations, which displayed bimodal-type distribution.
Keywords: PM2.5; diurnal variation; positive matrix factorization (PMF); size distribution; source apportionment.