Seasonal variations and chemical characteristics of sub-micrometer particles (PM ₁) in Guangzhou, China

Daily samples of ambient sub-micrometer particles (PM ₁, particles with an aerodynamic diameter≤1.0μm) were collected from July 2009 to April 2010 at an urban site over Guangzhou in southern China. Mass concentrations of water-soluble inorganic ions, organic carbon (OC) and elemental carbon (EC) were determined to characterize the chemical composition of PM ₁. The mass concentration of PM ₁ ranged from 14.6μgm ^-3 to 143.3μgm ^-3, with an annual mean value of 52.4±27.3μgm ^-3. Seasonally-averaged PM ₁ concentrations decreased in the order winter>autumn>spring>summer. The annual mean concentrations of OC and EC were 6.2±3.5 and 5.0±2.9μgm ^-3, respectively. The OC and EC concentrations were measured following the IMPROVE_A thermal/optical reflectance (TOR) protocol. Total carbonaceous aerosol (the sum of organic matter and elemental carbon) accounted for 23.0±4.4% of PM ₁ mass. Clear seasonal variations in OC and EC suggested sources of these two constituents were remarkable difference among the four seasons. Seasonally averaged OC/EC ratios were 1.2, 1.7, 1.4, and 1.5, from spring to winter respectively. Low OC/EC ratios in comparison with other cities in China revealed that vehicle emissions play an important role in carbonaceous aerosol levels in Guangzhou. SO ₄ ^2-, NO ₃ ^- and NH ₄ ⁺ were the three major inorganic ions in PM ₁, collectively contributing 30.0%±6.3% of the PM ₁ mass. SO ₄ ^2- and NH ₄ ⁺ were both the highest in autumn and the lowest in summer. In contrast, NO ₃ ^- was the highest in winter. Sulfur oxidation ratio was positively correlated with solar radiation and O ₃, but negatively correlated with SO ₂. Nitrogen oxidation ratio was positively correlated with NO ₂, NH ₄ ⁺ and Cl ^-, but showed a negative correlation with temperature. By applying the IMPROVE equation, PM ₁ mass was reconstructed and showed that (NH ₄) ₂SO ₄, NH ₄NO ₃, OM and EC accounted for (30.7±11.4) %, (9.7±5.2) %, (22.6±5.0) % and (9.7±2.3) % of PM ₁, respectively. Finally, source apportionment by positive matrix factorization revealed that (1) secondary aerosol and biomass burning, (2) diesel emissions, (3) gasoline emissions and sea salt, and (4) coal combustion were the greatest contributors to PM ₁.