Characterization and source apportionment of aerosol light extinction in Chengdu, southwest China

•Aerosol optical and chemical properties in an urban environment were characterized.•Aerosol hygroscopic curves were developed based on field measurements.•Source apportionment of aerosol light extinction was estimated.

To investigate aerosol properties in the Sichuan Basin of China, field aerosol sampling was carried out in Chengdu, China during four one-month periods, each in a different season in 2011. Aerosol scattering coefficient (bsp) at dry (RH<40%) and wet (40% < RH<90%) conditions and aerosol absorption coefficient (bap) were measured. Additionally, daily PM2.5 and PM10 samples were also collected. PM2.5 samples were subject to chemical analysis for various chemical components including major water-soluble ions, organic and elemental carbon (OC and EC), trace elements, as well as anhydrosugar Levoglucosan (LG) and Mannosan (MN). A multiple linear regression analysis was applied to the measured dry bsp against (NH4)2SO4, NH4NO3, organic mass (OM), fine soil (FS), and coarse mass (CM, PM2.5-10), and to the measured bap against EC in all the four seasons to evaluate the impact of individual chemical components of PM2.5 and CM on aerosol light extinction (bext = bsp + bap). Mass scattering efficiency (MSE) and mass absorption efficiency (MAE) of the individual chemical components of PM2.5 were estimated based on seasonal regression equations and were then used for estimating bext. The annual bsp, bap and single scattering albedo (SSA) at dry conditions were 456 ± 237 Mm−1, 96 ± 48 Mm−1 and 0.82 ± 0.05, respectively. The annual average bsp at ambient conditions estimated through hygroscopic curve of aerosol (f(RH)) was 763 ± 415 Mm−1, which was 1.7 times of the dry bsp. The annual average SSA at ambient conditions also increased to 0.88 ± 0.04. The estimated dry bext was only 2 ± 9% higher than the measurements and the estimated ambient bext from individual chemical components was only 1 ± 10% lower, on an annual basis, than that estimated from using f(RH). Secondary inorganic aerosols, coal combustion, biomass burning, iron and steel industry, Mo-related industry, soil dust, and CM to bext were estimated to account for 41 ± 19%, 18 ± 12%, 14 ± 13%, 13 ± 11%, 5 ± 4%, 5 ± 7% and 4 ± 3%, respectively, of the estimated ambient bext.