Aerosol physical and chemical properties retrieved from ground-based remote sensing measurements during heavy haze days in Beijing winter

摘要

With the increase in economic development over the past thirty years, many largecities in eastern and southwestern China are experiencing increased hazeevents and atmospheric pollution, causing significant impacts on the regionalenvironment and even climate. However, knowledge on the aerosol physical andchemical properties in heavy haze conditions is still insufficient. In thisstudy, two winter heavy haze events in Beijing that occurred in 2011 and 2012 wereselected and investigated by using the ground-based remote sensingmeasurements. We used a CIMEL CE318 sun–sky radiometer to retrieve haze aerosoloptical, physical and chemical properties, including aerosol optical depth(AOD), size distribution, complex refractive indices and aerosol fractionsidentified as black carbon (BC), brown carbon (BrC), mineral dust (DU),ammonium sulfate-like (AS) components and aerosol water content (AW). Theretrieval results from a total of five haze days showed that the aerosolloading and properties during the two winter haze events were comparable.Therefore, average heavy haze property parameters were drawn to present aresearch case for future studies. The average AOD is about 3.0 at440 nm,and the Ångström exponent is 1.3 from 440 to 870 nm. The fine-mode AOD is2.8 corresponding to a fine-mode fraction of 0.93. The coarse particlesoccupied a considerable volume fraction of the bimodal size distribution inwinter haze events, with the mean particle radius of 0.21 and2.9 μm for the fine and coarse modes respectively. The real part ofthe refractive indices exhibited a relatively flat spectral behavior with anaverage value of 1.48 from 440 to 1020 nm. The imaginary part showedspectral variation, with the value at 440 nm (about 0.013) higher than the otherthree wavelengths (about 0.008 at 675 nm). The aerosol composition retrievalresults showed that volume fractions of BC, BrC, DU, AS and AW are 1,2, 49, 15 and 33%, respectively, on average for the investigatedhaze events. The preliminary uncertainty estimation and comparison of theseremote sensing results with in situ BC and PM measurements are alsopresented in the paper.