Deriving aerosol hygroscopic mixing state from size-resolved CCN activity and HR-ToF-AMS measurements. (October 2016)

• Record Type: Journal Article
• Authors: Bhattu, Deepika; Tripathi, S.N.; Chakraborty, Abhishek
• Is Part Of: Atmospheric environment. Volume 142(2016)
• Page Start: 57
• Page End: 70
• Publication Date: 2016-10
• Subjects: Accumulation -- Aitken mode -- Hygroscopic -- CCN-inactive -- Chemically-resolved; Air -- Pollution -- Periodicals; Air -- Pollution -- Meteorological aspects -- Periodicals; 551.51
• DOI: 10.1016/j.atmosenv.2016.07.032
• Languages: English
• ISSNs: 1352-2310

Abstract

Abstract: The ability of a particle to uptake water and form a cloud droplet depends on its hygroscopicity. To understand its impact on cloud properties and ultimately radiative forcing, knowledge of chemically-resolved mixing state information or the one based on hygroscopic growth is crucial. Typically, global models assume either pure internal or external mixing state which might not be true for all conditions and sampling locations. To investigate into this, the current study employed an indirect approach to infer the probable mixing state. The hygroscopic parameters derived from κ -Kohler theory using size-resolved CCN measurements ( κ CCN ) and bulk/size-resolved aerosol mass spectrometer (AMS) measurements ( κ AMS ) were compared. The accumulation mode particles were found to be more hygroscopic ( κ CCN = 0.24) than Aitken mode ( κ CCN = 0.13), perhaps due to increased ratio of inorganic to organic mass fraction. The activation diameter calculated from size-resolved CCN activity measurements at 5 different supersaturation (SS) levels varied in the range of 115 nm–42 nm with κ CCN = 0.13–0.23 (avg = 0.18 ± 0.10 (±1σ)). Further, κ AMS > κ CCN was observed possibly due to the fact that organic and inorganic mass present in the Aitken mode was not correctly represented by bulk chemical composition and size-resolved fractional contribution of oxidized OA was not accurately accounted. Better correlation of organic fraction (forg ) and κ CCN at lower SS explained this behaviour. The decrease in κ CCN with the time of the day was more pronounced at lower SS because of the relative mass reduction of soluble inorganic species by ∼17%. Despite the large differences between κ measured from two approaches, less over-prediction (up to 18%) between measured and predicted CCN concentration suggested lower impact of chemical composition and mixing state at higher SS. However, at lower SS, presences of externally mixed CCN-inactive aerosols lead to CCN over-prediction reflecting the significance of aerosol mixing state information. Further examination of the effect of biomass burning aerosols (∼35% in least oxidized biomass burning organic aerosol (BBOA-2 fraction) on hygroscopicity and CCN activity showed increase in the concentration of all AMS measured species (except NH4 + and SO4 2− ), less O:C ratio, and organic mass fraction (forg ) peak shift to lower diameter range caused ∼13% change in critical diameter (Da ) and ∼40% change in κ CCN . Increased deviation of ∼100% between κ CCN and κ AMS due to sudden influx of internally mixed BBOA caused suppressed hygroscopic growth. This study finally suggests the assumption of pure internally mixed aerosol does not completely hold true for this anthropogenically polluted site. Highlights: Accumulation mode particles are more hygroscopic and homogeneously mixed compared to Aitken mode. κ AMS is over-predicted compared to κ CCN as organic and inorganic mass in Aitken mode is not appropriately approximated by bulk composition. Increased deviation of ∼100% between κ CCN and κ AMS due to sudden influx of freshly emitted BBOA caused suppressed hygroscopic growth.