Role of organic aerosols in CCN activation and closure over a rural background site in Western Ghats, India

Abstract

The cloud condensation nuclei (CCN) closure study was performed to exemplify the effect of aerosol chemical composition on the CCN activity of aerosols at Mahabaleshwar, a high altitude background site in the Western Ghats, India. For this, collocated aerosol, CCN, Elemental Carbon (EC), Organic Carbon (OC), sub-micron aerosol chemical speciation for the period from 3rd June to 19th June 2015 was used. The chemical composition of non-refractory particulate matter (<1 μm) as measured by Time of Flight – Aerosol Chemical Speciation Monitor (ToF-ACSM) was dominated by organics with average concentration of 3.81 ± 1.6, 0.32 ± 0.06, 0.15 ± 0.02, 0.13 ± 0.03 and 0.95 ± 0.12 μg m−3 for organics, ammonium, chloride, nitrate and sulphate, respectively. The PM1 number concentration as obtained by Wide Range Aerosol Spectrometer (WRAS) varied from 750 to 6480 cm−3. The average mass concentration of elemental carbon (EC) as measured by OC-EC analyzer was 1.16 ± 0.4 μg m−3. The average CCN concentrations obtained from CCN counter (CCNC) at five super-saturations (SS's) was 118 ± 58 cm−3 (0.1% SS), 873 ± 448 cm−3 (0.31% SS), 1308 ± 603 cm−3 (0.52% SS), 1610 ± 838 cm−3 (0.73% SS) and 1826 ± 985 cm−3 (0.94% SS). The CCN concentrations were predicted using Köhler theory on the basis of measured aerosol particle number size distribution, size independent NR-PM1 chemical composition and calculated hygroscopicity. The CCN closure study was evaluated for 3 scenarios, B-I (all soluble inorganics), B-IO (all soluble organics and inorganics) and B-IOOA (all soluble inorganic and soluble oxygenated organic aerosol, OOA). OOA component was derived from the positive matrix factorization (PMF) analysis of organic aerosol mass spectra. Considering the bulk composition as internal mixture, CCN closure study was underestimated by 16–39% for B-I and overestimated by 47–62% for B-IO. The CCN closure result was appreciably improved for B-IOOA where the knowledge of OOA fraction was introduced and uncertainty reduced to within 8–10%.