Characterization of aerosol hygroscopicity, mixing state, and
CCN activity at a suburban site in the central North China Plain
Abstract. Aerosol hygroscopicity, mixing state and CCN activity were investigated as a part of the Atmosphere-Aerosol-Boundary Layer-Cloud (A2BC) Interaction Joint Experiment carried out at Xingtai (XT), a suburban site in the center of the North China Plain (NCP). In general, the probability density function of the hygroscopicity parameter (κ-PDF) for 40–200 nm particles had a unimodal distribution and mean κ-PDF patterns for different sizes were similar, suggesting that the particles were highly aged and internally mixed because of strong photochemical reactions. The κ calculated from the hygroscopic growth factor in the daytime and at nighttime showed that photochemical reactions largely enhanced the aerosol hygroscopicity, and the effect became weaker as the particle size increased. In addition, the aerosol hygroscopicity was much larger at XT than at sites in the northern part of the NCP, illustrating that the hygroscopicity of particles varies due to different emissions and chemical processes in the NCP. Measurement results also showed that new particle formation events occurred frequently at XT, one of the most polluted city in China. The evolution of the planetary boundary layer played a dominant role in aerosol mass concentration changes while particle formation and growth had a greater influence on the variation in aerosol number concentrations. Particle size was the most important factor influencing the ability of aerosols to activate, especially at higher levels of supersaturation (SS). The cloud condensation nuclei (CCN) number concentration (NCCN) derived from chemical composition was highly correlated with the measured NCCN (R2 ≥ 0.85), but was generally overestimated due to measurement uncertainties. The effect of chemical composition on NCCN was weaker relative to the particle size. NCCN sensitivity tests showed that the impact of chemical composition on NCCN became weaker with increasing SS, suggesting that chemical composition played a less role in NCCN estimations at higher SS levels. A good proxy for the chemical comical composition (κ = 0.31) was found, which can simplify the calculation of NCCN on models.