Assessment of cloud condensation nucleus activation of urban aerosol particles with different hygroscopicity and the application to the cloud parcel model
Assessment of cloud condensation nucleus activation of urban aerosol particles with different hygroscopicity and the application to the cloud parcel model
复制标题
DOI：
10.1002/2013jd020827
复制
发表时间：
2014-03
期刊：
影响因子：
--
通讯作者：
K. Kawana;N. Kuba;M. Mochida
中科院分区：
文献类型：
--
作者：
K. Kawana;N. Kuba;M. Mochida
对日本名古屋上空的城市气溶胶进行了尺寸分辨测量，测量了具有不同吸湿生长因子(g)的粒子的云凝结核（CCN）与凝结核的比率，以及相对湿度为85%时g的分布。吸湿性较弱的颗粒（g为1.0和1.1）与吸湿性较强的颗粒（g为1.25和1.4）的CCN效率谱差异很大。而由g （dact,g85）预测的CCN活化直径与实测值（dact，CCN）之间的差异在12%以内，对于吸湿性较强的颗粒，差异较大（16%-41%）。可能的原因包括表面张力的降低，κ对溶液浓度的依赖，少溶物质的存在，以及颗粒的非球形。利用云包模型从g的分布估计了CCN （NCCN）和云滴（Ncd）的数量浓度和云滴的有效半径（Reff）。dact、g85和dact、CCN之间的差异以及CCN无活性颗粒的存在对模型评估的影响较小。与只存在更多吸湿性颗粒的假设情况相比，在高上升气流速度下，将更少和更多吸湿性颗粒纳入模式导致NCCN和Ncd大幅增加，Reff减少。结果表明，吸湿性较小的颗粒对云滴的形成有显著的促进作用，对g分布的评估在这方面是有用的。
Size‐resolved measurements of the ratios of cloud condensation nuclei (CCN) to condensation nuclei for particles with different hygroscopic growth factors (g) and distributions of g at 85% relative humidity were performed for urban aerosols over Nagoya, Japan. The CCN efficiency spectra of less hygroscopic particles (g of 1.0 and 1.1) were very different from those of more hygroscopic particles (g of 1.25 and 1.4). While the differences between the CCN activation diameters predicted from g (dact,g85) and those measured (dact,CCN) were within 12% for more hygroscopic particles, the differences were larger (16%–41%) for less hygroscopic particles. Possible causes of this included surface tension reduction, the dependence of κ on the concentration of the solution, the existence of sparingly soluble materials, and asphericity of particles. The number concentrations of CCN (NCCN) and cloud droplets (Ncd) and the effective radius of cloud droplets (Reff) were estimated from the distributions of g using a cloud parcel model. The influences of the differences between dact,g85 and dact,CCN and the existence of CCN‐inactive particles on the model assessment were small. With high updraft velocity, incorporating both less and more hygroscopic particles into the model led to substantial increases in NCCN and Ncd and a decrease in Reff as compared to the hypothetical cases that only more hygroscopic particles were present. The results indicated that less hygroscopic particles significantly contribute to cloud droplet formation and assessments of g distributions are useful in this regard.