Efficiency of cloud condensation nuclei formation from ultrafine particles

[摘要] Atmospheric cloud condensation nuclei (CCN) concentrations are a keyuncertainty in the assessment of the effect of anthropogenic aerosol onclouds and climate. The ability of new ultrafine particles to grow to becomeCCN varies throughout the atmosphere and must be understood in order tounderstand CCN formation. We have developed the Probability of Ultrafineparticle Growth (PUG) model to answer questions regarding which growth andsink mechanisms control this growth, how the growth varies between differentparts of the atmosphere and how uncertainties with respect to the magnitudeand size distribution of ultrafine emissions translates into uncertainty inCCN generation. The inputs to the PUG model are the concentrations ofcondensable gases, the size distribution of ambient aerosol, particle depositiontimescales and physical properties of the particles and condensable gases.It was found in most cases that condensation is the dominantgrowth mechanism and coagulation with larger particles is the dominant sinkmechanism for ultrafine particles. In this work we found that theprobability of a new ultrafine particle generating a CCN varies from<0.1% to ~90% in different parts of the atmosphere, though in theboundary layer a large fraction of ultrafine particles have a probabilitybetween 1% and 40%. Some regions, such as the tropical freetroposphere, are areas with high probabilities; however, variability withinregions makes it difficult to predict which regions of the atmosphere aremost efficient for generating CCN from ultrafine particles. For a given massof primary ultrafine aerosol, an uncertainty of a factor of two in the modaldiameter can lead to an uncertainty in the number of CCN generated as highas a factor for eight. It was found that no single moment of the primaryaerosol size distribution, such as total mass or number, is a robustpredictor of the number of CCN ultimately generated. Therefore, a completedescription of the emissions size distribution is generally required for globalaerosol microphysics models.

[学科分类] 大气科学