资源环境科技发展态势分析平台

Ice nucleation efficiencies, expressed in terms of either the active site density or the ice nucleation rate coefficient, are widely used to estimate ice formation rates in atmospheric models. Most estimates are, however, subject to bias since composition and surface area variation between particles is commonly neglected. This may amount to several orders of magnitude error in active site densities and introduce substantial error in cloud freezing temperatures impacting the accuracy of atmospheric models. Here it is shown that by performing droplet freezing experiments, varying mean particle surface area along with the temperature removes such a bias. The proposed method offers, for the first time, a solution to the long-standing problem of differentiating the "freezing rate" from the ice nucleation rate, or the apparent and the actual active site density of a material, and will likely improve the estimation of ice crystal formation rates in clouds.

Plain Language Summary At temperatures between 0 degrees C and approximately -37 degrees C clouds can exist as both liquid and ice. The fraction of condensate that remains liquid depends among other things on the presence of certain materials being immersed within, or in contact with, the cloud droplets, known as ice nucleation particles. Since, in a cloud, the particles impacting each droplet may differ in size and composition, they may induce freezing at slightly different temperatures. It has been difficult to account for this variation when estimating the rate of ice formation in a cloud. This work develops, for the first time, a method to eliminate such an error, and which will improve estimates of cloud droplet freezing in atmospheric models.