On the Nucleation of Ice in Highly Supersaturated Regions of Clouds

We simply but quantitatively analyze the hypothesis that the common observation, that ice crystal concentrations in clouds often far exceed measured ice nucleus concentrations, can be explained by our inability to reproduce cloud supersaturations during laboratory measurements of ice nuclei together with the fact that the probability of ice nucleation increases with supersaturation. We find the general hypothesis, while qualitatively appealing, is quantitatively unlikely. Especially unlikely is the particular hypothesis that the extremely high supersaturation regions that form around freezing drops contribute significantly to ice in clouds. The reasons for this are twofold: 1) the drop also warms the region of high supersaturation and 2) the region is very small compared to the regions of cloud with high ice concentrations.