The Mystery of Ice Crystal Multiplication in a Laboratory Experiment

This paper addresses the problem of the large discrepancies between ice crystal concentrations in clouds and the number of ice nuclei in nearby clear air reported in published papers. Such discrepancies cannot always be explained, even by taking into account both primary and secondary ice formation processes. A laboratory experiment was performed in a cylindrical column placed in a cold room at atmospheric pressure and temperature in the −12° to −14°C range. Supercooled droplets were nucleated in the column, in the absence of aerosol ice nuclei, by injecting ice crystals generated outside in a small syringe. A rapid increase in the ice crystal concentration was observed in the absence of any known ice multiplication. The ratio between the mean number of ice crystals in the column, after complete droplet vaporization, and the number of ice crystals introduced in the column was about 10:1. The presence of small ice crystals (introduced at the top of the column) in the unstable system (supercooled droplets) appears to trigger the transformation in the whole supercooled liquid cloud. A possible explanation could be that the rapidly evaporating droplets cool sufficiently to determine a homogeneous nucleation.