The formation of atmospheric ice crystals by the freezing of droplets

TLDR

The study examines the freezing behavior of purified water droplets in cloud and rain conditions using laboratory supercooling experiments. Results indicate that below about −30 °C droplets freeze as pure water, whereas above −20 °C they freeze more readily, implying that ice nuclei are unnecessary below −20 °C but become significant above that temperature, with implications for cirrus, stratiform, and convective cloud freezing.

Abstract

Abstract Using the results of laboratory experiments on the supercooling of purified water, the freezing of cloud and raindrops is examined. It is shown that at temperatures lower than about −30°C in cold‐box or expansion‐chamber experiments, the drops freeze in approximately the numbers that would be expected if they were pure water. Those that freeze at temperatures higher than about −20°C seem to be more numerous than would be the case if the drops were pure. An interpretation of the main features of the Findeisen and Schulz expansion‐chamber experiments is found to be possible without appealing to the action of foreign ice‐forming nuclei, although there is a discrepancy between calculation and experiment at temperatures higher than about −20°C. In the atmosphere, formation of cirrus clouds is shown to become possible at temperatures below about −35°C, glaciation in stratiform clouds to become appreciable at temperatures below about −20°C and freezing of raindrops in strongly convective clouds to become important below about −13°C, without the presence of ice nuclei. It is concluded that freezing nuclei may be important at temperatures above about −20°C, while their presence at lower temperatures will be masked by the freezing of uncontaminated drops.