On the Production of Large Ice Number Concentrations in Maritime Cumuli

A long-standing challenge in cloud physics has been to explain high number concentrations of ice in tropical maritime clouds with cloud tops warmer than -15 °C. Some studies have implicated the Hallet-Mossop (rime-splintering) mechanism, while others have suggested raindrop shattering, desert dust intrusions, or even questioned the reported ice crystal number concentrations.

Three days of in situ observations of ice nuclei and cloud properties from the ICE-T field campaign, which took place in the Caribbean in 2011, are used to constrain high-resolution 3D simulations of a tropical maritime cumulus congestus. Desert dust was absent on these particular days. Our goal is to see if the occasional occurrence of ice crystal number concentrations exceeding 100 per liter, with cloud tops warmer than -15°C, can be explained by observed ice nuclei and known secondary nucleation mechanisms.

Even when rime-splintering is included, the model is unable to replicate this high number concentration of ice crystals without greatly increasing the number of ice nuclei active at temperatures lower than -8 °C. The number of ice nuclei required is orders of magnitude greater than that observed. New simulations including effects such as raindrop shattering, and variations in the number of ice nuclei active over particular temperature intervals, will be examined to see if the observed ice number concentrations in excess of 100 per liter can be explained.