Integrated Calendar

Supercooled clouds substantially impact polar surface energy budgets but large-scale models often underestimate their occurrence, which motivates accurately establishing metrics of basic processes. A polar stratiform cloud’s lifecycle is determined by a set of complex interactions and feedbacks between different micro-physical and macro-physical processes, some of which are not fully understood or quantified, leading to uncertainty in climate predictions. These polar clouds are commonly presupposed as being turbulent as a result of intense cloud-top longwave radiative cooling, while experiencing desiccation dominated by precipitating ice. In this talk, I examine some of these underlying assumptions and provide applicable guidance for large-scale model evaluation.
I first present observations of persistent formation of drizzle drops at cloud temperatures below -25 °C detected over McMurdo Station, Antarctica. These supercooled drizzle observations supported by large-eddy simulations (LES) used to examine the cloud’s formation and evolution under initially stable, nonturbulent conditions, suggest that drizzle can be common over polar regions and serve as the main cloud moisture sink even well below the freezing temperature. A persistent nonturbulent cloud state suggested by the LES leads to the examination of nonturbulent cloud occurrence in observational datasets from Arctic and Antarctic ground-based sites. Such stable, nonturbulent conditions, surmised to preferentially occur early in cloud lifecycles, are estimated to prevail in a quarter of cloud occurrences over these polar sites. I use LES sensitivity tests to examine how short to intermediate period gravity waves, which are supported by such stable conditions, may catalyze turbulence formation when aerosol particles available for activation are sufficiently small. The observational datasets are also utilized to examine ice precipitation processes, and show that the vast majority of polar supercooled clouds are at least weakly precipitating ice at the cloud base even when they are not seeded from above, consistent with commonly observed supercooled cloud longevity. These results indicate that supercooled cloud layers are a sustained source of ice precipitation, and suggest that ground-based statistics offer valuable guidance for large-scale models. Finally, as an example of how some of these observational and modeling results may be used to evaluate the representations of polar clouds in large-scale models, I briefly describe using the GISS E3 climate model in single-column model (SCM) mode applied to the supercooled drizzle case study.