What Controls Crystal Diversity and Microphysical Variability in Cirrus Clouds?

Variability of ice microphysical properties like crystal size and density in cirrus clouds is important for climate through its impact on radiative forcing, but challenging to represent in models. For the first time, recent laboratory experiments of particle growth (tied to crystal morphology via deposition density) are combined with a state-of-the-art Lagrangian particle-based microphysics model in large-eddy simulations to examine sources of microphysical variability in cirrus. Simulated particle size distributions compare well against balloon-borne observations. Overall, microphysical variability is dominated by variability in the particles' thermodynamic histories. However, diversity in crystal morphology notably increases spatial variability of mean particle size and density, especially at mid-levels in the cloud. Little correlation between instantaneous crystal properties and supersaturation occurs even though the modeled particle morphology is directly tied to supersaturation based on laboratory measurements. Thus, the individual thermodynamic paths of each particle, not the instantaneous conditions, control the evolution of particle properties. Thin, high-altitude ice clouds-cirrus-are critical to climate through their interactions with incoming solar radiation and outgoing longwave radiation. With a zoomed-in view, observations show that cirrus comprise ice crystals of diverse sizes and shapes. These crystal properties determine the bulk properties of the clouds, like their ice water content. What controls this diversity of crystal properties? To address this question, we performed numerical simulations using a state-of-the-art model that tracks the growth of individual particles, informed by laboratory crystal growth experiments, in a realistic turbulent cloud flow. We show that variability in the local environmental conditions (e.g., relative humidity) that particles experience in a turbulent environment is a primary driver of crystal diversity in cirrus. Diversity is also driven by different shapes of newly formed crystals, leading to different growth characteristics. This diversity drives spatial variability of properties like mean crystal density across the cloud layer. Climate models that represent cirrus using simple relationships based on observations that are essentially snapshots of average crystal properties cannot capture this variability. Overall, our analysis indicates that the history of environmental conditions along particle trajectories, not just instantaneous conditions, are required to understand crystal growth and variability in cirrus. Variability in the thermodynamic histories of ice crystals are a primary driver of particle variability in cirrus Diversity in crystal morphology also drives spatial variability of mean particle size and density, particularly at mid-levels in the cloud Thermodynamic histories along particle trajectories, not just local conditions, are required to understand crystal growth in cirrus