Abstract

Cloud droplets grow by the diffusion of water vapor before collisional growth turns cloud droplets into drizzle and rain drops. A simple model of droplet growth inside an adiabatic air parcel rising through a cloud gives extremely narrow unimodal droplet size distributions. At the same time, obse... Show moreCloud droplets grow by the diffusion of water vapor before collisional growth turns cloud droplets into drizzle and rain drops. A simple model of droplet growth inside an adiabatic air parcel rising through a cloud gives extremely narrow unimodal droplet size distributions. At the same time, observed distributions are typically wide and multimodal. Cloud turbulence and turbulent entrainment are often invoked to explain this discrepancy. This study investigates the mechanism affecting diffusional growth of cloud droplets referred to as the "large-eddy hopping" [1]. The key idea is that droplets arriving at a given location within a cloud follow different trajectories through a cloud. Variability of the supersaturation along those trajectories results in the width of the resulting droplet distribution. The supersaturation variability comes from turbulent eddies, often resulting from cloud-edge instabilities and entrainment. Cloud droplets "hop" these eddies and grow in response to fluctuations of the supersaturation. This mechanism was suggested a few decades ago [2], and was investigated using a cumbersome numerical approach in subsequent studies [3,4]. We are developing a novel numerical model that merges Eulerian approach for the cloud-scale flow with Lagrangian approach for cloud droplets following the "super-droplet" method of Shima et al. [5]. The key element of the model is the subgrid-scale (SGS) scheme that affects super-droplet motion and includes prediction of the SGS supersaturation fluctuations along the droplet trajectory in addition to the supersaturation predicted by the resolved cloud-scale flow model. A Langevin equation is formulated for the SGS velocity and supersaturation fluctuations and its discrete-in-time Monte Carlo solutions are used in the SGS scheme. Show less