| It is well known that liquid droplets or solid particles in a turbulent gas tend to cluster outside of the vortices of the flow. The vortices "centrifuge" the particles out of their cores causing them to collect in the high strain regions of the flow. The resulting biased sampling of the flow they encounter affects single-particle statistics (e.g., particle accelerations, sedimentation, etc.) and multi-particle statistics (e.g., collision rates, dispersion, etc.) in profound ways. It has been hypothesized that these modifications in the dynamics of the particles can play an important role in initiating warm rain in cumulus clouds. Using a combination of direct numerical simulations, laboratory experiments and theory, we have analyzed the clustering phenomenon. In this talk, we will present detailed comparisons between experiments and simulations. We also will present estimates of the role of these effects on droplet growth at parameter values similar to those found in cumulus clouds. |
|