| This dissertation concerns effects of air turbulence on the collision rate of atmospheric cloud droplets. This research was motivated by the speculation that air turbulence could enhance the collision rate thereby help transform cloud droplets to rain droplets in a short time as observed in nature. The air turbulence within clouds is assumed to be homogeneous and isotropic, and its small-scale motion (1 mm to 10 cm scales) is computationally generated by direct numerical integration of the full Navier-Stokes equations. Typical droplet and turbulence parameters of convective warm clouds are used to determine the Stokes numbers (St) and the nondimensional terminal velocities (Sv) which characterize droplet relative inertia and gravitational settling, respectively.; A novel and efficient methodology for conducting direct numerical simulations (DNS) of hydrodynamically-interacting droplets in the context of cloud microphysics has been developed. This numerical approach solves the turbulent flow by the pseudo-spectral method with a large-scale forcing, and utilizes an improved superposition method to embed analytically the local, small-scale (10 mum to 1 mm) disturbance flows induced by the droplets. This hybrid representation of background turbulent air motion and the induced disturbance flows is then used to study the combined effects of hydrodynamic interactions and airflow turbulence on the motion and collisions of cloud droplets.; Hybrid DNS results show that turbulence can increase the geometric collision kernel relative to the gravitational geometric kernel by as much as 42% due to enhanced radial relative motion and preferential concentration of droplets. The exact level of enhancements depends on the Taylor-microscale Reynolds number, turbulent dissipation rate, and droplet pair size ratio. One important finding is that turbulence has a relatively dominant effect on the collision process between droplets close in size as the gravitational collision mechanism diminishes. A theory was developed to predict the radial relative velocity between droplets at contact. The theory agrees with our DNS results to within 5% for cloud droplets with strong settling. In addition, an empirical model is developed to quantify the radial distribution function. (Abstract shortened by UMI.)... |
