Abstract
In a recent paper on the effect of turbulence on droplet collisions in clouds, Woods, Drake and Goldsmith (1972) proposed that the principal factor was the strong spectral peak of shear near the Kolmogoroff microβscale. This hypothesis led to an important simplification, namely that spatial and temporal fluctuations in the turbulent shear field could be ignored during an encounter between a pair of cloud droplets. However, the authors pointed out that their simple model of (isotropic, homogeneous) turbulence contained an assumption, namely that the shears are normally distributed, which would have to be reβexamined if it turned out that droplet collision efficiency increased rapidly after a threshold value is exceeded. Subsequent measurements by Jonas and Goldsmith (1972) showed that this is indeed the case. In this paper we reconsider the efficiency of turbulent coalescence in the light of the new experimental data, and taking account of the intermittency of shear distribution. It is concluded that significant increases of collection efficiency will occur in clouds which are only weakly turbulent. For example, the collection efficiency for droplets of radii R = 20 ΞΌm and Ξ³ = 9 ΞΌm in a cloud with energy dissipation rate Ξ΅ = 55Β·5 cm^2^ s^β3^ will be approximately 14 per cent compared with 2 per cent for the same droplets falling though still air. The shear zone in the wind tunnel experiment designed by Woods et al. (1972) is a realistic approximation to the shear zones in cloud turbulence.