Performance of a Triclass Parameterization for the Collision-Coalescence Process in Shallow Clouds

Focusing on the formation of precipitation in marine stratiform clouds, a two-moment bulk parameterization for three liquid water classes (cloud, drizzle, and rain) is proposed to describe the process of collision-coalescence. Based on the stochastic collection equation and making use of partial moments to improve the physical representation of the shape of the drop size distribution, new rate equations for both number and mass densities are derived using the modified gamma distribution and an adapted collection kernel. Based on observations and spectral model results, the free shape parameters of the modified gamma distribution of each class are determined closing the set of equations. Idealized simulations of the new parameterization compare well to other studies and prove that the closure assumptions are appropriate, especially as the rate equations are invariant under time-stretching transformations-a key property of the stochastic collection equation. The framework of the one-dimensional kinematic cloud model is used to compare the new bulk parameterization to existing formulations and a spectral model. These results show a good agreement, especially in the sensitivity to the aerosol background concentration and the general development for updraft velocities relevant for shallow clouds. Furthermore, as drizzle dominates the formed precipitation for stratocumulus it becomes a pure transition class for more convective type clouds. The analysis reveals a different quantitative behavior of the various parameterizations in the drizzle regime, which is of special importance for precipitating stratocumulus clouds.