Utilization of a multi-phase particle model to develop self-consistent bulk microphysical parameterizations for hurricane models

A bulk cloud model typically contains various processes describing diffusional growth, collision, melting, freezing, and activation of various types of cloud particles. Though each process may reasonably represent what occurs in nature, the sum of all the processes may not accurately symbolize reality nor be numerically sound, i.e., the smallest numerical time scale produced by the bulk cloud model is significantly smaller than the sound wave time scale. Further, because of the inherent assumptions found within a bulk cloud model, the model may only be appropriate for a given realization, such as a hurricane, and may have to be “retuned” depending on the new application. Before addressing the issue of whether or not a self-consistent bulk cloud model can be used for a variety of realizations, a self-consistent cloud model was developed for use in a single application, i.e., for modeling hurricanes. Specifically, the various processes found in the new bulk cloud model will be formulated by averaging in time and space statistics from a Monte-Carlo like multi-phase particle model run for select hurricanes. Because the statistics, such as the collision rate between snow and rain particles, are obtained from a higher-order model in which all microphysical processes are occurring simultaneously, the end result should be a consistent bulk model that induces reasonable time scales and ultimately better forecasts of convective phenomena embedded within a hurricane. In this presentation, details concerning the multi-phase particle model and the averaging of results from the particle model towards development of a consistent bulk model will be illustrated along with how this procedure could be used to refine bulk cloud models used in operational hurricane models.