The Spectral Ice Habit Prediction System (SHIPS). Part III: Description of the Ice Particle Model and the Habit-Dependent Aggregation Model

The purpose of this paper is to describe a numerical scheme of the Spectral Ice Habit Prediction System (SHIPS) that simulates the dependency of aggregation process explicitly on crystal habit and size in cloud-resolving models (CRMs). The sizes and shapes of ice crystals are known to modulate the aggregation process, which is a critical part of physical processes leading to precipitation in addition to vapor deposition and riming processes. A problem with conventional formulation of aggregation process in CRMs is that it is not designed to predict the aggregates’ properties based on the information on crystals. To simulate such dependency, SHIPS solves a quasi-stochastic model that describes growth tendency for a group of particles, together with particle property variables (PPVs) that carry information on habit and types of ice particles. SHIPS diagnoses the ice particle properties based on the PPVs for each mass bin at given a time and space, which are used to calculate the collision cross-sectional area and terminal velocity differences based on crystal habits explicitly. To achieve prediction of properties of aggregates and rimed particles, SHIPS introduces 1) the use of a conceptually based, circumscribing shape, called the ice particle model, and 2) the explicit prediction of the circumscribing sphere volume based on a simple growth model of the maximum dimension. Based on these properties, SHIPS is able to predict the mass–dimensional relationships of aggregates so that they are physically consistent with the growth history of the particles. In addition, the information about the crystals making up the aggregates is predicted. Part IV of this series will present the results and evaluation of the application of this formulation to idealized tests in a Lagrangian “box model” setup.