The Geometry of Rimed Aggregate Snowflakes: A Modeling Study

Computer simulations of the aggregation and riming of ice crystals are performed to investigate the geometry of rimed aggregate snowflakes. Due to the universality of the geometry of aggregates, the conversion to a graupel‐like particle is self‐similar and independent from specific properties of the aggregate, when formulated in properly normalized variables. Hence, the particle habit of the primary crystals, the size of the aggregate or the density of the rime mass, does not lead to a structural change in the transition to graupel. Therefore, this transition can be parameterized by a similarity approach using simulations of many individual rimed aggregates. These parameterizations can replace the classic fill‐in model used in many cloud models. The parameterizations are applied and tested in a one‐dimensional Lagrangian superparticle model to simulate the growth of aggregates in a liquid layer. We find that the similarity model for the geometry of rimed snowflakes leads to a more rapid growth by riming and, hence, an increased precipitation rate compared to the fill‐in model. The main reason for this is that the increase of the maximum dimension during the early stages of riming is properly taken into account by the similarity model, whereas it is neglected by the fill‐in model.