Model Predictability of Hail Precipitation with a Moderate Hailstorm Case. Part II: Impact of Improved Treatment of Graupel in the Thompson–Eidhammer Microphysics Scheme

Abstract

Abstract Explicit simulation of hailstorms remains a challenge partly due to the lack of accurate representations of both initial conditions and microphysical processes. Using a moderate hailstorm case that occurred in Beijing on 10 June 2016, the impact of the initial conditions on explicit hail prediction has been studied in Part I of this two-part work via high-resolution data assimilation. This Part II paper examines the role of improved graupel treatment by comparing the recently upgraded Thompson–Eidhammer microphysics scheme (MP38) with two previous versions. MP38 is a double-moment hail-aware scheme with the ability to additionally predict the graupel number concentration and density. This case study showed that the addition of these predictive variables improved the simulation of the mass-weighted mean diameter of hail and thereby reduced the overestimation of hail size. However, the hail size was significantly underpredicted without the prediction of hail density, indicating that both quantities must be prognosed for skillful hail prediction. It was further shown that the revised graupel treatment also influenced hailstorm dynamics. The smaller hail size in MP38 led to a stronger graupel melting process, which further promoted a stronger cold pool and downdraft. By assessing the efficiency of the upgraded Thompson–Eidhammer microphysics scheme, the current study shed some light on the importance of the accurate representation of microphysical processes in numerical models for explicit hailstorm prediction.