Evaluation of riming in the P3-2ice Microphysics Scheme: A case study of Cold-Air Outbreak Snowfall in Eastern China

Using the multi-frequency radar suite, ground-based observations, and the Weather Research and Forecasting (WRF) model with the Predicted Particle Properties microphysics scheme with the version of two-ice category (P3-2ice), we study the characteristics of a cold-air outbreak snowfall event on 6-7 December 2024. The cloud system is shallow convection, with the microphysical characteristics differing between the first and second snowfall periods. The WRF model reproduces the synoptic-scale weather system and mesoscale structure well, but underestimates the event-averaged snow intensity and rime mass fraction by approximately 52% and 24%, respectively. Within the riming process, the collection of cloud water by ice is the dominant pathway. However, the fixed collection efficiency decouples riming from wind shear. To study the impact of collection efficiency on snow intensity, we increase the efficiency as a simplified approach to represent the potential effect of shear‑generated turbulence on riming. The results of sensitivity experiments reveal that the snow intensity exhibits a positive correlation with collection efficiency, but with contrasting behaviors between the two snowfall periods. In the first period, persistent severe underestimation occurs; in the second period, overestimation occurs when efficiency is high, and the rime mass fraction shows a nonlinear response with efficiency. These results indicate that a fixed collection efficiency is inadequate and a dynamic environment‑constrained efficiency is required. The research provides new insights into how to improve the riming process parameterization, contributing to a better understanding of the dynamic and microphysical characteristics of cold-air outbreak snowfall. Plain Language Summary Cold-air outbreak snowfall is produced when the cold air flows over a relatively warm water surface. Snowfall involves a series of complex processes of the mass growth of ice phase particles, in which riming—the collection of supercooled water by falling ice crystals—plays an important role in the surface snow mass. However, few studies have diagnosed the simulation ability of the numerical model for cold-air outbreak snowfall events dominated by the riming process. In this study, by using the multi-frequency radar suite, ground-based observations, and the WRF model with the Predicted Particle Properties microphysics scheme with the version of two-ice category (P3-2ice), we show that the model can well capture the characteristics of weather system and mesoscale structure corresponding to the shallow convective cold-air outbreak snowfall on 6-7 December 2024, but significantly underestimate the snow intensity and rime mass fraction. The enhanced collection efficiency between ice and cloud droplets still results in severe underestimation of snow intensity in the first snowfall period, but leads to overestimation in the second period when efficiency is high. This study serves as a useful reference to improve the parameterization of the riming process, which provides new implications for the forecasting of cold-air outbreak snowfall.