Assessing the relative importance of the prognostic hail mixing ratio and predicted graupel density on the vertical reflectivity structure in bulk cloud microphysics schemes

This study compares the performance of two versions of the Weather Research and Forecasting (WRF) Double-Moment 6-class (WDM6) microphysics scheme—one additionally incorporating the prognostic hail mixing ratio and the other additionally including the predicted graupel density—in terms of cloud morphology based on Contoured Frequency by Altitude Diagrams (CFADs) of radar reflectivity and simulated precipitation. The WDM6 scheme incorporating the predicted graupel density is based on the scheme of Park et al. (2024). The WDM6 scheme incorporating a hail category is newly developed in this study. The new hail properties include the size distribution, fall velocity–diameter relationships, mass–diameter relationships, and density with relevant hail microphysical processes. Thirteen precipitation events—including winter snowfall, summer rainfall, and hail—are evaluated to investigate the effect of the prognostic hail mixing ratio and predicted graupel density on the vertical distribution of radar reflectivity and simulated surface precipitation. The CFAD analysis reveals that, in most cases, the scheme with the predicted graupel density better simulates the reflectivity patterns compared with observational data, while the scheme with a hail category tends to overestimate the frequency of strong reflectivity due to the hail generation; additionally, the latter scheme simulates unrealistically high reflectivity of up to 60 dBZ. The analysis of vertical profiles of mixing ratios and associated microphysical processes indicates that, in the scheme incorporating the predicted graupel density, the reduction in the graupel mixing ratio is a key factor in improving the simulated radar reflectivity. Although the revised scheme incorporating the hail mixing ratio shows a reduction in the graupel mixing ratio, the incorporation of the hail mixing ratio leads to the overestimation of radar reflectivity compared to the observed values. These findings suggest that representing the properties of solid-phase hydrometeors is more important for simulating realistic vertical radar-reflectivity profiles, which implies the simulation of more realistic microphysical processes, than increasing the number of solid-phase hydrometeor categories.