Three-dimensional nonhydrostatic simulations of summer thunderstorms in the humid subtropics versus High Plains

Abstract

This article presents a detailed comparison of cloud microphysical evolution among six warm-season thunderstorm simulations using a time-dependent three-dimensional model WISCDYMM. The six thunderstorms chosen for this study consist of three apiece from two contrasting climate zones, the US High Plains (one supercell and two multicells) and the humid subtropics (two in Florida, US and one in Taipei, Taiwan, all multicells). The primary goal of this study is to investigate the differences among thunderstorms in different climate regimes in terms of their microphysical structures and how differently these structures evolve in time. A subtropical case is used as an example to illustrate the general contents of a simulated storm, and two examples of the simulated storms, one humid subtropical and one northern High Plains case, are used to describe in detail the microphysical histories. The simulation results are compared with the available observational data, and the agreement between the two is shown to be at least fairly close overall. The analysis, synthesis and implications of the simulation results are then presented. The microphysical histories of the six simulated storms in terms of the domain-integrated masses of all five hydrometeor classes (cloud water, cloud ice, rain, snow, graupel/hail), along with the individual sources (and sinks) of the three precipitating hydrometeor classes (rain, snow, graupel/hail) are analyzed in detail. These analyses encompass both the absolute magnitudes and their percentage contributions to the totals, for the condensate mass and their precipitation production (and depletion) rates, respectively. Comparisons between the hydrometeor mass partitionings for the High Plains versus subtropical thunderstorms show that, in a time-averaged sense, ice hydrometeors (cloud ice, snow, graupel/hail) account for ∼ 70–80% of the total hydrometeor mass for the High Plains storms but only ∼ 50% for the subtropical storms, after the systems have reached quasi-steady mature states. This demonstrates that ice processes are highly important even in thunderstorms occurring in warm climatic regimes. The dominant rain sources are two of the graupel/hail sinks, shedding and melting, in both High Plains and subtropical storms, while the main rain sinks are accretion by hail and evaporation. The dominant graupel/hail sources are accretion of rain, snow and cloud water, while its main sinks are shedding and melting. The dominant snow sources are the Bergeron-Findeisen process and accretion of cloud water, while the main sinks are accretion by graupel/hail and sublimation. However, the rankings of the leading production and depletion mechanisms differ somewhat in different storm cases, especially for graupel/hail. The model results indicate that the same hydrometeor types in the different climates have their favored microphysical sources and sinks. These findings not only prove that thunderstorm structure depends on local dynamic and thermodynamic atmospheric conditions that are generally climate-dependent, but also provide information about the partitioning of hydrometeors in the storms. Such information is potentially useful for convective parameterization in large-scale models.