Impacts Of Drop Size Distribution Characteristics On Tornadogenesis In A Simulated Supercell Over Eastern China

Tornado is an extremely severe convective system with violently rotating destructive winds. Mostly, the strongest tornados are produced within a supercell thunderstorm. Eastern China is one of the favorable regions for the occurrence of tornadoes. Because of the small scale and short life cycle, it is very difficult to detect and even harder to predict the tornado, which will definitely cause huge losses of both life and wealth. Currently, the formation mechanism of the tornadic supercell storms over eastern China is still poor understood. Besides that, how will the cloud microphysics affect tornadogenesis is not clear and deserve detailed investigation.In order to explore the formation mechanisms of tornadic supercell storms and the impacts of drop size distributions (DSDs) on tornadogenesis over eastern China, numerical simulations of a subtropical supercell thunderstorm occurred in Wuwei, Anhui Province, July 8th 2003 (Ju108 case), were first performed with the Advanced Regional Prediction System (ARPS) model. Sensitivity experiments with various DSDs were conducted to investigate the cloud microphysical impacts on the tornadogenesis. The present results were also compared with previous studies focusing on a U.S. great plain case. Main results of this study are summarized as follows:The evolution of an organized convective rainband embodied in a large area of stratiform precipitation during Meiyu period is well reproduced in the 3-km grid real-data simulation. Although some biases exist, the model captures the main features of this mesoscale convective system when it moved eastward across the Anhui Province. Applying the model extracted sounding from the real-data simulation, idealized simulation at 100-m grid could successfully produce the target tornadic supercell thunderstorm featured with obvious hook echo, mesocyclone, gust front, bounded weak echo region, and so on. Moreover, a tornadic vortex is formed and lasts for 14 min at the occlusion point of the forward and rear flank gust front when the storm starts to dissipate. The maximum low-level vertical vorticity and wind are 0.39 s-1 and 45.7 m s-1 (EF1 rank). Trajectory analyses emphasize the importance of the rear flank downdraft region on tornadogenesis.Idealized sensitivity experiments with different intercept parameters of rain, hail and snow DSDs in the Lin-type microphysics scheme were conducted to further study the DSDs impacts on the downdraft microphysical processes and tornadogenesis. Results show that DSDs with larger (smaller) intercepts which favorable for smaller (larger) particles would result relatively stronger (weaker) and wider (narrower) Surface Cold Pools (SCP) through enhanced (reduced) rain evaporation and hail melting processes in downdraft region. Compared with the hail DSD, the rain DSD affects the rainwater evaporation at lower levels, thus more important for the SCP intensity. The configuration between the outflow and the inflow is found to be critical to tornadogenesis of supercell storm:with a weak SCP, less horizontal vorticity is generated via baroclinic process, which is not favorable for the mid-level mesocyclone maintaining and low level vortex formation; However, when the SCP is too strong, the intense updraft will be titled rearward by the strong surface gust front, cutting off the connection between the mid-level mesocyclone and low-level vortex and less favorable for tornadogenesis yet.Compared with a typical U.S. Great Plain tornadic supercell case at mid-latitude, the present extracted sounding has a higher melting level and deeper warm layer, thus larger potentials for hail melting and rain evaporation. Consequently, the responses of the microphysical processes to the DSDs are more pronounced and vertically deeper in the present Ju108 case. This suggests that DSD-related cloud microphysics has a stronger influence on cold pool in downdraft region and thus the tornadogenesis of supercell storm in our extracted sounding environment.The Milbrandt and Yau (MY) multi-moment scheme was also used for further investigation. Different from the Lin-type single-moment scheme, more hydrometeor particles can extend to higher levels in MY multi-moment simulations as the particle fall speeds are calculated in a different way. The influences of the shape parameter on the SCP and tornadogenesis are relatively less pronounced than intercept parameters. Tornadic vortices are reproduced almost in all MY multi-moment experiments: generally, stronger tornadic vortex with longer lifetime is produced in weak-SCP experiments, while weaker or no tornadic vortex is formed in strong-SCP experiments.In a word, our simulation results have shown great impacts of DSDs characteristics on tornadogenesis of a simulated tornadic supercell thunderstorm over Eastern China. Compared with a typical tornadic supercell over U.S. Great Plain, the one focused on in our study is more sensitive to microphysics characteristics. Our findings provide some useful insights for improving the understanding about the formation mechanisms of tornadic supercell storms in Eastern China and the prediction and artificial suppression of tornadic supercell in the future.