Improvements in the treatment of evaporation and melting in multi-moment versus single-moment bulk microphysics: results from numerical simulations of the 3 May 1999 Oklahoma tornadic storms

In a previous study, we reported on the results of numerical experiments designed to test the impact of single and multi-moment bulk microphysical parameterizations (BMP) on high-resolution convection-resolving simulations of the 3 May 1999 Oklahoma tornado outbreak using the Advanced Regional Prediction System (ARPS) model. This outbreak was characterized by severe, long-lived supercells that often tracked in close proximity (within a few storm diameters) to each other without significant destructive interference, and displayed relatively weak and localized cold pools. Both real-data and idealized single-sounding simulations indicated better agreement with the observed reflectivity and cold-pool structures of the storms when using a multi-moment BMP, particularly in the relatively high-resolution (500 m grid spacing and smaller) idealized simulations.

In this study, we build upon the previous work by providing physical explanations for the superior performance of the multi-moment schemes--the Milbrandt and Yau (2005) BMP--at high grid resolutions. In particular, we find that the size-sorting process, which is modeled in the multi-moment schemes, but not in the single-moment schemes, plays a significant role within the downdraft region of the simulated storms, in general yielding a preponderance of large mean particle sizes (mainly raindrops and hailstones) at low-levels, reducing the evaporation and melting rates over those found in a single-moment treatment. In addition, the flexibility of allowing the total number concentration N_t of hydrometeors to vary independently of the mixing ratio q in the multi-moment scheme allows for more physically-realistic treatment of the effect of the evaporation and melting processes on changes in the distribution. These and other effects, and general implications for high-resolution thunderstorm modeling using the multi-moment BMP approach will be discussed.

Preliminary results from a new set of real-data numerical simulations, nested down to sub-1km grid spacings, of the 3 May 1999 outbreak will be presented. These experiments are designed to test the results of the previous idealized simulations in the more complicated real-data prediction framework, and to assess the potential for the multi-moment microphysics approach to improve the prediction of severe thunderstorms and associated circulations, such as mesocyclones and tornadoes.