Turbulence Closure Parameterization and Grid Spacing Effects in Simulated Supercell Storms

Many meteorological organizations plan to substantially increase the resolution of the limited-area models used for severe weather prediction. Such an approach does not guarantee a priori the reduction of the uncertainty of the decision maker in the prediction of severe weather impact. A deep moist convective process, a supercell, is studied in a simplified atmospheric scenario by means of high-resolution numerical simulations with the Consortium for Small-Scale Modeling (COSMO) model. Different turbulence closure models and their impact on the spatiotemporal properties of storm processes are discussed. In the range of grid spacing between 1 km and 100 m, also termed “terra incognita,” the simulations of a supercell converge with respect to flow field structure, transport properties, and precipitation fields when a turbulence closure derived from large-eddy simulation (LES) is used. In contrast, more simplified turbulence closures such as 1D (vertical) boundary layer approximations yield substantially worse results than the 0.2-km LES reference simulation.