Stratospheric turbulence (altitudes 10-30km) is characterized by patchy high frequency fluctuations in the stratospheric wind fields and long-lived energetic eddies with a few hundred meters scales. While there is a need for real time forecast of these thin stratospheric CAT layers,

they cannot be resolved by even the latest generation of Weather Research and Forecasting (WRF) mesoscale meteorological codes. Our approach is based on vertical nesting and adaptive vertical gridding in nested mesoscale WRF/microscale codes. The inner nest of WRF (1km grid in the horizontal, 150 levels in the vertical) is coupled with a sequence of embedded microscale nests, both horizontally and vertically (microscale nests with 450 and 750 vertical levels). The fully three-dimensional, compressible Navier-Stokes equations with non-homogeneous

stratification (buoyancy) and background rotation (Coriolis) are solved with a stretched, adaptive grid in the vertical. An adaptive, staggered grid mesh is used in the vertical, with a grid spacing down to a few meters

in thin CAT layers where strong turbulent mixing occurs. For nesting, both lateral and vertical boundary conditions are treated via relaxation zones where the velocity and temperature fields are relaxed to those obtained from the WRF inner nest. This methodology is applied to the analysis of field data from the T-REX Campaign (Terrain-induced Rotor Experiment), Owens Valley, CA, 2006. Our simulations are based on initial and boundary conditions from both

GFS and high resolution T799 L91 ECMWF analysis datasets.

Our results demonstrate how mountain waves induced by topography propagate through the tropopause with polarization and experience wavelength shortening in the lower stratosphere. The structure is clearly seen from the finest microscale nest. There are intense wave breaking events above the tropopause resulting in formation of several sharp adiabatic layers. Below adiabatic layers at the tropopause level there are regions of high stability characterized by large vertical gradients of potential temperature. This has been observed in several radiosonde profiles during the TREX campaign of measurements. These laminated structures with regions of inhomogeneous gradients can further impact the propagation and transmission of upward propagating topographic waves. These are strongly 3D nonlinear regimes. Thin adiabatic layers of potential temperature and strong patches of vertical velocity cannot be resolved even by latest generation of Weather Research and Forecasting (WRF) mesoscale codes. They are resolved by the microscale nests with our coupled WRF-microscale nest simulations. We discuss the impact of laminated structures on transport of passive scalars such as ozone and vertical redistribution of tracer.