4.3 Coupled mesoscale-LES modeling to improve the representation of stably-stratified atmospheric boundary layers

Munoz-Esparza, Domingo, Branko Kosovic, Jeremy Sauer, National Center for Atmospheric Research, and Julie Lundquist, Department of Atmospheric and Oceanic Sciences, University of Colorado Boulder

The stable boundary layer (SBL) is a challenging scenario for numerical weather prediction (NWP) models due to the complex dynamics involved including Kelvin–Helmholtz instabilities, turbulent intermittency, low-level jets and gravity waves, among others. Therefore, SBLs remain the most difficult regime for planetary boundary layer (PBL) schemes to forecast, and where the largest errors occur. In this talk, we will present coupled WRF mesoscale-LES simulations of the SBL during one of the nights of the CWEX-13 field experiment. Dynamical downscaling from synoptic-scale down to resolved three-dimensional eddies in the SBL was performed, spanning four orders of magnitude in horizontal grid resolution: from 111 km down to 8.2 m. Turbulence generation at the mesoscale-LES transition was enabled by the cell perturbation method. By comparing to wind profiler and sonic anemometer observations at the CWEX-13 site, we show that multi-scale modeling is capable of improving vertical wind speed shear, turbulent kinetic energy and surface temperature predictions in the SBL. Remarkably, the multiscale WRF simulation improved not only the turbulence fluctuations but also variability at scales from 1-min to 1-h, not captured by the mesoscale model with a PBL parameterization. These improvements result from resolving part of the turbulent and sub-meso spectra, which in turn allows for explicit representation of SBL phenomena such as global intermittency in our finest LES domain. This work demonstrates the great potential of multi-scale modeling to improve prediction of real-world stable boundary layers.