Representation of Turbulence in Multiscale WRF-IBM Simulations Over Complex Urban Terrain

Multiscale atmospheric simulations over highly complex urban terrain have recently been demonstrated using a modified version of the Weather Research and Forecasting model, referred to as WRF-IBM. Downscaling from the mesoscale has the potential to improve many microscale simulations by including the effects of regional meteorology and large-scale turbulence that are typically absent in microscale-only simulations. Multiscale simulations in WRF-IBM are possible due to several major developments to the model; including, an immersed boundary method, the ability to refine vertically when downscaling with grid nesting, and a cell perturbation method (CPM) for improving the development of turbulence after grid refinement interfaces.

In the simulations presented here, we analyze a multiscale configuration centered over Oklahoma City, OK that downscales from the mesoscale (Δ x = 2 m). Unlike microscale-only simulations, this multiscale simulation is forced using a meteorological forecast product, the North American Regional Reanalysis, and is executed similar to a traditional mesoscale forecasting simulation. Data from the Joint Urban 2003 field campaign is used to evaluate the model's skill at predicting wind speed/direction and passive tracer concentrations. Additionally, we compare spectra of turbulent kinetic energy generated using model output and sonic anemometers located throughout the central business district. Results from the multiscale simulation are also compared to a microscale-only simulation and the multiscale configuration is evaluated with and without the CPM. The results of this study emphasize the importance of including regional meteorology in microscale large-eddy simulations and highlight some of the challenges associated with downscaling from the mesoscale to microscale.