Large-eddy Simulations over Granite Mountain using WRF

The Weather Research and Forecasting (WRF) model is used to perform a nested large eddy simulation of flow over the Granite Mountain Atmospheric Science Testbed (GMAST) southwest of Salt Lake City, Utah. Simulations are validated against observation data taken as part of the Mountain Terrain Atmospheric Modeling and Observations (MATERHORN) project (Fernando et al. 2015). Particular attention is given to the near-surface temperature because underpredicting the diurnal temperature range over complex terrain is a known WRF bias (Zhang et al. 2013, Massey et al. 2014). By modifying the soil moisture initialization, two-meter temperature error is reduced several degrees over a diurnal cycle. Results are also improved by implementing the 'hybrid' soil thermal conductivity scheme used by Massey et al. 2014. Accurate prediction of the near surface temperature is necessary for the model to reproduce a nocturnal cold air pool in the basin east of Granite Mountain. Additional model sensitivity studies address the effects of domain size, grid resolution, input resolution, and nesting configurations on model accuracy. While the WRF model performs reasonably well, it remains difficult to resolve strong gradients near the surface of steep complex terrain. Numerical limitations associated with the terrain following coordinate system used in WRF are partly to blame for this inadequacy.