A case study of the weather research and forecasting model applied to the joint urban 2003 tracer field experiment. Part 1: Wind and turbulence

Numerical-weather-prediction models are often used to supply the mean wind and turbulence fields for atmospheric transport and dispersion plume models as they provide dense horizontally- and vertically-resolved geographic coverage in comparison to typically sparse monitoring networks. Here, the Weather Research and Forecasting (WRF) model was run over the month-long period of the Joint Urban 2003 field campaign conducted in Oklahoma City and the simulated fields important to transport and dispersion models were compared to measurements from a number of sodars, tower-based sonic anemometers, and balloon soundings located in the greater metropolitan area. Time histories of computed wind speed, wind direction, turbulent kinetic energy (e), friction velocity (u_∗), and reciprocal Obukhov length (1/L) were compared to measurements over the 1-month field campaign. Vertical profiles of wind speed, potential temperature (θ), and e were compared during short intensive operating periods. The WRF model was typically able to replicate the measured diurnal variation of the wind fields, but with an average absolute wind direction and speed difference of 35° and 1.9 ms⁻¹, respectively. Using the Mellor-Yamada-Janjic (MYJ) surface-layer scheme, the WRF model was found to generally underpredict surface-layer TKE but overpredict u_∗ that was observed above a suburban region of Oklahoma City. The TKE-threshold method used by the WRF model’s MYJ surface-layer scheme to compute the boundary-layer height (h) consistently overestimated h derived from a θ gradient method whether using observed or modelled θ profiles.