WRF Model Study of the Great Plains Low-Level Jet: Effects of Grid Spacing and Boundary Layer Parameterization

In association with the Plains Elevated Convection at Night (PECAN) field campaign, the ability of the Weather Research and Forecasting (WRF) model to accurately resolve the Great Plains low-level jet (LLJ) was investigated. WRF-modeled LLJs were compared to high-resolution observations collected during phase one of the Lower Atmospheric Boundary Layer Experiment (LABLE) at the Southern Great Plains Atmospheric Radiation Measurement site. Seeking the grid spacing that most accurately reproduces the observed LLJ at a reasonable computational expense, model runs were performed with various horizontal and vertical spacings. WRF-modeled LLJs were evaluated on grids with 4-, 2-, and 1-kilometer horizontal spacing, and on the default stretched and non-stretched 20- and 40-meter vertical grids. Finally, three boundary layer parameterization schemes were tested for applicability to reproduce LLJs. A local transport scheme (Mellor-Yamada Nakanishi Niino - MYNN), a non-local transport scheme (Yonsei University - YSU), and a local transport scale elimination scheme (Quasi-Normal Scale Elimination - QNSE) were compared. Results indicate that refinement of the horizontal spacing does not necessarily improve the modeled wind speed. It was found furthermore that increasing the number of vertical levels on a non-stretched vertical grid provides additional information about the structure of the LLJ, but the benefit is limited by the increase in computational expense. Lastly, this work indicates that the QNSE boundary layer scheme offers the best parameterization for resolving the LLJ compared to the YSU and MYNN schemes.