Impact of physical parameterizations on wind simulation with WRF V3.9.1.1 under stable conditions at planetary boundary layer gray-zone resolution: a case study over the coastal regions of North China

Reliable simulation of wind fields under stable weather conditions is vital for preventing air pollution. In this study, we investigate how different physical parameterizations impact simulated near-surface wind at 10 m height over the coastal regions of North China using the Weather Research and Forecasting (WRF) model with a horizontal grid spacing of 0.5 km. We performed 640 simulations using combinations of 10 planetary boundary layer (PBL), 16 microphysics (MP), and four shortwave–longwave radiation (SW–LW) schemes. Model performance is evaluated using measurements from 105 weather station observations. The results show that the WRF model can reproduce the temporal variation of wind speed in a reasonable way. The simulated wind speed is most sensitive to the PBL schemes, followed by SW–LW schemes and MP schemes. Among all PBL schemes, the MYJ scheme shows the best temporal correlation with the observed wind speed, while the Yonsei University (YSU) scheme has the lowest model bias. Dudhia–RRTM and MYDM7 show the best model performances out of all SW–LW and MP schemes, respectively, and the interactions among schemes also have large influences on wind simulation. Further investigation indicates that model sensitivity is also impacted by ocean proximity and elevation. For example, for coastal stations, MYNN shows the best correlation with observations among all PBL schemes, while Goddard shows the smallest bias of SW–LW schemes; these results are different from those of inland stations. In general, according to the bias metrics, WRF simulates wind speed less accurately for inland stations compared to coastal stations, and the model performance tends to degrade with increasing elevation. The WRF model shows worse performance in simulating wind direction under stable conditions over the study area, with lower correlation scores compared to wind speed. Our results indicate the role parameterizations play in wind simulation under stable weather conditions and provide a valuable reference for further research in the study area and nearby regions.