Sensitivity of near-ground wind to PBL schemes, reanalysis data, and nudging in month-long WRF simulations in the Tibetan Plateau with highly complex terrain

The construction of wind-sensitive structures in the Tibetan Plateau requires detailed information about the near-ground wind. Using the Weather Research and Forecasting Model (WRF), one-month numerical investigations on the boundary-layer wind and temperature were conducted considering the Yarlung Zangbo River Valley in the southeast of the Tibetan Plateau. By comparing with site observation, WRF has proved accurate in surface winds over highly complex terrain using a relatively coarse grid. Range analysis is applied to compare model parameters including four-dimensional data assimilation (FDDA) option, reanalysis data, and planetary boundary layer scheme. FDDA option has the strongest impact on wind results. The combination of grid-point data nudging, high-precision reanalysis data, and Quasi–normal Scale Elimination Scheme is recommended. Typical valley wind is observed and successfully simulated, characterized by a diurnal change of near-ground wind and temperature. Stronger winds and neutral air occur around sunset, while weaker winds and stable air appear near sunrise. The wind direction is restricted along the valley. Strong synoptic weather events disturb the local climate, minimize the terrain effects, and produce uniform winds and temperatures. A relatively large wind speed enhanced by the V-shape valley is found. • WRF accurately simulates 1-month complex terrain wind in Tibet using coarse grids. • Surface wind is most sensitive to FDDA, followed by reanalysis data and finally the PBL scheme. • The combination of FDDA, ERA5 reanalysis data and QNSE scheme is recommended. • Valley wind with strong wind around sunset and weak wind near sunrise is dominant. • Strong synoptic event reduces terrain effects and unifies winds and temperatures.