基于WRF-LES的崇礼复杂地形局地风场模拟研究

Local Wind Simulation over Complex Terrain of Chongli Using WRF-LES

采用四重嵌套的WRF-LES,针对2022年北京冬奥会张家口崇礼赛区开展局地风场模拟试验,基于地面自动气象站和激光雷达观测资料,对一次晴空高压系统控制下的具有明显局地风环流特征的天气个例模拟结果进行检验评估。文中引入了STRM130 m地形数据、glc201527 m土地利用数据和CL‐DAS的土壤湿度数据用以提高模拟结果的准确性,并设计了敏感性试验来探讨不同资料对模拟结果的影响。结果表明:(1)WRF-LES能够呈现出复杂地形下局地风场的时空变化特征,各站风向绝对误差在10°~60°,风速绝对误差在0.5~2 m·s^(-1)。在山谷和山沟区域,模拟风场和观测风场都表现出明显的日变化特征,海拔较高站点的误差比海拔相对较低站点的误差更小。海拔较低站点在山谷风或上下坡风发展稳定时段风向误差较小,风向转换时段误差较大。(2)更新地形、土地利用以及CLDAS土壤湿度初始场对模拟结果都有一定程度改善。其中更新CLDAS土壤湿度初始场对风向和2 m气温的改善效果最为明显,风向绝对误差减小4.26°,2 m气温绝对误差减小0.84℃。更新土地利用对风速的改善效果最明显,风速绝对误差减小0.32 m·s^(-1)。(3)CLDAS土壤湿度较ERA5的土壤湿度明显偏小,较小的土壤湿度会使土壤具有低传导率或低热容量,这种状态会导致地表温度在白天的升温幅度和夜间的降温幅度都增大,较强的白天增温垂直高度可延伸至700~900 m,较强的夜间降温垂直高度在200 m以下,与此同时,较小土壤湿度也会使白天低层风速增大,这些特征在山谷和山沟中更为明显。

Utilizing four nested domains of large eddy simulation with Weather Research and Forecasting Model(WRF-LES),this paper carries out wind simulation test in Chongli area,where is one of the host locations of2022 Beijing Winter Olympics.Based on surface and lidar data,a clear-sky case with high-pressure system is chosen to evaluate the performance of the model.STRM130 m topographic data,glc201527 m land use data,and China Meteorological Administration Land Data Assimilation(CLDAS)soil moisture data are introduced to improve the simulated results.In addition,sensitive tests are conducted to evaluate effects of different input data.Results show that:(1)Spatial and temporal changes of simulated winds reasonably agree with observations.The mean absolute error(MAE)of wind direction is 10°~60°and the wind speed is 0.8~2 m·s^(-1).For wind direction,errors of higher-elevation areas are less than that of lower-elevation areas.The errors are less during the devel‐oped period of valley-wind or slope-wind but grow higher when wind shifts.(2)Positive effects are concluded after the update of terrain,land use and soil moisture data.The improvements are most obvious for both wind di‐rection and 2-meter temperature after using CLDAS soil moisture,with MAE values reduced by 4.26°and0.84°C,respectively.The wind speed is more sensitive to land use,and the error is reduced by 0.32 m·s^(-1).(3)The CLDAS soil moisture is obviously lower than that of ERA5.Lower soil moisture can create lower con‐ductivity or lower heat capacity,this state can create stronger surface warming during daytime and cooling at night,and the vertical height of the stronger daytime warming can reach 700~900 m,while the intense nighttime cooling is below 200 m.An apparent wind increase is also noted in lower layers during daytime.These changes are more obvious in valleys and ravines.