Scanning Doppler lidars for Wind Energy research and validation of NWP model wind forecast in complex terrain

摘要

Measurements from scanning Doppler lidars at three sites in the Columbia River Gorge during the second Wind Forecast Improvement Project (WFIP-2) are used to analyze the temporal and site-to-site variability of the wind profile, to better understand physical processes related to the terrain effects, weather conditions, and presence of wind farms. Continuous measurements from two 200S lidars for 18-months and year-long measurements from a Streamline lidar provide a comprehensive dataset to validate NWP model wind forecasts and evaluate model accuracy over days, months and seasons, as well as for periods of interesting meteorological events including wind ramps, cold fronts, or marine push in the study area. Verification metrics such as bias, RMSE, MAE, and the correlation coefficient between observed and modeled wind variables are analyzed as a function of height, time, and forecast hour, revealing meteorological features that are not well represented by models and provide insight to potential model improvements.