Evaluating Wind Velocity Measurement Errors in Ground-Based Doppler LiDAR Using Virtual Doppler LiDAR and Large Eddy Simulation

Doppler Wind LiDARs (DWLs) are remote sensing devices that measure wind speed and direction by analysing the Doppler shift of the light backscattered from atmospheric particles along the lidar beam's line of sight. Hence, DWLs are extensively employed in boundary layer meteorology to analyse wind flow due to their ability to provide high-resolution wind measurements. Recently, there has been growing interest in deploying DWLs in urban environments, where mast-based cup anemometers or sonic anemometers face challenges. However, DWL scanning techniques typically assume a homogeneous, stationary wind field, assumptions which often break down in urban boundary layers due to turbulence caused by buildings and complex topography that significantly influences wind profiles. Moreover, the selection of DWL scanning patterns and their configuration should be carefully tailored to the specific application.

One of the most-used scanning methods for measuring vertical wind velocity profiles is the Velocity Azimuth Display (VAD). The technique involves scanning the laser beam around the zenith in a conical pattern at a fixed elevation angle. However, completing the full 360° requires a finite time, during which the wind speed is assumed to be constant. Additionally, if the wind varies significantly within the sampling volume (e.g., due to turbulence or flow inhomogeneity) the calculated wind profiles may be inaccurate.

Large-Eddy Simulation (LES), with a sufficiently high grid resolution to resolve turbulent motions, provides a means to evaluate potential errors in DWL sampling strategies. This study uses a Virtual Doppler LiDAR (VDL) tool (Rahlves et al., 2022) within the Parallelized Large-Eddy Simulation Model (PALM, version 6.0) to estimate velocity profiles derived from simulated radial velocities along virtual laser beam paths under the VAD scheme. The research is part of the ASSURE Project (Across-Scale Processes in Urban Environments), which focuses on Bristol, UK. The project investigates urban wind flow using DWLs deployed across the city, employing scanning strategies utilised during a one-year field campaign beginning in May 2024.

Bristol was chosen for its compact urban layout and distinct topographic features, including the Avon Gorge and a central valley. The city serves as a case study for examining urban wind dynamics. This study's objectives are twofold: (1) to identify and quantify errors between the vertical wind profile derived from a VAD scan using the VDL and the profile directly taken from the PALM model and (2) to facilitate comparisons between PALM-simulated wind profiles and observations from ground-based DWL. By addressing the discrepancies arising from topographically induced flow, this research aims to enhance the reliability of DWL data in urban settings and improve our understanding of urban boundary layer processes. Results will be presented for a case study of flow channelled by a deep valley interacting with a city-centre boundary layer.

Rahlves, C., Beyrich, F., and Raasch, S. (2022). ‘‘Scan strategies for wind profiling with Doppler lidar – an large-eddy simulation (LES)-based evaluation’’, Atmospheric Measurement Techniques, 15(9), 2839-2856