Virtual LiDAR in WRF-LES

Abstract:

Recent strides in wind-profiling LiDAR technology have opened the doors for an increasing number of field campaigns employing these instruments, including cases focused on flow in complex terrain and heterogeneous phenomena like wind turbine wakes (e.g. Perdigão, XPIA in Colorado, and the Inn River Valley). LiDAR exploits the doppler backscatter effect off of aerosols carried in the flow to measure line-of-sight (LoS) wind velocities along projected beams. However, intercomparison of observational LiDAR datasets and computational models is nontrivial. Previous work implementing online, within-code simulators for instruments such as sodar, profiling lidar, as well as offline scanning lidar simulators has proved fruitful for exploring the nuance of the measurements and connecting them with our understanding of processes as represented in the computational models.

We implement a virtual scanning LiDAR in WRF-LES to sample the simulated flow as would be done by the instrument. Beam paths are generalized to accommodate different typical scanning patterns. Most commonly, scans will hold either the azimuthal or elevation angle constant while varying the other to create, e.g., conical scans (VAD), conical sections (PPI), or a vertical plane sampling (RHI). In addition to incorporating the geometry of the beams, we also implement an equipment-defined spatial weighting function that can be adjusted based on the hardware of the instrument considered. The temporal averaging of the virtual LiDAR measurements can also be modified based on experimental data collection protocols, varying from instantaneous measurements to half-hourly (or longer) averages.

Initial testing has focused on the eXperimental Planetary boundary layer Instrumentation Assessment (XPIA) campaign at the Boulder Atmospheric Observatory (BAO), from 2 March to 31 May 2015. This campaign has been well-modeled and validated using WRF-LES, and five scanning LiDARs were deployed at the campaign. We will present a comparison of virtual LiDAR scans with WRF-LES as compared to actual data collected during XPIA, including RHI scans and stares at tower-based sonic anemometers.

Although the performance of WRF-LES is based on a number of phenomena, the accessibility of a virtual LiDAR within this widely-used code can enable improved model validation, experiment design, effective visualization, and research of processes exemplified in observed case studies.