Untangling atmospheric flows through the lenses of wind lidars

Doppler wind lidars are optimal tools to study the planetary boundary layer (PBL) and to characterize wind and turbulence, which is of utmost importance for wind energy applications. However, flow measurements using multiple lidars over complex terrain are still scarce and there is need to further compare lidar measurements and outputs from numerical models.

This thesis starts by demonstrating the usage of multiple coordinated scanning lidars to observe multi-scale flow phenomena over complex terrain. A novel experimental dataset, mainly composed by measurements from scanning lidars, was gathered from the planning, execution and data analysis of the Alaiz Experiment, in northern Spain.

Analysis of multilidar measurements at Alaiz, along with multiple meteorological observations, identified atmospheric hydraulic jumps, mountain waves, valley flow stagnation, among other flow patterns. It was found that lee-side hydraulic jumps at the Alaiz mountain can potentially take place up to 10% of the time, due to the frequent combination of topographical and flow characteristics ideal for the development of such jumps.

Furthermore, simulations using the Weather Research and Forecast (WRF) model, with a nesting domain setup in which the innermost domains were run using the large-eddy simulation (LES) capability, reproduced the hydraulic jump episode in high detail and agreed with the observations both in timing and flow features.

Subsequently, long-range profiling lidar observations were compared with the outputs from the New European Wind Atlas (NEWA) project over both offshore and flat terrain conditions. The NEWA outputs are a product of numerical simulations using the WRF model with a local PBL scheme that assumes K-theory, i.e., that the vector of vertical flux of horizontal momentum and the vector of mean vertical gradient of horizontal velocity are parallel.

The observations showed a misalignment between these vectors from 100 m up to 500 m for both onshore and offshore conditions. Idealized LESs carried out with the WRF model also showed an almost perfect alignment, further demonstrating that there is a need to investigate the conditions in which this misalignment occurs.

The findings in this thesis show that both multi-lidar measurements and meso- to micro-scale numerical models are able to capture large scale atmospheric phenomena over complex terrain and that a long-range lidar aids for the improvement of the PBL schemes used in numerical weather models.