The LOLland offshore Lidar EXperiment (LOLLEX) dataset for studying wind farm flow with a ship-mounted lidar

The LOLLEX measurement campaign was carried out to accumulate knowledge on the atmospheric layer and the flow within and outside the wind farm. The campaign was held between September 2022 and September 2023 at Rødsand II offshore wind farm, located in the Eastern Baltic Sea, south of the island of Lolland, Denmark. The nearby landmass is notable for its flat terrain, adding little disturbance beyond a change in land-sea roughness length.

 

The primary measurements were performed by scanning and profiling lidars. The scanning lidar WindCube 100S had a vertical range up to 2.5 km and worked in 30-minute cycles, alternating between a staring mode with a 1 Hz sampling frequency for vertical wind velocity scans and a DBS mode for horizontal wind speed profiles. The profiles were recorded every 30 minutes during a gap between two scans. The profiling lidar WindCube V2 retrieved wind speed profiles for a shorter vertical range of 40-290 m with a sampling frequency of 0.25 Hz. During the first phase of the campaign, September 2022-January 2023, both lidars were stationed at Rødbyhavn and performed onshore measurements. During the second phase, January-August 2023, the lidars were mounted on the crew transport vessel (CTV) and travelled with it to Rødsand II when the weather permitted. While the rapid CTV movement disrupted WindCube 100S' performance, the lidar collected data during long maintenance stays near the turbines, capturing the vertical speed component of the wake flow.

 

The dataset was processed to filter noise from lidar scans and apply motion correction to WindCube V2 profiles. Vertical scans taken continuously for 10-25 minutes constitute one-third of the dataset scans and cover about 1500 hours; half of this subset demonstrates at least 50% data availability below 500 m. The dataset is classified by data availability, atmospheric boundary layer height, and various events, such as recorded wakes and atmospheric events, e.g., observations of gravity waves or Kelvin–Helmholtz billows.

 

The dataset is planned for release with a throughout documentation at the end of 2026. Processing scripts are also included, along with data visualization scripts and routines to highlight cases of interest, such as the presence of wake or gravity waves in a scan.