10 Oct 2023
 | 10 Oct 2023
Status: this preprint is currently under review for the journal WES.

Machine Learning Methods to Improve Spatial Predictions of Coastal Wind Speed Profiles and Low-Level Jets using Single-Level ERA5 Data

Christoffer Hallgren, Jeanie A. Aird, Stefan Ivanell, Heiner Körnich, Ville Vakkari, Rebecca J. Barthelmie, Sara C. Pryor, and Erik Sahlée

Abstract. Observations of the wind speed at heights relevant for wind power are sparse, especially offshore, but with emerging aid from advanced statistical methods, it may be possible to derive information regarding wind profiles using surface observations. In this study, two machine learning (ML) methods are developed for predictions of (1) coastal wind speed profiles and (2) low-level jets (LLJs) at three locations of high relevance to offshore wind energy deployment; the U.S. Northeastern Atlantic Coastal Zone, the North Sea, and the Baltic Sea. The ML models are trained on multiple years of lidar profiles and utilize single-level ERA5 variables as input. The models output spatial predictions of coastal wind speed profiles and LLJ occurrence. A suite of nine ERA5 variables are considered for use in the study due to their physics-based relevance in coastal wind speed profile genesis, and the possibility to observe these variables in real-time via measurements. The wind speed at 10 m a.s.l. and the surface sensible heat flux are shown to have the highest importance for both wind speed profile and LLJ predictions. Wind speed profile predictions output by the ML models exhibit similar root mean squared error (RMSE) with respect to observations as is found for ERA5 output. At typical hub heights, the ML models show lower RMSE than ERA5 indicating approximately 5 % RMSE reduction. LLJ identification scores are evaluated using the Symmetric Extremal Dependence Index (SEDI). LLJ predictions from the ML models outperform predictions from ERA5, demonstrating markedly higher SEDIs. However, optimization utilizing the SEDI results in a higher number of false alarms when compared to ERA5.

Christoffer Hallgren et al.

Status: final response (author comments only)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on wes-2023-122', Anonymous Referee #1, 25 Oct 2023
  • RC2: 'Comment on wes-2023-122', Anonymous Referee #2, 30 Nov 2023

Christoffer Hallgren et al.

Christoffer Hallgren et al.


Total article views: 311 (including HTML, PDF, and XML)
HTML PDF XML Total Supplement BibTeX EndNote
232 73 6 311 23 2 2
  • HTML: 232
  • PDF: 73
  • XML: 6
  • Total: 311
  • Supplement: 23
  • BibTeX: 2
  • EndNote: 2
Views and downloads (calculated since 10 Oct 2023)
Cumulative views and downloads (calculated since 10 Oct 2023)

Viewed (geographical distribution)

Total article views: 309 (including HTML, PDF, and XML) Thereof 309 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
Latest update: 03 Dec 2023
Short summary
To know the wind speed across the rotor of a wind turbine is key in making good predictions of the power production. However, models struggle capturing both the speed and the shape of the wind profile. Using machine learning methods based on the model data, we show that the predictions can be improved drastically. The work focuses on three coastal sites, spread over the northern hemisphere (the Baltic Sea, the North Sea, and the US Atlantic coast) with similar results for all sites.