Toward improved offshore wind design: Variability of wind veer revealed by hybrid cluster analysis of lidar data

Understanding offshore wind veer, i.e., the vertical variation of wind direction with height, is critical for improving wind turbine design, structural load assessment, and wake interaction modeling in offshore wind farms. This study aims to explore the variability of offshore wind veer and identify its primary atmospheric drivers across multiple temporal and dynamic regimes. To achieve this, a hybrid clustering framework was developed and applied to a multi-year dataset of high-resolution Doppler lidar measurements collected from an offshore platform near Hong Kong. The methodology integrates shape-oriented time series clustering with directional similarity metrics, enabling the classification of full-profile veer structures under varying conditions. The results indicate that vertical wind veer profiles vary systematically with changes in season, wind speed, and atmospheric stability. Seasonal groupings exhibited distinct profile shapes, reflecting changes in atmospheric structure likely associated with shifts in large-scale flow regimes and mesoscale conditions. Wind speed was found to influence the curvature and directional trend of veer profiles, with low and high wind speed regimes displaying markedly different vertical patterns. Similarly, variations in atmospheric stability corresponded with identifiable shifts in the shape and smoothness of veer profiles, suggesting a sensitivity to stratification and mixing conditions. While the precise physical mechanisms could not be fully resolved from the available dataset, the consistent emergence of distinct veer structures across environmental categories highlights the importance of incorporating dynamic veer characterization into offshore wind assessments.