Bias-Corrected MOST Hub-Height Wind Estimates from Dynamical Downscaling for Offshore Wind Assessment

In offshore wind energy assessment, the vertical profile of near-surface wind speed is frequently modulated by regional atmospheric phenomena, including low-level jets (LLJs), variations in atmospheric stability, and land–sea thermal contrasts, thereby inducing pronounced variability in the synoptic wind field. Under certain conditions, wind speed may even decrease with height, exhibiting negative wind shear. Such non-monotonic wind profiles are particularly common over monsoon-dominated marine regions such as the Taiwan Strait. Previous studies have indicated that conventional vertical extrapolation approaches and Monin–Obukhov similarity theory (MOST) often fail to adequately represent the actual wind field under these conditions, consequently reducing the reliability of wind resource assessment and power generation forecasting.

This study uses near-surface meteorological data from the TReAD, a dynamical downscaling model(2km spatial resolution) developed by the National Science and Technology Center for Disaster Reduction (NCDR), along with MOST to estimate wind speed at hub height. The estimates are compared against observations from the nearest meteorological mast (approximately 725 m away) to develop a data-driven bias-correction approach. A bias prediction model is trained on 2019 data and then applied to independent datasets from 2022 and 2023 to evaluate its generalization across interannual variability and varying boundary-layer conditions.

The results show that, even when trained on data from a single year, the bias-corrected MOST wind speeds consistently reduce overall errors and improve the error across all evaluated years, suggesting that the proposed method can effectively correct MOST's systematic biases under negative wind shear and cross-year nonstationary conditions. Overall, this study presents a MOST bias-correction algorithm integrated with a dynamical downscaling model to produce near-surface wind speed estimates that more closely align with meteorological mast observations, thereby providing a practical approach for offshore wind energy assessment in the Taiwan Strait.