Estimating Offshore Wind Power Potential by Utilizing Gray-Zone Atmospheric Modeling

The global reduction of fossil fuel reserves, coupled with increasing awareness of human impact on climate change, has significantly accelerated the deployment of renewable energy technologies. Among these, offshore wind energy stands out as a promising avenue for achieving sustainability goals for two key reasons. First, offshore wind resources are abundant and benefit from higher energy fluxes due to reduced surface roughness, which allows for higher wind speeds. Second, technological advancements have enabled the deployment of the largest wind turbines to date, such as the MySE 16-260, which boasts a 16 MW capacity, a 260 m rotor diameter, and a 152 m hub height. This enables the harnessing of abundant winds at higher altitudes, enhancing the potential of large-capacity wind turbines for electricity generation.

Furthermore, new methods of deploying turbines far offshore, such as floating wind farms, have effectively removed the constraints associated with proximity to shore. This development allows for the exploitation of extensive wind resources in deep offshore locations. Coastal effects in the atmosphere arise from differences between the physical properties of land and ocean surfaces. Two main factors are at play: ocean surfaces exhibit shorter roughness length scales compared to land, and ocean water possesses a significantly greater heat capacity than typical soil. These differences contribute to phenomena such as Internal Boundary Layers (IBLs), low-level jets (LLJs), and sea breezes, all of which can influence wind farms.

With the growing size of wind turbines and hub heights reaching 150 m, accurate characterization of wind resources at these altitudes is crucial. Typically, long-term wind data at higher altitudes are obtained using wind masts, SODAR, and LiDAR. While these methods provide accurate measurements at high temporal frequencies, they are often expensive to install in deep waters and are sparsely distributed, making them inadequate for detailed spatial wind monitoring.

As a result, atmospheric reanalysis has proven to be a great alternative for offshore wind resource assessment, particularly due to its extensive spatial coverage. Researchers have utilized global and regional reanalysis datasets like ECMWF-Re-Analysis Interim (ERA-Interim), Modern-Era Retrospective Analysis for Research and Applications version 2 (MERRA-2), and the fifth generation ECMWF atmospheric reanalysis (ERA5) to address the spatiotemporal variability of offshore wind resources and to understand the influence of atmospheric circulation patterns on wind power production. However, the coarse resolution of reanalysis data often fails to capture coastal effects accurately. The current state-of-the-art regional reanalysis, Copernicus Regional Reanalysis for Europe (CERRA), offers a resolution of 5.5 km, which still may not be sufficient in some cases.

To overcome these limitations, regional numerical models, particularly the Weather Research and Forecasting (WRF) model, provide a cost-effective solution with high spatial and temporal resolution. Traditional mesoscale model simulations, which typically have a horizontal grid size of 1 km or coarser, cannot resolve turbulence, necessitating the use of a planetary boundary layer (PBL) scheme in conjunction with the Reynolds-Averaged Navier-Stokes (RANS) equations. In typical mesoscale modeling, vertical gradients are much larger than horizontal ones, so PBL schemes are usually one-dimensional and capture only vertical turbulent exchanges. However, when atmospheric simulations are performed with sub-kilometer grid spacing, turbulence is partially resolved, and the assumptions inherent in RANS modeling begin to break down. For grid spacing on the order of a few meters, large-eddy simulations can be performed, but for intermediate grid spacing (e.g., a few hundred meters), a gray-zone parameterization is necessary. For such simulations, traditional PBL schemes are not recommended. Despite the widespread availability of high-performance computing resources, which have made sub-kilometer modeling increasingly popular in wind energy studies, many studies incorrectly use traditional PBL schemes.

This study presents a calibrated ultra-high wind resource assessment at a sub-kilometer (500 m) resolution. This allows us to identify the gray-zone issue for the first time and apply an appropriate gray-zone PBL scheme for wind energy assessment applications in complex offshore environments. The wind resource assessment is utilized with two turbine models: an 8 MW model with a hub height of 120 m, currently employed in coastal waters, and a state-of-the-art 15 MW model with a hub height of 150 m, representing future developments in offshore wind technology. The higher resolution dataset reveals impacts on power production and identifies physical phenomena, such as the IPCJ, that are detrimental to power generation.

Our unique dataset also assesses resource variability at diurnal, interannual, and annual scales, providing additional insights into the suitability of these models for capturing the dynamic nature of offshore wind resources. This work showcases the importance of fine-tuning offshore wind energy selection at the resource assessment level. We provide a framework demonstrating that even for smaller regions, power production implications can arise from effects captured only by calibrated high-resolution models, which are otherwise overlooked.