Hurricane Impacts on US East Coast Offshore Wind Energy Lease Areas

The US east coast has excellent offshore wind resources, and wind turbines deployed in offshore wind energy lease areas there have the potential to significantly contribute to national electricity supply (Pryor et al. 2021: Wind power production from very large offshore wind farms, Joule, 5, 2663-2686). Analyses of hourly ERA5 reanalysis wind speeds at 100 m indicate expected design-standard 50-year return period wind speeds (U₅₀) of 30 to 40 ms^-1 within these lease areas (Barthelmie et al. 2021: Extreme Wind and Waves in U.S. East Coast Offshore Wind Energy Lease Areas, Energies, 14, 1053). After applying corrections to account for spectral smoothing and the 20-minute averaging period, the associated pseudo-point U₅₀ estimates are 34 to 46 ms^-1. Analysis of output from ERA5 indicate that the 50-year return period maximum wave height may substantially exceed 15 m. However, parts of the US east coast episodically experience tropical cyclone passages and this study is designed to examine extremes associated with historical hurricanes.

Here we perform and present high-resolution simulations of so-called ‘Superstorm Sandy’ – Hurricane Sandy that made landfall on the New Jersey coastline causing significant economic losses (Letson et al. 2021: Intense Windstorms in the Northeastern United States, Natural Hazards and Earth System Sciences, 21, 2001–2020 and Zambon et al. 2014: Tropical to extratropical: Marine environmental changes associated with Superstorm Sandy prior to its landfall, Geophys. Res. Lett., 41, 8935-8943). The track of Hurricane Sandy is such that it, and its associated rain bands, travelled close to or over many of the active US east coast lease areas. These simulations are performed with the Weather and Research Forecasting (WRF) model and the Coupled Ocean Atmosphere Wave Sediment Transport (COAWST) model and are designed to answer three linked research questions:

1. Are simulated extreme wind speeds at/close to wind turbine hub-height in these offshore wind energy lease areas during Hurricane Sandy significantly influenced by use of more comprehensive atmosphere-wave-ocean coupling? i.e., How different are 150 m hub-height wind speeds when simulated with WRF versus COAWST. Does either simulation imply the need for deployment of wind turbine classes designed for wind speeds of 57 ms^-1?
2. How long would wind turbines deployed in the lease areas have produced low/no power due to wind speeds above cut-out of 25 ms^-1? Is the research finding critically contingent on whether WRF or COAWST is used?
3. What level of significant wave heights is simulated for the offshore lease areas using COAWST?

The Fitch wind-farm parameterization (Fitch et al. 2012: Local and mesoscale impacts of wind farms as parameterized in a mesoscale NWP model, Mon. Wea. Rev., 140, 3017-3038) is applied to simulated wind power production and wind farm wake generation. Following Pryor et al. (2021), IEA 15 MW reference wind turbines with a hub-height of 150 m are deployed in the lease areas with a spacing of 1.85 km. This results in a total of 2069 wind turbines within the simulation domain. For both the WRF-only and COAWST simulations, two atmospheric domains are used. The outer atmospheric domain has a grid spacing of 4 km and the inner domain has a grid spacing of 1.33 km. Both domains use 57 vertical levels and wind speeds at the model level at 155 m are used as equivalent to the wind turbine hub-height. In the COAWST simulation, WRF, Regional Ocean Modeling Systems (ROMS), and Simulating Waves Nearshore (SWAN) models are coupled using the Model Coupling Toolkit (MCT). These models represent the atmosphere, ocean, and waves, respectively. ROMS and SWAN use an outer domain with a grid spacing (dx) of 10 km, and an inner domain with dx of 3.33 km. ROMS is configured with 16 vertical levels, while SWAN has one vertical level.

The WRF-only simulation indicates highest hub-height wind speeds in the offshore wind energy lease areas of 44.5 ms^-1, which is in good agreement with U₅₀ estimates from Barthelmie et al. (2021). The period of highest wind speeds is centered at 0000 UTC on 30 October 2012, and during this time there was wide-spread exceedance of 37.5 ms^-1. However, the highest wind speed is well-below the 50-year return period wind speed for a wind turbine class I (50.0 ms^-1). Electrical power production dropped to below 20% of possible power for a period of 5 hr 30 min due to high sustained wind speeds of over 25 ms^-1 over the offshore lease areas. Two additional periods with low system-wide power production occurred prior to and after this extended period of low power production but both lasted only 1 hour. Thus, the WRF-only simulations imply that Hurricane Sandy was not associated with wind conditions likely to cause structure failures and electrical power production would be only fairly briefly interrupted. Simulations that are ongoing with COAWST will be used to evaluate the degree to which the WRF-only simulation results are robust.