Validation of the Mann spectral tensor for offshore wind conditions at different atmospheric stabilities

Simulated wind fields are very useful when predicting loads on structures subjected to turbulent winds, wind turbines being a prime example. Knowledge of statistical properties such as the spatial and temporal correlations of real turbulent wind fields increases the realism of the simulated simulated wind fields. The statistical properties of real turbulent wind fields have been shown to depend on quantities such as the surface roughness, the mean wind speed, measurement height and atmospheric stability. The Mann spectral tensor attempts to predict all spatial correlations of shear generated turbulence given only three input parameters. The most suitable such input values have been investigated for different onshore surface roughnesses, but so far not for typical offshore conditions. The meteorological mast at the Rødsand II offshore wind farm has among other instruments sonic anemometers mounted at 15, 40 and 57 meters above sea level. Wind speed spectra at the three heights are calculated and binned with respect to both wind speed and atmospheric stability. The three parameters of the Mann spectral tensor are determined to ensure best fit to the spectra of each of the bins and are presented as a function of mean wind speed, measurement height and atmospheric stability. The behaviour of the presented parameters values are largely consistent with the previous onshore results. The parameter values are also compared to potentially related quantities and a constant quantity is derived. Given optimal parameters the spectral tensor is found to reproduce the surface layer generated turbulence well, also for different atmospheric stabilities, however in the wind speed spectra a contribution from the very large scale quasi-geostrophic turbulence is also observed, a contribution the spectral tensor does not attempt to model.