Seasonality of wave prediction skills for supporting marine operations in an offshore wind farm

Results of a one year hindcast wave simulation using four different modelling approaches are compared against measurements in Liverpool Bay in the Irish Sea, where an offshore wind farm is located. The different modelling approaches include the use of global or mesoscale wind as forcing, the use of different complexity in shallow water parameterizations and different spatial resolution. Comparisons with observations at a buoy located in the offshore windfarm showed a high improvement due to the use of mesoscale wind (from WRF atmospheric model) instead of global fields (From GFS model). Additionally, the increase in resolution and/or the use of more complex shallow water parameterization (4km for WW3 and 250m for SWAN) improved the skills allowing us to obtain of very good results in terms of wave significant height and mean period. No more improvements were found when SWAN used an additional domain with 100m grids. The work shows the high impact of using proper modelling approach in terms of spatial scales in the skills of the wave simulation in an offshore wind farm during a year. Differences in skills between approaches show that the use of high resolution simulations have benefits in the forecasting of meteorological and oceanographic (metocean) conditions for supporting marine operations in the offshore windfarm and, consequently, in a cost-effectively execution of the works involved.