Abstract
In the present work, wind and wave climate baselines are derived on the basis of historical hindcast data. The data covers an area in the Atlantic Ocean around an existing wind farm at a high resolution (∼1km for wind and ∼100 m for waves), which is located west of Portugal (WindFloat Atlantic: WFA). For this, two regional numerical models, atmospheric model WRF and wave model SWAN, have been run for ten (10) years, and results have been generated every 1 hr. Detailed geographical information has been acquired from EMODNET database. Wind and wave boundary conditions have been taken from global database ERA5. Then, wind and wave data have been statistically analyzed, and various aspects of their probabilistic structure have been investigated. Results include mean annual and inter-annual variability, study of the probability structure in the form of probability quantiles (50%, 90%, 95%, 90%, 99.9%), directional distribution, extreme-value estimation for return periods (50, 100 years), long-term trends, study of the probability of the duration of an event (e.g., “the time that wave height lies below 3m for more than 12 hours”), as well as the waiting time (weather window) until this event will happen again.