SWAN Wave Modelling in Fetch-Limited Seas: Influence of Wind Data and White-Capping Formulations on Wave Modelling in Öresund

Sammanfattning

Flooding of coastal infrastructure due to rising sea levels and storm surges can be mitigated by coastal protection. The design of such infrastructure depends not only on the storm surge still-water level but also strongly on storm-wave characteristics. Therefore, numerical wave modelling is essential in the absence of wave observations. Fetch-limited seas with low moderate wave-energy climates are understudied environments in storm-wave modelling, with the Öresund Strait as an example. Among others, high-quality wind data and suitable white-capping formulation are essential when modelling storm-waves in fetch-limited seas. The aim of this study is to evaluate the reliability of the ERA-5 reanalysis wind data set, investigate how different wind data inputs affects wave modelling in Öresund, and how the choice of white-capping formulation influences the model performance focusing on large wave heights. A SWAN-model is used in the study as well as a regional model for comparison. The results showed that ERA-5 underestimates large wind velocities, and was calibrated by linear regression for a better fit in Öresund. The magnitude of the underestimation varied between the analysed wind stations, depending on their surroundings. Different wind data had a significant impact on the wave model, where reanalysis datasets achieved a better model performance than windfields based on measured data. The model performance of the reanalysis datasets ERA-5 and CERRA differed due to differences in spatial and temporal resolution. Using calibrated ERA-5, (adjusted ERA-5), resulted in overestimation of wave heights. However, when combining it with a white-capping formulation adapted for wind-sea conditions a good model performance was achieved. The study highlights the importance of both good-quality wind data and well-suited white-capping formulation when modelling Öresund.