Jianting Du

On the impact of wind on the development of wave field during storm Britta

The observation of extreme waves at FINO 1 during storm Britta on the 1st November 2006 has initiated a series of research studies regarding the mechanisms behind. The roles of stability and the presence of the open cell structures have been previously investigated but not conclusive. To improve our understanding of these processes, which are essential for a good forecast of similarly important events offshore, this study revisits the development of storm Britta using an atmospheric and wave coupled modeling system, wind and wave measurements from ten stations across the North Sea, cloud images and Synthetic Aperture Radar (SAR) data. It is found here that a standard state-of-the-art model is capable of capturing the important characteristics of a major storm like Britta, including the storm path, storm peak wind speed, the open cells, and peak significant wave height (Hs) for open sea. It was also demonstrated that the impact of the open cells has negligible contribution to the development of extreme Hs observed at FINO 1. At the same time, stability alone is not sufficient in explaining the development of extreme Hs. The controlling conditions for the development of Britta extreme Hs observed at FINO 1 are the persistent strong winds and a long and undisturbed fetch over a long period.

The use of a wave boundary layer model in SWAN

A Wave Boundary Layer Model (WBLM) is implemented in the third-generation ocean wave model SWAN to improve the wind-input source function under idealized, fetch-limited condition. Accordingly, the white capping dissipation parameters are re-calibrated to fit the new wind-input source function to parametric growth curves. The performance of the new pair of wind-input and dissipation source functions is validated by numerical simulations of fetch-limited evolution of wind-driven waves. As a result, fetch-limited growth curves of significant wave height and peak frequency show close agreement with benchmark studies at all wind speeds (5 ∼ 60 ms−1) and fetches (1 ∼ 3000 km). The WBLM wind-input source function explicitly calculates the drag coefficient based on the momentum and kinetic energy conservation. The modeled drag coefficient using WBLM wind-input source function is in rather good agreement with field measurements. Thus, the new pair of wind-input and dissipation source functions not only improve the wave simulation but also have the potential of improving air-sea coupling systems by providing reliable momentum flux estimation at the air-sea interface. This article is protected by copyright. All rights reserved.

Wave boundary layer model in SWAN revisited

In this study, we extend the work presented in Du et al. (2017) to make the wave boundary layer model (WBLM) applicable for real cases by improving the windinput and white-capping dissipation source functions. Improvement via the new source terms includes three aspects. First, the WBLM wind-input source function is developed by considering the impact of wave-induced wind profile variation on the estimation of wave growth rate. Second, the white-capping dissipation source function is revised to be not explicitly dependent on wind speed for real wave simulations. Third, several improvements are made to the numerical WBLM algorithm, which increase the model’s numerical stability and computational efficiency. The improved WBLM wind-input and white-capping dissipation source functions are calibrated through idealized fetch-limited and depth-limited studies, and validated in real wave simulations during two North Sea storms. The new WBLM source terms show better performance in the simulation of significant wave height and mean wave period than the original source terms.

Coupling atmospheric and ocean wave models for storm simulation

This thesis studies the wind-wave interactions through the coupling between the atmospheric model and ocean surface wave models. Special attention is put on storm simulations in the North Sea for wind energy applications in the coastal zones. The two aspects, namely storm conditions and coastal areas, are challenging for the wind-wave coupling system because: in storm cases, the wave field is constantly modified by the fast varying wind field; in coastal zones, the wave field is strongly influenced by the bathymetry and currents. Both conditions have complex, unsteady sea state varying with time and space that challenge the current coupled modeling system. The conventional approach of estimating the momentum exchange is through parameterizing the aerodynamic roughness length (z0) with wave parameters such as wave age, steepness, significant wave height, etc. However, it is found in storm and coastal conditions, z0 parameterization method often fails in reproducing z0 because the complexity of the sea state cannot be represented by a few selected wave parameters. Different from the parameterization method, physics-based methods take the idea that the loss of momentum and kinetic energy from the atmosphere must, by conservation, result in the generation of the surface waves and currents. The physics-based methods are sensitive to the choice of wind-input source function (Sin), parameterization of high-frequency wave spectra tail, and numerical cut-off frequencies. Unfortunately, literature survey shows that in most wind-wave coupling systems, either the Sin in the wave model is different from the one used for the momentum flux estimation in the atmospheric model, or the methods are too sensitive to the parameterization of high-frequency spectra tail and numerical cut-off frequencies. To confront the above mentioned challenges, a wave boundary layer model (WBLM) is implemented in the wave model SWAN as a new Sin. The WBLM Sin is based on the momentum and kinetic energy conservation. The wave-induced mean wind profile changes at all vertical levels within the wave boundary layer, and the spectral sheltering effect at each frequency within the wave spectrum are explicitly considered. The WBLM Sin is used for both the calculation of the wave growth and the estimation of the air-sea momentum flux. Moreover, the WBLM Sin extended the model ability in high-frequency ranges so that the issue of high-frequency spectra tail and numerical cut-off frequencies are automatically solved. The new WBLM method is proved to be able to improve both the wave simulation and stress estimation in idealized fetch-limited wind-wave evolution studies. To apply the WBLM method in real cases, proper setup of the dissipation source function, numerical stability and model efficiency are needed to be considered. Therefore, a revised dissipation source function for the wave model and a refinement of the numerical algorithm of WBLM Sin is done. The new pair of wind-input and dissipation source functions are evaluated with point measurements through wave simulations during offshore and onshore storms in the west coast of Denmark. The WBLM method is proved to provide significant wave height and mean wave period that outperforms the other approaches in SWAN when compared with measurements. The WBLM method is further applied in the wind-wave coupling system during a number of North Sea storms. In comparison, six other coupling method have also been used for one of the storms. Results of wind, wave, and stress have been validated with point measurements at a coastal, shallow water site. In particular, the spatial distribution of z0 from WBLM is found to have similar spatial patterns as the Advanced Synthetic Aperture Radar (ASAR) radar backscatter; both show features of the bathymetry. Analysis of the wind field from the non-coupled and WBLM coupled experiments show that the wind-wave coupling is important in strong wind conditions, varying wind conditions (e.g. front system, open cellular convections during a storm), and coastal areas. The thesis is submitted to the Danish Technical University in partial fulfillment of the requirements for the PhD degree. DTU PhD-0074(EN) April 2017 DOI 10.11581/DTU:00000020 Sponsorship: Danish Forskel (PSO-12020) “X-WiWa” project Pages: 129 References: 173 Figures: 67 Tables: 9 Wind Energy Department Technical University of Denmark P.O.Box 49 DK-4000 Roskilde Denmark Telephone +45 93511127 jitd@dtu.dk www.dtu.dk