Complex factors including steep slopes, intense wave breaking, large bottom friction and remarkable wave setup should be considered while studying wave propagation over coral reefs, and how to simulate wave propagation and setup on coral reefs efficiently has become a primary focus. Several wave models can be used on coral reefs as have been published, but further testing and comparison of the reliability and applicability of these models are needed. A comparative study of four numerical wave models (i.e., FUNWAVE-TVD, Coulwave, NHWAVE and ZZL18) is carried out in this paper. These models’ governing equations and numerical methods are compared and analyzed firstly to obtain their differences and connections;then the simulation effects of the four wave models are tested in four representative laboratory experiments. The results show that all four models can reasonably predict the spectrum transformation. Coulwave, NHWAVE and ZZL18 can predict the wave height variation more accurately;Coulwave and FUNWAVE-TVD tend to underestimate wave setup on the reef top induced by spilling breaker, while NHWAVE and ZZL18 can predict wave setup relatively accurately for all types of breakers;NHWAVE and ZZL18 can predict wave reflection by steep reef slope more accurately. This study can provide evidence for choosing suitable models for practical engineering or establishing new models.
Boundary Collocation Method （BCM） based on Eigenfunction Expansion Method （EEM）, a new numerical method for solving two-dimensional wave problems, is developed. To verify the method, wave problems on a series of beaches with different geometries are solved, and the errors of the method are analyzed. The calculation firmly confirms that the results will be more precise if we choose more rational points on the beach. The application of BCM, available for the problems with irregular domains and arbitrary boundary conditions, can effectively avoid complex calculation and programming. It can be widely used in ocean engineering.
The energy characteristics in the evolution of the wave train are investigated to understand the inherent cause of the freak wave generation. The Morlet wavelet spectrum method is employed to analyze the numerical, laboratory and field evolution data of this generation process. Their energy distributions and variations are discussed with consideration of corresponding surface elevations. Through comparing the energy characteristics of three cases, it is shown that the freak wave generation depends not only on the continuous transfer of wave train energy to a certain region where finally the maximum energy occurs, but also on the distinct shift of the converged energy to high-frequency components in a very short time. And the typical energy characteristics of freak waves are also given.
