A numerical method based on a fully-coupled solution of 2D Navier- Stokes equations with non-linear free surface boundary conditions is used to compute propagation of waves up to breaking on a sloping beach. The interface tracking is based on a moving grid technique. The purpose of this paper is to study the capability of such a numerical method to properly detect the inception of wave breaking. This objective is addressed by studying three different breaking criteria.
The wave breaking is a very significant phenomenon in many water waves processes and especially in coastal engineering. It can cause wave damping, currents, can throw sediments into suspension and put them in offshore or onshore movements. During the last decades significant advances have been done concerning both experimental and theoretical aspects of wave breaking. With the growth of computer power two or three-dimensional numerical wave tanks (NWTs) based on potential flow and more recently on viscous flow theory have been developed by many researchers. These softwares give very good and interesting results in a number of applications such as : periodic wave or solitary wave propagation, deep water or shallow water cases, wave diffraction on submerged or free-surface piercing bodies … NWTs using potential flow theory are usually based on Boundary Elements Methods. They are efficient for the computation of wave propagation and shoaling over a slope up to the wave overturning. Contributions of Longuet-Higgins and Cokelet (1976), Tullin and Cointe (1988) or Grilli et al.(1997) for 2D cases can be mentioned among many others. Use of Lagrangian markers can lead simulations to go a bit further but the flow after wave breaking cannot be calculated and the simulation is stopped. Some authors (Subramani et al., 1998) use a criterion able to detect likelihood of wave breaking.
