Breakwaters are built in coastal zones to alter sediment transport, or protect threatened habitats and port facilities. The rising sea level is causing more water to overtop these structures. Increasing amounts of overtopping discharge can compromise the security of people, or equipment standing on the crest of breakwaters. The overtopping flow causes a peak force that drives an instability on these subjects. The flow depths and velocities depend on the hydrodynamic conditions and structure geometry. Up to date, there is limited data that characterize this flow for a wide range of wave conditions and breakwater configurations. In principle, it is possible to perform physical model tests to fill this gap, but these tests are expensive and time-consuming. Furthermore, extracting the flow depths and velocities from experiments is complex and most intrusive measuring instruments can alter the flow. An alternative or addition to physical model experiments is the use of numerical models. In the present research, a numerical model based in OpenFOAM® is used to evaluate the effects of different wave conditions and protrusion heights on the flow depths and velocities at the crest of a rubble mound breakwater.

The numerical model was validated with physical model tests performed on a rubble mound breakwater. Overall, the numerical model simulated the incident waves accurately, but overestimated the mean overtopping discharge. The overprediction of the overtopping discharge and the different methodologies of computation of the flow velocities made it difficult to validate the overtopping flow. However, the numerical model was still valuable to study the physical processes occurring during overtopping events, and the trends on the modelled flow depths and velocities when changing the wave conditions and protrusion heights.

It was determined that wave gauges and probes are the optimal methods to extract the flow depths and velocities from the numerical model. They were placed along (and over) the crest. For the specific set-up of the numerical model, it was found that the most extreme events impact the horizontal part of the crest wall in between the measuring devices. Therefore, for some instruments, the flow depths and velocities are extracted when the events are still in the air or at the moment of collision with the crest. In these cases, the trends had a different behavior than the expected one once the events are propagating attached to and along the crest. More detailed analysis and future validation are required for such circumstances.

Events associated to high exceedance probabilities showed trends aligned with the expected tendencies (for events propagating attached to the crest). This is because, these events were produced by smaller wave heights. Hence, the overtopping events collided with the horizontal part of the crest wall before the first measuring devices. For these events, it was observed that the flow depths and velocities decreased the lower the significant wave height, the larger the wave steepness, and the longer the distances from the seaward boundary. In addition, for a smaller protrusion height, more events were captured, and their flow depths and velocities were larger.