Abstract
Traditional coastal protection solutions, such as dykes and groynes, are now recognised as having a negative impact on the environment. In such a context, soft solutions have emerged with the aim to integrate environmental considerations into their conception. Soft solutions include hybrid solutions (combining traditional and natural structures), natural habitats (coral, seagrass, mangroves) and biomimetic solutions. The design and deployment of these solutions depend on a deep understanding of their control over hydrodynamical processes. The aim of this thesis is to study the impact of different soft solutions on wave height dissipation and to propose a new formalism that improves the representation of a complex, flexible structure in traditionnal models.Historically, the theory of wave dissipation on soft solutions supposed that the structure can be approximated by a rigid cylinder, even if it is flexible. Although more recent methods attempt to represent the motion of a single flexible structure, there is no universal formalism that links this motion to dissipation. The application of wave dissipation theory requires knowledge of the value of the drag coefficient. To date, the drag coefficient is the only parameter that can be fully determined in the form of a law using experimental data. The empirical laws obtained are then incorporated into numerical models.Three in-situ experimentations were carried out on different soft solutions to improve understanding of the wave dissipation mechanisms. The results show that the three solutions studied all dissipate wave height in function of wave frequency. Short waves tend to be more easily dissipated, while long waves can gain energy. Experimental measurements are used to generate drag coefficient empirical laws. These new laws, integrated into the theory of dissipation, can estimate the average dissipation of wave height through the soft solutions. However, in in-situ conditions, the diversity of meteorological and marine forcings increases the spread of drag coefficient values, which empirical laws struggle to represent.The representation of soft protection solutions in numerical models is essential to improve their design. The drag coefficient laws defined experimentally are essential for setting up such a model. The lack of variability of the drag coefficient in empirical laws does not allow robust results to be obtained for different forcings. Two new methods for calculating drag coefficient are therefore proposed in this thesis. The first proposes to define an empirical law based on dissipation intervals. The second is based on the force balance to propose a new drag coefficient formula. The methods are tested and validated on the basis of measured data and the literature. Although the results are encouraging, further applications and tests are still needed to improve the robustness of these methods.