A methodology for the analysis of damage progression in rubble mound breakwaters

Nowadays, risk based designs as well as reliable rehabilitation and maintenance strategies are essential when dealing with coastal structures. In this sense, the probability of failure due to instability of the armour layer is one of the main issues in rubble mound breakwaters, and so is improving the knowledge on its deterioration rate. Both stability and damage progression on rubble mound breakwaters have been studied for more than 80 years, using different approaches, under regular/irregular waves and testing diverse geometries and armor units. However, due to the complexity of the processes involved, there are some modelling difficulties and the applicability of the different proposals is usually restricted to the typologies and specific range of the parameters tested in laboratory for deriving their empirical coefficients. Also, there are not many experimental results focused on the stochastic nature of damage progression, which is indeed an essential part of the problem. Despite being relatively common in coastal laboratories, a lack of standards for damage progression experiments in rubble mound breakwaters has been perceived. What is more, the concept of damage itself and how it is measured seems to be a debatable issue, specially taking into account the potential of the recent advances in scanning techniques for monitoring small scale model as well as prototypes. Regarding the large amount of information on this topic, part of the present dissertation is focused on stating a discussion on damage definition, describing the parameters influencing armor layer stability and providing a detailed historical overview of the most relevant models and contributions in the study of damage progression on rubble mound breakwaters. Also, a methodological approach for damage progression experiments is provided, describing, step-by-step, its main peculiarities and considerations. Following this approach, a set of experiments have been carried out in order to characterize the stochastic nature of damage progression in a quarry stone model, in which an innovative post-processing based on the results of a structured light scanner is presented. These results, together with the ones from Melby and Kobayashi (1999), are used for accomplishing an initial calibration of the damage progression probability model of Castillo et al. (2012), which was fully derived under probabilistic assumptions and following premises of general validity. A methodology for calibrating this model is also proposed and some preliminary conclusions derived from the results are proportioned. However, further experimentation covering a wider range of design parameters, ideally both in model and prototype, is needed for achieving a fully developed tool for assisting rubble mound breakwater design and conservation