Statistical Analysis of the Wave Runup at Walls in a Changing Climate by Means of Image Clustering

This contribution builds on an existing methodology of image clustering analysis, conceived for modelling the wave overtopping at dikes from video records of laboratory experiments. It presents new procedures and algorithms developed to extend this methodology to the representation of the wave runup at crown walls on top of smooth berms. The upgraded methodology overcomes the perspective distortion of the native images and deals with the unsteady, turbulent and bi-phase flow dynamics characterizing the wave impacts at the walls. It accurately reconstructs the free surface along the whole structure profile and allows for a statistical analysis of the wave runup in the time and spatial domain. The effects of different structural configurations are investigated to provide key information for the design of coastal defences. In particular, the effects of increased sea levels in climate change scenarios are analysed. Innovative results, such as profiling of the envelopes of the runup along the wall cross and front sections, and the evidencing of 3D effects on the runup are presented. The extreme runup is estimated for the definition of the design conditions, while the envelopes of the average and minimum runup heights are calculated to assess the normal exercise conditions of existing structures.