Extreme waves in coastal areas : landslide tsunami modeling and storm waves long term statistcs

Abstract

Nowadays many coastal disaster have been experienced all around the world and some of them can be linked to extreme events, as flooding, extreme sea waves, tsunamis, etc. In this thesis two different coastal extreme events, landslide generated tsunamis and extreme sea waves, have been studied because of the high risk related to their occurrence and to the high economical relevance of coastal areas. Because of the different rate of occurrence of the two kind of events, different approaches have been used. The commercial fully 3D numerical model, FLOW 3D, has been used to study landslide generated tsunamis. In particular it has been used to model and analyze fluid-structure interaction and also the generation and propagation in the near and far field of the tsunamis. The model has been firstly successfully validated, comparing the results of numerical simulations to the results of 2D and 3D physical experiments, in particular focusing on landslide generated tsunamis around a conical island. Then the spatial distribution of the generated run-up around the island has been analyzed. The high accuracy achieved suggests the use of the model for few, accurate precomputed inundation maps. The model has also been used to carry out parametric analysis, showing how different parameters influence landslide generated tsunamis and how the numerical model allows to obtain numerous and reliable data with reasonable computational costs. It can finally be noted that the model could be considered an helpful tool for an Early Warning System of tsunami events. The analysis of extreme sea waves has been carried out considering the fact that almost 20 years of recorded data along the Italian coast are available and so that an accurate extreme statistical analysis can be carried out. In this thesis two different statistical distributions have been applied on the data recorded by eight RON (Rete Ondametrica Nazionale) buoys. In particular a stationary and then a no-stationary GPD-Poisson model have been used. The first one has given results similar to the one obtained with the statistical analysis reported on the Italian Wave Atlas (2004),the second has analyzed the intra-annual variability of extreme sea waves. This analysis has been carried out considering the statistical distribution parameters as time dependent, varying throughout a year. This procedure allows a better explanation of the variability of extreme significant wave heights throughout a year, its seasonality, and could help, for instance, to define working windows for maritime works in the open sea or to define design waves for seasonal structure.