Caratterizzazione delle onde marine estreme in condizioni di mare incrociato

Wind waves drive many geophysical processes at the ocean-atmosphere interface, which have a large impact on the climate and weather, and affect recreational and productive human activities both offshore and along the coastlines. Understanding the behaviour of the highest waves is thus central to many offshore activities. During the last decades, much effort has been devoted in trying to accurately measure and model extreme waves. State-of-the-art stereoscopic imaging techniques employed at offshore platforms recently provided evidence that the maximum sea surface elevation occurring over a given area in a short-crested sea is larger than that occurring at a fixed location, as measured by a wave buoy, due to the inclusion of the wave group dynamics. A stochastic approach was thus developed to provide the expected maximum individual waves over a given sea surface area (space) and duration (time). Such space-time statistics were recently included in operational spectral wave models to provide accurate short-term predictions of extreme conditions, although they are still not fully suited to representing wave extremes in all realistic sea conditions. In the case of crossing seas, when more than one wave system is present at the same time and location, our understanding of the underlying physics is still incomplete and their accurate description is therefore lacking. The main aim of this doctoral thesis is to modify existing formulations in order to provide more physically consistent and reliable predictions of extremes in crossing sea conditions. To this end, the extreme crest height probability of different crossing seas during a Tropical Cyclone in the Yellow Sea was investigated using a phase-resolving numerical model, whose reliability was assessed for the first time against stereo wave measurements of the sea surface elevation from an offshore platform in the region of interest. Particular crossing conditions were found to lead to higher extreme crest heights compared to the statistics of the underlying wind sea or swell, due to strong second-order non resonant interactions between the two wave systems. In the light of this evidence, a relatively simple analytical formulation for the upper bound of the wave steepness, which relies on the characteristics of the two constituent partitions and also on the local crossing conditions, was proposed and tested. The novel steepness formulation was then combined with phase-averaged wave model spectra for the first characterization of maximum individual waves during three Tropical-Like Cyclones in the Mediterranean Sea with the aim of studying the influence of the various combinations of swell and wind wave partitions around the cyclone eye on the spatial distribution of extreme waves. New geographical regions of enhanced crest height values, compared to those predicted with current space-time statistics, were highlighted for the first time.