Study Of Freak Waves Based On Numerical Ocean Waves Simulation

Freak wave is one of the most concerned issues in the ocean wave studies in recent decades, on which the studies are still in a preliminary stage.There are a lot of problems to be solveed. At first, in this dissertation the status of the-state of-art has been reviewed concerning the linear and nonlinear generation mechanism of freak waves proposed so far.The third generation ocean wave model, WAVEWATCH III (WW3),is applied to investigate the linear generation mechanism of freak waves due to the effect of currents for an ideal case. It is shown that huge wave could be generated by the modulation of non-uniform current. The waves tend to be convergent as the waves propagate against the current flow, while they tend to be divergent as the waves propagate in the direction of the flow. It is well known that the instability of ocean wave could arise not only from the wind forcing, but also from the interactions among its components.The well-known classical theory concerning water wave instability is the so-called B-F instability mechanism proposed by Benjamin and Feir(1967).At present this mechanism is considered as the most promising theory to understand the generation of freak waves due to nonlinear effect. The third-order nonlinear Schrodinger (NLS) equation, which could quite well discribe the B-F instability, has been derived in detail by multiple scale method. This thesis gives a detailed derivation process of the third-order nonlinear Schrodinger (NLS) equation using multiple scales method,which can discribe the B-F instability quite well.The derived NLS equation predicts the conditions in which freak waves tend to occur.Up to now very few field observations of freak waves are available, most of which is in the form of time series measured at single points.This kind of field observation could not reveal the characteristics of the evolution of the wave field as a whole. Recently the third generation wave models haved been applied to investigate the freak waves from the viewpoint of wave field evolution.In the present dissertation, WW3 wave model is used to investigate the statistical characteristics of wave attributes as freak waves occur. First of all, WW3 model is quantitatively validated with observational wave data, which is driven by QSCAT/NCEP blended wind field. Compared to significant wave height (SWH) measured by buoy and altimeter satellite, the relative error of the modeled SWH is about 15%,its absolute mean error is 0.49m, and the correlation coefficient between the modeled and measured SWHs is 0.89. WW3 model also showed a quite good performance in simulating the wave field with the presence of swell, and the unimodal and bimodal wave spectra could be reproduced fairly. The wave evolution is simulated with the forecast wind field, both for a general wave process in North-West Pacific over complex topography and for Morakot typhoon process.Compared with measured data, the relative errors of the modeled SWH is about 20% and 18% for the two processes respectively.Chosen from historical literature and shipwreck reports in which the moment and location of freak wave occurrence are clearly reported, six freak wave events have been studied. The wave field is simulated using WW3 model by multiple nesting technique for the chosen freak wave events.It is shown that the freak waves always occur, when the sea state is severe, wave steepness is large, directional spreading is small, spectral width is narrow, spectral kurtosis is large and four-wave nonlinear interaction play a leading role on energy balance at low frequencies.The temporal-spatial evolution characteristics of wave attributes have been analyzed in detail as freak wave occurs, and it is found that the correlation between the freak wave occurrence and that single wave attribute is not obviously evident.The Benjiemin-Feir Index (BFI) factor, which is the measure of probability of freak wave occurrence proposed by Janssen (2003),is computed with the WW3 model results.It is shown that the values of BFI are all smaller than 0.4, which is far away from the criterion for water wave instability that BFI>1 predicted by Janssen (2003)with the laboratory experiment results for one dimensional case. This means that the criterion is not applicable to the waves in the ocean.On the basis of the analysis to the scatter diagram of the joint distribution of wave steepness and directional spreading during the freak wave occurrence, it is found the freak waves always tend to occur at or the points of the inflexion in the scatter diagram, at which the data points are high densely distributed. At these points of the inflexion, the wave steepness is large, with the value above 0.08,while the directional spreading is small between 20～27 degrees.To some sense, the large wave steepness and small directional spreading are necessary conditions for freak wave occurrence.This thesis also gives an ocean wave simulation of Krosa Typhoon which passed through China in 2007 with WAVEWATCH III wave model.The ieal Typhoon wind filed is obtained according to Jelesnianski1965 ideal Typhoon model.18 points are taked for the research. The results show that most BFI values are less than 0.5,and the data points in the 0.09-0.1 range accounting for about 0.04% in the total sample, and the BFI values larger than 0.8 mainly appear in the initial stage of development of wave height. These results are the similar with the results obtained by Bertotti and Cavaleri (2008).The envolements of BFI, wave steepness, spectral kurtosis,direction angle, wave height and direction spreading for the 18 points, show that there are two fast-changing time segment. One is in the initial stage of wave height growth, and the other is the rapid increase stage. From the characteristics of wave height, buoy data shows good correspondence between the two times period of relatively high frequency freak wave occurs and the rapid changing time period of simulated wave parameters. Simulation results show that the BFI values have a remarkable positive correlation with the wave steepness, while less correlation with direction spreading. As the direction spreading in the ocean are always big, so it is not enough to measure the instability of ocean waves using a single BFI factor.