Validation of extreme wave and crest height distribution

Abstract

This thesis has focussed on the validation of extreme wave and crest height distribution. The empirical distribution was simulated based on laboratory and field measurement water elevation data. Both sets of data indicated the presence of abnormal wave ensuring the extreme condition aimed at in this study. The literature review was performed to identify improved predictions of wave and crest height distribution. Wave height predictions which are involved in this study were based on a modified Weibull distribution. Along with these predictions, a later formula that involved nonlinear factor of wave steepness is also applied. Meanwhile, nonlinear factors have been increasingly involved in crest height prediction. The modified formulae which are compared include the nonlinear factor of water depth, wave steepness, and directional factor. In all cases, standard linear Rayleigh distribution is referred to in relation to how good the improved formulae are in predicting the distributions. Based on wave height validation, Rayleigh shows better accuracy than the modified Weibull distributions. The standard Rayleigh distribution seems to fit the laboratory data, but deviates more widely from the field measurement data. The inadequacy of Rayleigh based on field validation showed the need of better prediction of nonlinear wave height in nature. Validation showed that the newly developed Rayleigh-Stokes prediction comes out with a slightly better prediction. Nonetheless, it still largely deviates from the observed distribution. On the other hand, the inadequacy of the Rayleigh distribution is seen very clearly in relation to crest height validation. Newer nonlinear formulae are found to give a better prediction showing a stronger nonlinearity affect on crest height in nature. However these models show discrepancies from one another. It is possibly caused by the different methods underpinning the development of these formulae and the way nonlinear factors are included in their prediction