Non-stationary extreme value models to account for trends and shifts in the extreme wave climate due to climate change

The extreme values of wave climate data are of great interest in a number of different ocean engineering applications, including the design and operation of ships and offshore structures, marine energy generation, aquaculture and coastal installations. Typically, the return values of certain met-ocean parameters such as significant wave height are of particular importance. There exist many methods for estimating such return values, including the initial distribution approach, the block maxima approach and the peaks-over threshold approach. In a climate change perspective, projections of such return values to a future climate are of great importance for risk management and adaptation purposes. However, many approaches to extreme value modelling assume stationary conditions and it is not straightforward how to include non-stationarity of the extremes due to for example climate change. In this paper, various non-stationary GEV-models for significant wave height are developed that account for trends and shifts in the extreme wave climate due to climate change. These models are fitted to block maxima in a particular set of wave data obtained for a historical control period and two future projections for a future period corresponding to different emission scenarios. These models are used to investigate whether there are trends in the data within each period that influence the extreme value analysis and need to be taken into account. Moreover, it will be investigated whether there are significant inter-period shifts or trends in the extreme wave climate from the historical period to the future periods. The results from this study suggest that the intra-period trends are not statistically significant and that it might be reasonable to ignore these in extreme value analyses within each period. However, when it comes to comparing the different data sets, i.e. the historical period and the future projections, statistical significant inter-period changes are detected. Hence, the accumulated effect of a climatic trend may not be negligible over longer time periods. Interestingly enough, such statistically significant shifts are not detected if stationary extreme value models are fitted to each period separately. Therefore, the non-stationary extreme value models with inter-period shifts in the parameters are proposed as an alternative for extreme value modelling in a climate change perspective, in situations where historical data and future projections are available.     

Characterizing the seasonal and directional varying properties in a marine environment

With noticing an increasing number of failure events for offshore structures in the present days, it is now realized that modeling the marine environment especially for exceptional environmental conditions is quite important. It is recognized that a possible improvement in the traditional modeling of environmental characteristics, which are the basis for the load models for structural analysis and design, may be needed. In this paper, the seasonal and directional varying properties in modeling the ocean parameter, the wave height, are studied. The peak over threshold (POT) method is selected to model the extreme wave height by utilizing a non-stationary discrete statistical extreme model. The varying parameters are taken into account with a changing pattern to reflect the seasonal and directional dependent behavior. Both the magnitude and the occurrence rate of the extreme values are investigated. Detailed discussion on the continuity of the established model is also given. The importance of the proposed model is demonstrated in reliability analysis for a jacket structure. The sensitivity to the changing marine environment in reliability analyses is investigated.     

Characterizing the Seasonal and Directional Varying Properties in A Marine Environment

With noticing an increasing number of failure events for offshore structures in the present days, it is now realized that modeling the marine environment especially for exceptional environmental conditions is quite important. It is recognized that a possible improvement in the traditional modeling of environmental characteristics, which are the basis for the load models for structural analysis and design, may be needed. In this paper, the seasonal and directional varying properties in modeling the ocean parameter, the wave height, are studied. The peak over threshold (POT) method is selected to model the extreme wave height by utilizing a non-stationary discrete statistical extreme model. The varying parameters are taken into account with a changing pattern to reflect the seasonal and directional dependent behavior. Both the magnitude and the occurrence rate of the extreme values are investigated. Detailed discussion on the continuity of the established model is also given. The importance of the proposed model is demonstrated in reliability analysis for a jacket structure. The sensitivity to the changing marine environment in reliability analyses is investigated.     

极值波高建模中不确定性的模糊量化方法

A non-traditional fuzzy quantification method is presented in the modeling of an extreme significant wave height. First, a set of parametric models are selected to fit time series data for the significant wave height and the extrapolation for extremes are obtained based on high quantile estimations. The quality of these results is compared and discussed. Then, the proposed fuzzy model, which combines Poisson process and generalized Pareto distribution(GPD) model, is applied to characterizing the wave extremes in the time series data. The estimations for a long-term return value are considered as time-varying as a threshold is regarded as non-stationary. The estimated intervals coupled with the fuzzy theory are then introduced to construct the probability bounds for the return values. This nontraditional model is analyzed in comparison with the traditional model in the degree of conservatism for the long-term estimate. The impact on the fuzzy bounds of extreme estimations from the non stationary effect in the proposed model is also investigated.     

The effect of directionality on extreme wave design criteria

Sea state design criteria for offshore facilities are frequently provided by direction. For example, it is typical for return-period values of the significant wave height to be specified for each of eight 45° sectors in addition to the omni-directional case. However, it is important that these criteria be consistent so that the probability of exceedance of a given wave height from any direction derived from the directional values is the same as for the omni-directional value. As recently demonstrated by Forristall it is not sufficient simply to scale the directional values so that the value of the wave height from the most severe sector is the same as the omni-directional value.We develop an approach for establishing appropriate directional criteria and an associated omni-directional criterion for a specific location. The inherent directionality of sea states is used to develop a model for the directional dependence of distributions of storm maxima. The directional model is applied to the GOMOS data, and the distributional properties of the 100-year significant wave height are estimated and the implications for design discussed. An objective risk-cost approach is proposed for optimising directional criteria, while preserving overall reliability. Simulation studies are performed, using realistic extreme value assumptions, to quantify the uncertainties.     

Modelling the long-term distribution of significant wave height with the Beta and Gamma models

The paper suggests modelling the long-term distribution of significant wave height with the Gamma, Beta of the first and second kind models. The three models are interrelated, flexible and cover the three different tail types of Extreme Value Theory. They can be used simultaneously as a means of assessing the uncertainty effects that result from choosing equally plausible models with different tail types. This procedure is intended for those applications that require the long-term distribution of significant wave height as input rather than the prediction of extreme values. The models are fitted to some significant wave data as an illustration. Details about maximum likelihood estimation are given in A.     

Statistical estimation of extreme ocean environments: The requirement for modelling directionality and other covariate effects

With increasing availability of good directional data, provision of directional estimates of extreme significant wave heights, in addition to the omni-directional estimates, is more common. However, interpretation of directional together with omni-directional design criteria is subject to inconsistency, even in design guidelines. In particular, omni-directional criteria are usually estimated ignoring directional effects. In this article, for data which exhibit directional effects, we show that a directional extreme value model generally explains the observed variation significantly better than a model which ignores directionality, and that omni-directional criteria developed from a directional model are different from those generated when directionality is not accounted for. We also show that omni-directional criteria derived from a directional model are more accurate and should be preferred in general over those based on models which ignore directional effects. We recommend use of directional extreme value models for estimation of both directional and omni-directional design criteria in future, when good directional data are available. If effects of other covariates (e.g. time or space) are suspected, we similarly recommend use of extreme value models which adequately capture sources of covariate variability for all design analysis.     

基于WW3的南海北部有效波高时空变化及其极值重现期估算方法分析

本文基于第3代海浪模式WAVEWATCH Ⅲ (WW3）模拟的1996–2015年海浪后报数据,分析了南海北部有效波高及其极值的时空变化特征,并采用Pearson-Ⅲ和Gumbel两种极值分布方法对该区极值波高重现期进行了估算。结果表明,南海北部有效波高的季节变化和空间分布与季风风场基本一致,呈现秋冬高春夏低,并自吕宋海峡西侧向西南降低的特征,与ERA5再分析数据结果高度相似。有效波高极值（简称极值波高）的时空分布特征受时间分辨率强烈影响,采用极值数据的分辨率越高（如逐小时）,所展现的台风型波浪特征越显著。扣除季节变化信号后的有效波高和年极值波高均体现出较强的线性增高趋势,近20年升高的比例分别为7.7%和31.6%,值得警惕和关注。该区多年一遇极值波高存在若干个大值区,且与台风的路径、强度有直接联系,表明台风是引发该区域极端大浪的最主要机制。对比Pearson-Ⅲ和Gumbel极值分布估算结果发现：若极值波高较低,频率随极值波高升高缓慢降低,此时两种极值分布的估算都比较准确,差异极小,可忽略不计;但当研究时间范围内,某年极值波高远超其他年份时,Pearson-Ⅲ极值分布估算结果明显高...     

The influence of seasonality on estimating return values of significant wave height

A time-dependent generalized extreme value (GEV) model for monthly significant wave heights maxima is developed. The model is applied to several 3-hour time series from the Spanish buoy network. Monthly maxima show a clear non-stationary behavior within a year, suggesting that the location, scale and shape parameters of the GEV distribution can be parameterized using harmonic functions. To avoid a possible over-parameterization, an automatic selection model, based on the Akaike Information Criterion, is carried out. Results show that the non-stationary behavior of monthly maxima significant wave height is adequately modeled, drastically increasing the significance of the parameters involved and reducing the uncertainty in the return level estimation. The model provides new information to analyze the seasonal behavior of wave height extremes affecting different natural coastal processes.     

Methods of fitting the Fisher-Tippett type 1 extreme value distribution

Methods are described for estimating the parameters of the Fisher-Tippet Type 1 extreme value distribution and associated return values from measured extremes, such as maximum wave height. A comparison of these methods, with simulated data, shows that those using Gumbel's plotting position are least satifactory. Maximum likelihood methods give the smallest mean square errors, but the very much simpler method of moments is nearly as good.     

Joint modelling of extreme ocean environments incorporating covariate effects

Characterising the joint distribution of extremes of ocean environmental variables such as significant wave height (H_S) and spectral peak period (T_P) is important for understanding extreme ocean environments and in the design and assessment of marine and coastal structures. Many applications of multivariate extreme value analysis adopt models that assume a particular form of extremal dependence between variables without justification. Models are also typically restricted to joint regions in which all variables are extreme, but regions where only a subset of variables is extreme can be equally important for design. The conditional extremes model of Heffernan and Tawn (2004) provides one approach to overcoming these difficulties.Here, we extend the conditional extremes model to incorporate covariate effects in all of threshold selection, marginal and dependence modelling. Quantile regression is used to select appropriate covariate-dependent extreme value thresholds. Marginal and dependence modelling of extremes is performed within a penalised likelihood framework, using a Fourier parameterisation of marginal and dependence model parameters, with cross-validation to estimate suitable model parameter roughness, and bootstrapping to estimate parameter uncertainty with respect to covariate.We illustrate the approach in application to joint modelling of storm peak H_S and T_P at a Northern North Sea location with storm direction as covariate. We evaluate the impact of incorporating directional effects on estimates for return values, including those of a structure variable, similar to the structural response of a floating structure. We believe the approach offers the ocean engineer a straightforward procedure, based on sound statistics, to incorporate covariate effects in estimation of joint extreme environmental conditions.     

A methodology for deriving extreme nearshore sea conditions for structural design and flood risk analysis

Extreme sea conditions in the nearshore zone are required for coastal flood risk analysis and structural design. Many multivariate extreme value methods that have been applied in the past have been limited by assumptions relating to the dependence structure in the extremes. A conditional extremes statistical model overcomes a number of these previous limitations. To apply the method in practice, a Monte Carlo sampling procedure is required whereby large samples of synthetically generated events are simulated. The use of Monte Carlo approaches, in combination with computationally intensive physical process models, can raise significant practical challenges in terms of computation. To overcome these challenges there has been extensive research into the use of meta-models. Meta-models are approximations of computationally intensive physical process models (simulators). They are derived by fitting functions to the outputs from simulators. Due to their simplified representation they are computationally more efficient than the simulators they approximate.Here, a methodology for deriving a large Monte Carlo sample of extreme nearshore sea states is described. The methodology comprises the generation of a large sample of offshore sea conditions using the conditional extremes model. A meta-model of the wave transformation process is then constructed. A clustering algorithm is used to aid the development of the meta-model. The large sample of offshore data is then transformed through to the nearshore using the meta-model. The resulting nearshore sea states can be used for the probabilistic design of structures or flood risk analysis. The application of the methodology to a case study site on the North Coast of Spain is described.     

Joint modelling of wave spectral parameters for extreme sea states

Characterising the dependence between extremes of wave spectral parameters such as significant wave height (H_S) and spectral peak period (T_P) is important in understanding extreme ocean environments and in the design and assessment of marine structures. For example, it is known that mean values of wave periods tend to increase with increasing storm intensity. Here we seek to characterise joint dependence in a straightforward manner, accessible to the ocean engineering community, using a statistically sound approach.Many methods of multivariate extreme value analyses are based on models which assume implicitly that in some joint tail region each parameter is either independent of or asymptotically dependent on other parameters; yet in reality the dependence structure in general is neither of these. The underpinning assumption of multivariate regular variation restricts these methods to estimation of joint regions in which all parameters are extreme; but regions where only a subset of parameters are extreme can be equally important for design. The conditional approach of Heffernan and Tawn (2004), similar in spirit to that of Haver (1985) but with better theoretical foundation, overcomes these difficulties.We use the conditional approach to characterise the dependence structure of H_S and T_P. The key elements of the procedure are: (1) marginal modelling for all parameters, (2) transformation of data to a common standard Gumbel marginal form, (3) modelling dependence between data for extremes of pairs of parameters using a form of regression, (4) simulation of long return periods to estimate joint extremes. We demonstrate the approach in application to measured and hindcast data from the Northern North Sea, the Gulf of Mexico and the North West Shelf of Australia. We also illustrate the use of data re-sampling techniques such as bootstrapping to estimate the uncertainty in marginal and dependence models and accommodate this uncertainty in extreme quantile estimation.We discuss the current approach in the context of other approaches to multivariate extreme value estimation popular in the ocean engineering community.     

A maximum-entropy compound distribution model for extreme wave heights of typhoon-affected sea areas

A new compound distribution model for extreme wave heights of typhoon-affected sea areas is proposed on the basis of the maximum-entropy principle.The new model is formed by nesting a discrete distribution in a continuous one,having eight parameters which can be determined in terms of observed data of typhoon occurrence-frequency and extreme wave heights by numerically solving two sets of equations derived in this paper.The model is examined by using it to predict the N-year return-period wave height at two hydrology stations in the Yellow Sea,and the predicted results are compared with those predicted by use of some other compound distribution models.Examinations and comparisons show that the model has some advantages for predicting the N-year return-period wave height in typhoon-affected sea areas.     

基于随机集合的非传统型有效波极值模型

The analysis and design of offshore structures necessitates the consideration of wave loads. Realistic modeling of wave loads is particularly important to ensure reliable performance of these structures. Among the available methods for the modeling of the extreme significant wave height on a statistical basis, the peak over threshold method has attracted most attention. This method employs Poisson process to character- ize time-varying properties in the parameters of an extreme value distribution. In this paper, the peak over threshold method is reviewed and extended to account for subjectivity in the modeling. The freedom in selecting the threshold and the time span to separate extremes from the original time series data is incorpo- rated as imprecision in the model. This leads to an extension from random variables to random sets in the probabilistic model for the extreme significant wave height. The extended model is also applied to different periods of the sampled data to evaluate the significance of the climatic conditions on the uncertainties of the parameters.     

Prediction of Extreme Significant Wave Height from Daily Maxima

For prediction of the extreme significant wave height in the ocean areas where long term wave data are not available, the empirical method of extrapolating short term data (1-3 years) is used in design practice. In this paper two methods are proposed to predict extreme significant wave height based on short-term daily maxima. According to the da-a recorded by the Oceanographic Station of Liaodong Bay at the Bohai Sea, it is supposed that daily maximum wave heights are statistically independent. The data show that daily maximum wave heights obey log-normal distribution, and that the numbers of daily maxima vary from year to year, obeying binomial distribution. Based on these statistical characteristics, the binomial-log-normal compound extremum distribution is derived for prediction of extreme significant wave heights (50-100 years). For examination of its accuracy and validity, the prediction of extreme wave heights is based on 12 years' data at this station, and based on each 3 years' data respectively     

The distribution of zero-crossing wave heights and periods in a stationary sea state

Existing theoretical distributions of wave height and period do not reflect measured joint distributions from field data. A simulation methodology is introduced to retain the essential features of the theoretical background in Gaussian random noise but to avoid further compromising assumptions in the interpretation of height and period in the amplitude domain. A joint distribution can be associated directly with an empirical or measured variance spectrum. Spectral shape appears to dominate the detail of predicted joint distributions. There is generally a much sharper decay in probability levels at higher periods than is predicted by theoretical models. For Jonswap spectra, there is a dominant central ridge and a distinct bimodal structure in the joint distribution, features that are not evident in symmetric Gaussian spectral forms. The wave height distributions for Jonswap spectra differ little from the Rayleigh distribution, except at extreme wave heights where Rayleigh overpredicts. The period distributions are strongly sensitive to spectral shape. In the conditional distribution of periods, given the height, the asymptotic median period at extreme wave heights is significantly longer than the mean period for Jonswap spectra, but not for symmetric Gaussian forms.     

Wave energy assessment based on trivariate distribution of significant wave height,mean period and direction

Based on the Vine copula theory, a trivariate statistical model of significant wave height, characterized wave period and mean wave direction was constructed. To maintain the properties of the different types of variables, a special copula function was derived from the model developed by Johnson and Wehrly based on the maximum entropy principle. It was then combined with the Archimedean copulas to construct the proposed model. An effective algorithm for generating corresponding joint pseudo-random numbers was also developed. Statistical analysis of hindcast data for the significant wave height, mean wave period, and direction, which were collected from an observation point in the North Atlantic every three hours from 1997 to 2001, was performed. The marginal distributions of the significant wave height and mean wave period were fitted by a modified maximum entropy distribution, and the mean wave direction was fitted by a mixture of von Mises distributions. It was shown that the proposed model is a good fit for the data. The seasonal wave energy resources in the target area were assessed using the model estimates. Histograms of the directional wave energy, wave energy roses, and scatter and energy diagrams were presented.     

Some statistical characteristics of large deepwater waves around the Korean Peninsula

The relationship between significant wave height and period, the variability of significant wave period, the spectral peak enhancement factor, and the directional spreading parameter of large deepwater waves around the Korean Peninsula have been investigated using various sources of wave measurement and hindcasting data. For very large waves comparable to design waves, it is recommended to use the average value of the empirical formulas proposed by Shore Protection Manual in 1977 and by Goda in 2003 for the relationship between significant wave height and period. The standard deviation of significant wave periods non-dimensionalized with respect to the mean value for a certain significant wave height varies between 0.04 and 0.21 with a typical value of 0.1 depending upon different regions and different ranges of significant wave heights. The probability density function of the peak enhancement factor is expressed as a lognormal distribution, with its mean value of 2.14, which is somewhat smaller than the value in the North Sea. For relatively large waves, the probability density function of the directional spreading parameter at peak frequency is also expressed as a lognormal distribution.     

海洋重现期波高统一化计算方法

重现期波高是港口海岸及海洋工程设计中不可回避的一个重要设计参数,尤其对深水海港、海上平台、海底油气管道、沿海核电站等重大涉海工程设计具有巨大的经济价值和深远的社会效益。但是,现有重现期波高推算缺乏统一的计算方法,导致计算结果相差悬殊。研究重现期波高的统一化计算方法,分析重现期波高计算中存在的各种不确定因素,提出减少这些不确定因素的新方法,建立误差小、应用方便、方法统一的重现期波高计算方法。基于澳大利亚悉尼站的长期连续观测波浪数据,研究发现:广义帕累托函数(generalized Pareto distribution III,GPD-III)和威布尔(Weibull)是重现期波高计算的最佳候选极值分布函数,新推导的函数形状参数计算公式较好提高重现期波高的计算精度,极值波高数据的分析方法和样本大小是影响重现期波高计算精确度的两个重要因素,短期波浪资料和年极值法可能高估重现期波高值。逐个风暴的极值波高数据分析法及最佳候选极值分布函数GPD-III和Weibull建议应用于涉海工程设计的重现期波高推算。