No abstract

A method for specifying hurricane-generated waves in the Gulf of Mexico in terms of the directional wave spectrum has been developed. The method includes a description of the surface-wind field from historical meteorological data through the application of a dynamic-hurricane frictional boundary-layer model. Introduction One of the more critical factors affecting the design of an offshore structure is the estimate of the maximum wave condition that can be expected to affect the structure during its lifetime. In the Gulf of Mexico, the more severe wave conditions are associated with hurricanes. The maximum expected wave conditions are estimated using extreme-value statistical-analysis techniques. The wave data used in the extreme-value analyses must represent a statistically significant population of historical wave data. The use of population of historical wave data. The use of wave-hindcasting techniques in conjunction with the meteorological data that characterize historical storms is the most appropriate way of obtaining the required wave data. The hindcast approach has been applied in the Gulf of Mexico with wave-hindcast models that use the significant-wave concept. These models characterize sea state only in terms of the significant wave height and period. Such a study recently described by Bea used period. Such a study recently described by Bea used the deep-water hindcast model developed by Wilson as calibrated by Patterson. That hindcast model used descriptions of the hurricane wind field provided by the empirical model developed by the Office of Hydrology of the National Oceanic and Atmospheric Admin. To produce such a description, the wind model required only a specification of radius-to-maximum wind, magnitude of -storm pressure anomaly, the forward speed of the storm, and the storm track. These parameters are available for all storms that have crossed the U.S. coast line since 1900. The study reported here resulted in a state-of-the-art method for describing hurricane-generated winds and sea states. The development of the method dealt with both significant aspects of the problem-the hurricane wind-field model and the wave-hindcasting model. The wind-field model is based on a fundamental description of the interaction between the atmospheric pressure field and the wind field within a hurricane. However, even with this fundamental approach, the only input data required by the wind-field model is of the type available for historical hurricanes. The description of the sea state based on the concept of the directional wave spectrum. The hindcast model describes the growth, dissipation, propagation, and shallow-water modifications (refraction, shoaling, and bottom friction) of the waves in terms of the directional spectra. The approaches taken in the development of these models were made possible by three important factors. First, a significant amount of wind and wave measurements in the hurricane environment has become available for model calibration and validation as a result of the Ocean Data Gathering Program (ODGP). Second, recent advances in dynamic hurricane modeling allowed the application of a numerical model of the hurricane frictional boundary layer to the wind-specification problem. Finally, the hindcast model could build upon the problem. Finally, the hindcast model could build upon the framework of contemporary deep-water directional spectral models that have been applied so successfully within the past decade to extratropical weather systems. past decade to extratropical weather systems. JPT P. 385

No abstract

This paper reports the results of an extensive study of hurricane-generated waves in the Gulf of Mexico. The study includes hindcasting of historical hurricanes and statistical interpretation of the hindcast wave data to determine maximum wave heights throughout the OCS of the central. and western Gulf of Mexico. Introduction The maximum wave height that reasonably can be expected to affect a structure during its lifetime is one of the more critical factors influencing the design of an offshore structure. By definition, such a wave height is a rare occurrence. The estimation of future occurrences of such waves must reflect, and indeed be derived from, the past history of storm-wave occurrences. Long-term past history of storm-wave occurrences. Long-term historical data bases are required to estimate these rare occurrences with meaningful reliability. Thus, estimating rare future occurrences poses two basic problems:developing a sufficiently long-term historical data base andinterpreting the historical data base to provide reliable estimates of future rare occurrences. The first problem can be solved effectively with hindcasting models. Hindcasting models can be developed and calibrated with contemporary data. These models then are applied with historical storm meteorological data to "re-create" an historical storm-wave data base. In this manner, the need for a long-term historical data base is satisfied by meteorological data.This paper emphasizes the solution of the second problem, which can be stated more concisely: Given an problem, which can be stated more concisely: Given an historical storm-wave data base, what is the most appropriate method for determining the cumulative probability distribution function for the largest waves probability distribution function for the largest waves that will affect a site in tY years? From that distribution function, a variety of wave statistics useful in design studies (e.g., expected maximum wave height in tY years or return-period wave heights) can be determined.This paper describes a study of hurricane-generated waves along the central and western OCS of the Gulf of Mexico. A hurricane climatology characterizing the geographical distribution, wave-generating capability, and frequency of severe hurricanes was derived from historical data. Based on this climatology, a model for estimating the cumulative probability distribution function for the largest wave in tY years from historical wave data was developed. The historical wave data base was generated by hindcasting the more severe historical hurricanes that occurred between 1900–74. Finally, the model was applied with the hindcast data and sample results are presented. Hindcasting Historical Hurricanes The historical wave data base required for this study was developed by hindcasting severe historical hurricanes. The hindcast model used has been described before and will be discussed only briefly here. Hindcast Model The hurricane hindcast model actually consists of two models - a wind-field model and a wave model. The wind-field model is based on a fundamental theoretical description of the relationship between the atmospheric pressure field and winds within a hurricane. The pressure pressure field and winds within a hurricane. The pressure field is described from historical meteorological data. JPT P. 632

PermlSSlon to copy lS restrlcted to an abstract of not more than 300 words. Illustrations may'not be copied. Such us~of an abstract should contain conspicuous acknowledgment of'where and by~ho~the paper lS presented.ABSTRACT A significant advancement in describing storm generated waves has been achieved with the completion of a method for predicting hurricane generated waves in the Gulf of Mexico. This method provides a more accurate and complete description of waves through the use of directional wave spectra concepts. Data from a number of recent hurricanes were used in the calibration and verification of the method. The correlation between measured and predicted data is very satisfactory.