ABSTRACT

A five parameter wind model and a modified PTB wave model are outlined. The re3ults of tuning the wave model are shown in comparisons with one-dimensional wave spectra measured during the joint industry Ocean Data Gathering Program. An example application of hindcast results for 28 historical hurricanes to estimate extreme wave directional statistics is outlined. 100- and 50-year wave height contours for the continental shelf offshore Louisiana are presented.

INTRODUCTION

In 1970, Amoco initiated a hurricane wave hindcast study aimed at utilizing measured hurricane data obtained during the Ocean Data Gathering Project.1 The aim was to improve our understanding of hurricane waves and the statistics of their occurrence. The program was divided into three parts. The first, conducted in collaboration with the Institute for Storm Research, was aimed primarily at the hindcasting of waves in deep water. The second and third parts, conducted in-house, focused on the influence of shallow water in modifying the wave propagating toward shore, and on the processing of hindcast data for 28 hurricanes affecting the Louisiana offshore area during a 70year span to obtain extreme wave statistics. An overview description of the entire effort has been given in reference 2. The work was performed independently of, but is in many respects parallel to, the joint industry study reported in the 1975 OTC by Cardone, Pierson and Ward.3

This paper provides a technical outline of the hindcasting model, indicates the degree to which model calibration has allowed a matching with measured hurricane data, and provides examples of extreme wave statistic results for the Continental Shelf area, offshore Louisiana.

DEVELOPMENT OF HURRICANE WIND FIELD MODEL

The approach taken was to expand upon the framework of the simple type of parameterized model of cyclonic surface winds in a hurricane. By and large, such models are characterized as three parameter models involving the following physical quantities: 4,5

  • Rm = radius to the region of maximum winds

  • ?P= barometric depression of the eye

  • Vf = forward velocity of the eye.

Such three-parameter models involve assumptions of irrotational flow outside of Rm and rigid body rotation of the fluid inside Rm' Some investigators have further modified such models using mathematical curve-fitting to obtain closer approximations to measured wind fields.6,7 However, pure curve-fitting of data without a theoretical framework has the disadvantages that it requires relatively more data to perform a curve-fitting, that it is difficult to distinguish bad data points, and that the resulting wind field may not be compatible with basic physical laws governing atmospheric flows.

This content is only available via PDF.
You can access this article if you purchase or spend a download.