Abstract

Hurricane Ivan (September 2004) generated the highest waves ever measured or hindcast in the Gulf of Mexico. The implications of Ivan on deepwater Gulf of Mexico metocean design criteria are examined from three aspects:

  1. whether there is an obvious deficiency in the ODGP wind and wave hindcast model,

  2. whether Ivan generated "freak" conditions that break the NTS/API gust factor relationships and wind spectra characterizations or the Forristall wave height and crest height distributions, and

  3. whether Ivan is, from a statistical standpoint, truly an "unexpected" event.

To address the first two issues, wind and wave measurements collected on two industry platforms near Ivan's track are examined. Excellent comparisons are found between the industry measurements and the ODGP hindcast model, which further corroborate the findings of [2]. Analysis of the industry wind measurements suggests NTS/API gust factors are valid for Ivan, while the NTS/API spectra are found to agree reasonably well with the measured spectra except perhaps for a tendency of API to overestimate low-frequency energy. Measured wave heights and wave crests are found to fit the Forristall distributions quite well, providing no support for "freak" waves during Ivan. To address the last issue, the Gulf of Mexico hindcast record is examined. The maximum wave heights generated by Ivan correspond to roughly a 2,500-year local return period, assuming the extremal distribution before Ivan was correct. Adding Ivan to the existing hindcast database is found to raise 100-year Hs values close to the storm track by 1.0 m or more. Through consideration of hurricane encounters in a finer spatial sense (160 km zones) rather than simply considering the entire Gulf as a single encounter zone, a strong argument can be made that the conditions generated by Ivan could reasonably be expected at a local level within the context of 105 years of hurricane experience throughout the Gulf. However, while Ivan may be described as a rare but not unexpected event at the local level, prudence dictates that winds and waves generated by Ivan be included in the hindcast database used for future Gulf of Mexico designs, as opposed to treating them as local outliers to existing extremal distributions and neglecting them.

Introduction

Hurricane Ivan entered the Gulf of Mexico on September 14, 2004, and made landfall at Gulf Shores, Alabama on September 16. Its course took it over some of the most developed areas of the Mississippi Canyon, Viosca Knoll and Main Pass regions (Figure 1). During its passage through these regions, Ivan declined from (barely) Category 4 intensity to Category 3 at landfall. While not a particularly intense storm in Saffir-Simpson scale terms, during its passage through the developed areas Ivan generated the largest waves ever measured in the Gulf of Mexico, well in excess of those recommended for deepwater design by the 21st edition of API RP-2A [1].

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