An Intercomparison of Measured Wave Grouping Parameters During a Storm

Abstract

ABSTRACT This paper describes the results of a wave measurement experiment carried out at the Kinsale Head Gas field, off the South coast of Ireland between 21 and 23 November, 1984. Simultaneous wave recordings, of length up to two and a half hours, were made using wave buoy and wave staff devices. These recordings include the growth, peak and decay of a severe winter storm. A comparison is made between the wave parameters obtained from the two devices. This includes first order wave spectra and their derived height and period parameters as well as the most commonly used wave grouping parameters. The characteristic wave height and periods show good correlation between the wave buoy and staff while the chief grouping parameters demonstrate weaker dependence. The spectral peakedness parameter gave the strongest correlation but the mean run lengths were virtually un-correlated between the two locations. Each system gave similar values for the auto-correlation between successive wave heights with a weak but significant relationship. The Funke and Mansardgroupiness factor 1showed an even weaker dependency. This was slightly improved by employing a new modified factor. The agreement between SIWEH spectra was dependent on the length of the wave record. 1. INTRODUCTION With the development of compliant and floating oil production systems, wave groups are increasingly important to offshore engineers. Many authors have demonstrated the relationship between the slow drift motions of moored floating structures and the presence of wave modulation due to groups. Despite this relatively little is known about the occurrence of wave groups during storms. Attempts have been made to characterize nonrandom behavior of waves under two broad categories. Firstly, wave groups have been described in terms of the dependency of successive wave heights. The second category is the study of low frequency modulations of the sea surface. The former may be described as counting methods, wherein the wave heights are discretized and examined as a series of statistical events. The most common definition of a wave group is a run of consecutive waves whose heights exceed some threshold value. Figure 1 demonstrates this definition showing two wave, groups of length 3 and 4. Goda considers the mean run length of a record for threshold values of Hs ' the significant wave height, and Havg, the average wave height. Goda also demonstrated that the individual run containing the largest wave of the record is on average longer than other wave groups. This particular wave group is termed the "extreme wave group" (EWG) and has been proposed as a design criterion for capsize tests.