Wave Crest Sensor Intercomparison Study: An Overview of WACSIS

Author(s):  
George Z. Forristall ◽  
Stephen F. Barstow ◽  
Harald E. Krogstad ◽  
Marc Prevosto ◽  
Paul H. Taylor ◽  
...  

The Wave Crest Sensor Intercomparison Study (WACSIS) was designed as a thorough investigation of the statistical distribution of crest heights. Measurements were made in the southern North Sea during the winter of 1997–1998 from the Meetpost Noordwijk in 18 m water depth. The platform was outfitted with several popular wave sensors, including Saab and Marex radars, an EMI laser, a Baylor wave staff and a Vlissingen step gauge. Buoys were moored nearby to obtain directional spectra. Two video cameras viewed the ocean under the wave sensors and their signals were recorded digitally. The data analysis focused on comparisons of the crest height measurements from the various sensors and comparisons of the crest height distributions derived from the sensors and from theories. Some of the sensors had greater than expected energy at high frequencies. Once the measurements were filtered at 0.64 Hz, the Baylor, EMI and Vlissingen crest height distributions matched quite closely, while those from the other sensors were a few percent higher. The Baylor and EMI crest distributions agreed very well with the statistics from second order simulations, while previous parameterizations of the crest height distribution were generally too high. We conclude that crest height distributions derived from second order simulations can be used with confidence in engineering calculations. The data were also used in investigations of crest and trough shapes and the joint height/period distribution.

2004 ◽  
Vol 126 (1) ◽  
pp. 26-34 ◽  
Author(s):  
George Z. Forristall ◽  
Stephen F. Barstow ◽  
Harald E. Krogstad ◽  
Marc Prevosto ◽  
Paul H. Taylor ◽  
...  

The Wave Crest Sensor Intercomparison Study (WACSIS) was designed as a thorough investigation of the statistical distribution of crest heights. Measurements were made in the southern North Sea during the winter of 1997–1998 from the Meetpost Noordwijk in 18 m water depth. The platform was outfitted with several popular wave sensors, including Saab and Marex radars, an EMI laser, a Baylor wave staff and a Vlissingen step gauge. Buoys were moored nearby to obtain directional spectra. Two video cameras viewed the ocean under the wave sensors and their signals were recorded digitally. The data analysis focused on comparisons of the crest height measurements from the various sensors and comparisons of the crest height distributions derived from the sensors and from theories. Some of the sensors had greater than expected energy at high frequencies. Once the measurements were filtered at 0.64 Hz, the Baylor, EMI and Vlissingen crest height distributions matched quite closely, while those from the other sensors were a few percent higher. The Baylor and EMI crest distributions agreed very well with the statistics from second order simulations, while previous parameterizations of the crest height distribution were generally too high. We conclude that crest height distributions derived from second order simulations can be used with confidence in engineering calculations. The data were also used in investigations of crest and trough shapes and the joint height/period distribution.


Author(s):  
Bas Buchner ◽  
George Forristall ◽  
Kevin Ewans ◽  
Marios Christou ◽  
Janou Hennig

The objective of the CresT JIP was ‘to develop models for realistic extreme waves and a design methodology for the loading and response of floating platforms’. Within this objective the central question was: ‘What is the highest (most critical) wave crest that will be encountered by my platform in its lifetime?’ Based on the presented results for long and short-crested numerical, field and basin results in the paper, it can be concluded that the statistics of long-crested waves are different than those of short-crested waves. But also short-crested waves show a trend to reach crest heights above second order. This is in line with visual observations of the physics involved: crests are sharper than predicted by second order, waves are asymmetric (fronts are steeper) and waves are breaking. Although the development of extreme waves within short-crested sea states still needs further investigation (including the counteracting effect of breaking), at the end of the CresT project the following procedure for taking into account extreme waves in platform design is recommended: 1. For the wave height distribution, use the Forristall distribution (Forristall, 1978). 2. For the crest height distribution, use 2nd order distribution as basis. 3. Both the basin and field measurements show crest heights higher than predicted by second order theory for steeper sea states. It is therefore recommended to apply a correction to the second order distribution based on the basin results. 4. Account for the sampling variability at the tail of the distribution (and resulting remaining possibility of higher crests than given by the corrected second order distribution) in the reliability analysis. 5. Consider the fact that the maximum crest height under a complete platform deck can be considerably higher than the maximum crest at a single point.


2020 ◽  
Vol 197 ◽  
pp. 106899 ◽  
Author(s):  
V. Sanil Kumar ◽  
S. Harikrishnan ◽  
Sourav Mandal

Author(s):  
Marc Prevosto ◽  
Geoerge Z. Forristall

The analysis phase of the Wave Crest Sensor Intercomparison Study (WACSIS) focussed on the interpretation of the wave data collected by the project during the winter of 1997–98. Many aspects of wave statistics have been studied, but the main emphasis has been on crest height distributions, and recommendations for crest heights to be used in air gap calculations. In this paper we first describe comparisons of the crest height distributions derived from the sensors (radars, wave staffs, laser) and from simulations based on 3D second order irregular wave models. These comparisons permit us to make conclusions on the quality of these models and to qualify the ability of some sensors to measure the crest heights accurately. In the second part two new parametric models of the crest height distributions are discussed and their superiority to standard parametric models is demonstrated.


Author(s):  
Felice Arena ◽  
Alfredo Ascanelli

The interest and the studies on nonlinear waves are increased recently for their importance in the interaction with floating and fixed bodies. It is also well known that nonlinearities influence wave crest and wave trough distributions, both deviating from Rayleigh law. In this paper a theoretical crest distribution is obtained taking into account the extension of Boccotti’s Quasi Determinism theory, up to the second order for the case of three-dimensional waves, in finite water depth. To this purpose the Fedele & Arena [2005] distribution is generalized to three-dimensional waves on an arbitrary water depth. The comparison with Forristall second order model shows the theoretical confirmation of his conclusion: the crest distribution in deep water for long-crested and short crested waves are very close to each other; in shallow water the crest heights in three dimensional waves are greater than values given by long-crested model.


2018 ◽  
Vol 140 (3) ◽  
Author(s):  
Muk Chen Ong ◽  
Dag Myrhaug

This paper provides a practical stochastic method by which the burial and scour depths of short cylinders and truncated cones exposed to long-crested (two-dimensional (2D)) and short-crested (three-dimensional (3D)) nonlinear random waves plus currents can be derived. The approach is based on assuming the waves to be a stationary narrow-band random process, adopting the Forristall second-order wave crest height distribution representing both 2D and 3D nonlinear random waves. Moreover, the formulas for the burial and the scour depths for regular waves plus currents presented by previous published work for short cylinders and truncated cones are used.


Author(s):  
Cuilin Li ◽  
Dingyong Yu ◽  
Yangyang Gao ◽  
Junxian Yang

Many empirical and theoretical distribution functions for wave crest heights have been proposed, but there is a lack of agreement. With the development of ocean exploitation, waves crest heights represent a key point in the design of coastal structures, both fixed and floating, for shoreline protection and flood prevention. Waves crest height is the dominant parameter in assessing the likelihood of wave-in-deck impact and its resulting severe damage. Unlike wave heights, wave crests generally appear to be affected by nonlinearities; therefore, linear wave theory could not be satisfied to practical application. It is great significant to estimate a new nonlinear wave crest height distribution model correctly. This paper derives an approximation distribution formula based on Stokes wave theory. The resulting theoretical forms for nonlinear wave crest are compared with observed data and discussed in detail. The results are shown to be in good agreement. Furthermore, the results indicate that the new theoretical distribution has more accurate than other methods presented in this paper (e.g. Rayleigh distribution and Weibull distribution) and appears to have a greater range of applicability.


Author(s):  
Felice Arena ◽  
Alfredo Ascanelli

The interest and studies on nonlinear waves are increased recently for their importance in the interaction with floating and fixed bodies. It is also well-known that nonlinearities influence wave crest and wave trough distributions, both deviating from the Rayleigh law. In this paper, a theoretical crest distribution is obtained, taking into account the extension of Boccotti’s quasideterminism theory (1982, “On Ocean Waves With High Crests,” Meccanica, 17, pp. 16–19), up to the second order for the case of three-dimensional waves in finite water depth. To this purpose, the Fedele and Arena (2005, “Weakly Nonlinear Statistics of High Random Waves,” Phys. Fluids, 17(026601), pp. 1–10) distribution is generalized to three-dimensional waves on an arbitrary water depth. The comparison with Forristall’s second order model (2000, “Wave Crest Distributions: Observations and Second-Order Theory,” J. Phys. Oceanogr., 30(8), pp. 1931–1943) shows the theoretical confirmation of his conclusion: The crest distribution in deep water for long-crested and short-crested waves are very close to each other; in shallow water the crest heights in three-dimensional waves are greater than values given by the long-crested model.


2004 ◽  
Vol 126 (1) ◽  
pp. 43-50 ◽  
Author(s):  
Marc Prevosto ◽  
George Z. Forristall

The analysis phase of the Wave Crest Sensor Intercomparison Study (WACSIS) focussed on the interpretation of the wave data collected by the project during the winter of 1997–98. Many aspects of wave statistics have been studied, but the main emphasis has been on crest height distributions, and recommendations for crest heights to be used in air gap calculations. In this paper, we first describe comparisons of the crest height distributions derived from the sensors (radars, wave staffs, laser) and from simulations based on 3-D second-order irregular wave models. These comparisons permit us to make conclusions on the quality of these models and to qualify the ability of some sensors to measure the crest heights accurately. In the second part, two new parametric models of the crest height distributions are discussed and their superiority to standard parametric models is demonstrated.


Author(s):  
Se´bastien Fouques ◽  
Carl Trygve Stansberg

Wave crest height and steepness are crucial parameters for the design of ships and offshore structures. For irregular sea states, they are commonly predicted by using linear wave theory and a Eulerian description of the fluid motion. This theory only applies when the wave steepness is small and it fails to capture extreme wave events. Such linear solutions can also be extended by including second-order terms in order to provide more realistic wave properties. The paper describes a model for irregular long-crested waves that is based on a modified linear solution derived from a Lagrangian description of the fluid, i.e. by considering the motion of individual fluid particles. Lagrangian solutions have the advantage of showing realistic wave profiles with sharp crests and broad troughs already at the first order, whereas these features only appear at the second order when using the Eulerian approach. Still, a severe drawback with the former is that the mass conservation is not fulfilled exactly. The aim of the modification in the present Lagrangian model is to ensure that the mass conservation is always fulfilled in the solution. This is done by using the second-order residual in the continuity equation to lift up the fluid particles vertically. Comparative investigations of wave properties such as the crest height and the wave steepness are further carried out by making use of both numerical case studies and wave tank recordings. The wave models used in the comparisons include linear and second-order Eulerian solutions as well as the modified linear Lagrangian one.


Sign in / Sign up

Export Citation Format

Share Document