Search Results
You are looking at 1 - 10 of 57 items for
- Author or Editor: David W. Wang x
- Refine by Access: All Content x
Abstract
It has been recognized that modulated wave groups trigger wave breaking and generate energy dissipation events on the ocean surface. Quantitative examination of wave-breaking events and associated turbulent kinetic energy (TKE) dissipation rates within a modulated wave group in the open ocean is not a trivial task. To address this challenging topic, a set of laboratory experiments was carried out in an outdoor facility, the Oil and Hazardous Material Simulated Environment Test Tank (203 m long, 20 m wide, 3.5 m deep). TKE dissipation rates at multiple depths were estimated directly while moving the sensor platform at a speed of about 0.53 m s−1 toward incoming wave groups generated by the wave maker. The largest TKE dissipation rates and significant whitecaps were found at or near the center of wave groups where steepening waves approached the geometric limit of waves. The TKE dissipation rate was O(10−2) W kg−1 during wave breaking, which is two to three orders of magnitude larger than before and after wave breaking. The enhanced TKE dissipation rate was limited to a layer of half the wave height in depth. Observations indicate that the impact of wave breaking was not significant at depths deeper than one wave height from the surface. The TKE dissipation rate of breaking waves within wave groups can be parameterized by local wave phase speed with a proportionality breaking strength coefficient dependent on local steepness. The characterization of energy dissipation in wave groups from local wave properties will enable a better determination of near-surface TKE dissipation of breaking waves.
Abstract
It has been recognized that modulated wave groups trigger wave breaking and generate energy dissipation events on the ocean surface. Quantitative examination of wave-breaking events and associated turbulent kinetic energy (TKE) dissipation rates within a modulated wave group in the open ocean is not a trivial task. To address this challenging topic, a set of laboratory experiments was carried out in an outdoor facility, the Oil and Hazardous Material Simulated Environment Test Tank (203 m long, 20 m wide, 3.5 m deep). TKE dissipation rates at multiple depths were estimated directly while moving the sensor platform at a speed of about 0.53 m s−1 toward incoming wave groups generated by the wave maker. The largest TKE dissipation rates and significant whitecaps were found at or near the center of wave groups where steepening waves approached the geometric limit of waves. The TKE dissipation rate was O(10−2) W kg−1 during wave breaking, which is two to three orders of magnitude larger than before and after wave breaking. The enhanced TKE dissipation rate was limited to a layer of half the wave height in depth. Observations indicate that the impact of wave breaking was not significant at depths deeper than one wave height from the surface. The TKE dissipation rate of breaking waves within wave groups can be parameterized by local wave phase speed with a proportionality breaking strength coefficient dependent on local steepness. The characterization of energy dissipation in wave groups from local wave properties will enable a better determination of near-surface TKE dissipation of breaking waves.
Abstract
Coexistence of wind sea generated locally and swell radiated from distant storms often results in double-peaked or multiple-peaked spectra. Identification and separation of the wave energies of wind sea and swell provide a more realistic description of the sea state, which is of great importance to scientific and engineering applications. This paper describes a method based on the peak frequency of a newly defined steepness function to separate the wave energies of wind sea and swell from the omnidirectional wave spectra. This steepness method does not rely on the availability of the information of wind velocities and wave directions and can be easily implemented for operational uses. Verification results using directional wave data collected from buoys in the Gulf of Mexico and offshore California are presented.
Abstract
Coexistence of wind sea generated locally and swell radiated from distant storms often results in double-peaked or multiple-peaked spectra. Identification and separation of the wave energies of wind sea and swell provide a more realistic description of the sea state, which is of great importance to scientific and engineering applications. This paper describes a method based on the peak frequency of a newly defined steepness function to separate the wave energies of wind sea and swell from the omnidirectional wave spectra. This steepness method does not rely on the availability of the information of wind velocities and wave directions and can be easily implemented for operational uses. Verification results using directional wave data collected from buoys in the Gulf of Mexico and offshore California are presented.
Abstract
Recently, directional wave spectra have been obtained by applying the two-dimensional fast Fourier transform (2D FFT) to the three-dimensional spatial topography of ocean surface waves collected by an airborne scanning laser ranging system during a quasi-equilibrium wind wave condition. The directional distributions show that most wave energy at wavenumbers larger than the peak wavenumber is in two sidelobes at directions symmetrically located about the wind direction. Presented in this study is an analysis of the Fourier harmonics derived from decomposition of the measured bimodal directional distributions. The similarity properties of the Fourier coefficients are analyzed. A nonlinear function is proposed to represent the similarity relation. A bimodal directional distribution model in the form of Fourier series expansion consisting of the first eight Fourier harmonics is developed. Application of this model to extend the directional distribution of buoy measurements is demonstrated.
Abstract
Recently, directional wave spectra have been obtained by applying the two-dimensional fast Fourier transform (2D FFT) to the three-dimensional spatial topography of ocean surface waves collected by an airborne scanning laser ranging system during a quasi-equilibrium wind wave condition. The directional distributions show that most wave energy at wavenumbers larger than the peak wavenumber is in two sidelobes at directions symmetrically located about the wind direction. Presented in this study is an analysis of the Fourier harmonics derived from decomposition of the measured bimodal directional distributions. The similarity properties of the Fourier coefficients are analyzed. A nonlinear function is proposed to represent the similarity relation. A bimodal directional distribution model in the form of Fourier series expansion consisting of the first eight Fourier harmonics is developed. Application of this model to extend the directional distribution of buoy measurements is demonstrated.
Abstract
To study the dispersion relation of short wind waves, a linear wave gauge array (WGA) is configured and mounted on a wave-following buoy to conduct in situ space–time measurements of short gravity waves. Results from two field deployments of the WGA buoy in growing seas are presented. The two-dimensional (2D) wavenumber–frequency spectra derived from the space–time measurements provide a direct examination on the relation of wave frequency and wavenumber of short waves in the along-wind direction. Both wavenumber-based and frequency-based phase velocities are extracted from the 2D spectra. The effect of higher harmonics resulting from the Fourier decomposition of nonlinear wave profiles is more prominent to the frequency-based phase velocity than the wavenumber-based phase velocity. The wavenumber-based phase velocity is consistent with that according to the linear dispersion relation, while the frequency-based phase velocity becomes larger due to the higher harmonics.
Abstract
To study the dispersion relation of short wind waves, a linear wave gauge array (WGA) is configured and mounted on a wave-following buoy to conduct in situ space–time measurements of short gravity waves. Results from two field deployments of the WGA buoy in growing seas are presented. The two-dimensional (2D) wavenumber–frequency spectra derived from the space–time measurements provide a direct examination on the relation of wave frequency and wavenumber of short waves in the along-wind direction. Both wavenumber-based and frequency-based phase velocities are extracted from the 2D spectra. The effect of higher harmonics resulting from the Fourier decomposition of nonlinear wave profiles is more prominent to the frequency-based phase velocity than the wavenumber-based phase velocity. The wavenumber-based phase velocity is consistent with that according to the linear dispersion relation, while the frequency-based phase velocity becomes larger due to the higher harmonics.
Abstract
A semi-implicit, two-dimensional (in a vertical plane) model is developed for circulation in the partially mixed estuary. Comparisons between the semi-implicit and explicit method are made in the simulation of tidal, wind-driven and density-driven circulations. In general, the two model results are in good agreement in velocity and density computation; the semi-implicit method, however, fails to simulate the surface seiche oscillation. On the other hand, the semi-implicit method is more efficient; depending on the horizontal space resolution, the semi-implicit method can result in orders of magnitude saving in computer time. Application of the semi-implicit model to the Potomac River indicates large longitudinal and vertical changes in tidal, density-driven and wind-driven circulations, which suggests that two-dimensional (in a vertical plane) modeling is essential in the transport and mixing study.
Abstract
A semi-implicit, two-dimensional (in a vertical plane) model is developed for circulation in the partially mixed estuary. Comparisons between the semi-implicit and explicit method are made in the simulation of tidal, wind-driven and density-driven circulations. In general, the two model results are in good agreement in velocity and density computation; the semi-implicit method, however, fails to simulate the surface seiche oscillation. On the other hand, the semi-implicit method is more efficient; depending on the horizontal space resolution, the semi-implicit method can result in orders of magnitude saving in computer time. Application of the semi-implicit model to the Potomac River indicates large longitudinal and vertical changes in tidal, density-driven and wind-driven circulations, which suggests that two-dimensional (in a vertical plane) modeling is essential in the transport and mixing study.
Abstract
Recent results of numerical wave models have shown that the presence of a bimodal directional spreading is a robust feature at wavenumbers above the spectral peak. This directional bimodality is controlled mainly by directional transfer of energy through nonlinear wave–wave interactions. The bimodal feature has also been observed in the directional spectra derived from the spatial topography of ocean surface waves acquired by stereo-photography, image radars, and an airborne scanning lidar system. In this study, a comprehensive data analysis of the evolution of the wave directional distribution during two active wave growth periods in Lake Michigan is conducted. The wind and wave measurements are acquired by two heave–pitch–roll buoys moored at a nearshore and an offshore station. An empirical method averaging the results of the maximum likelihood method and maximum entropy method is used to estimate the directional distribution from buoy measurements. The study shows that the bimodal distribution is a distinctive and persistent feature over a broad frequency range throughout the wave growth process. The characteristics of directional bimodality are quantified by parameters related to the separation angles and the amplitudes of the sidelobes. In general, the values of the parameters are smallest near the peak frequency and increase toward both lower and higher frequencies. This frequency-dependent pattern appears to be invariant to the change of wave age throughout the wave growth process. The persistent nature of the directional bimodality indicates that the nonlinear wave–wave interaction mechanism not only actively moves wave energy away from the peak frequency into both higher and lower frequency components but also constantly redistributes wave energy into directions oblique to the wind direction. At the offshore buoy site when the wind and peak wave directions align closely, the bimodal distribution is symmetric about the wind direction. At the nearshore buoy site when the local wind and the peak wave are not moving in the same direction or the wind field is less homogeneous, the bimodal distribution is asymmetric.
Abstract
Recent results of numerical wave models have shown that the presence of a bimodal directional spreading is a robust feature at wavenumbers above the spectral peak. This directional bimodality is controlled mainly by directional transfer of energy through nonlinear wave–wave interactions. The bimodal feature has also been observed in the directional spectra derived from the spatial topography of ocean surface waves acquired by stereo-photography, image radars, and an airborne scanning lidar system. In this study, a comprehensive data analysis of the evolution of the wave directional distribution during two active wave growth periods in Lake Michigan is conducted. The wind and wave measurements are acquired by two heave–pitch–roll buoys moored at a nearshore and an offshore station. An empirical method averaging the results of the maximum likelihood method and maximum entropy method is used to estimate the directional distribution from buoy measurements. The study shows that the bimodal distribution is a distinctive and persistent feature over a broad frequency range throughout the wave growth process. The characteristics of directional bimodality are quantified by parameters related to the separation angles and the amplitudes of the sidelobes. In general, the values of the parameters are smallest near the peak frequency and increase toward both lower and higher frequencies. This frequency-dependent pattern appears to be invariant to the change of wave age throughout the wave growth process. The persistent nature of the directional bimodality indicates that the nonlinear wave–wave interaction mechanism not only actively moves wave energy away from the peak frequency into both higher and lower frequency components but also constantly redistributes wave energy into directions oblique to the wind direction. At the offshore buoy site when the wind and peak wave directions align closely, the bimodal distribution is symmetric about the wind direction. At the nearshore buoy site when the local wind and the peak wave are not moving in the same direction or the wind field is less homogeneous, the bimodal distribution is asymmetric.
Abstract
Field observations show that the crosswind component constitutes a significant portion of the ocean surface mean square slope. The average ratio between the crosswind and upwind mean square slope components is 0.88 in slick-covered ocean surfaces. This large crosswind slope component cannot be explained satisfactorily based on our present models of a unimodal directional distribution function of ocean waves. Two-dimensional spectral analysis of the 3D ocean surface topography reveals that a bimodal directional distribution is a common feature for wave components shorter than the peak wavelength. The calculated result of the upwind and crosswind mean square slope components using a bimodal directional distributions yields substantial improvement in agreement with field measurements. Also discussed in this paper is the transition of the spectral function from an equilibrium form to a saturation form. Through comparison with the mean square slope data of the slick cases under which short waves are suppressed and calculation of the range of wavenumbers influenced by nonlinear wave–wave interaction, it is found that the transition from the equilibrium range to saturation range occurs at a wavenumber in the neighborhood of 6.5 times the peak wavenumber.
Abstract
Field observations show that the crosswind component constitutes a significant portion of the ocean surface mean square slope. The average ratio between the crosswind and upwind mean square slope components is 0.88 in slick-covered ocean surfaces. This large crosswind slope component cannot be explained satisfactorily based on our present models of a unimodal directional distribution function of ocean waves. Two-dimensional spectral analysis of the 3D ocean surface topography reveals that a bimodal directional distribution is a common feature for wave components shorter than the peak wavelength. The calculated result of the upwind and crosswind mean square slope components using a bimodal directional distributions yields substantial improvement in agreement with field measurements. Also discussed in this paper is the transition of the spectral function from an equilibrium form to a saturation form. Through comparison with the mean square slope data of the slick cases under which short waves are suppressed and calculation of the range of wavenumbers influenced by nonlinear wave–wave interaction, it is found that the transition from the equilibrium range to saturation range occurs at a wavenumber in the neighborhood of 6.5 times the peak wavenumber.
Abstract
The issue of duration-limited growth of wind-generated waves is of importance to wave studies. Most analytical solutions for wind waves are given in time rather than fetch domain. Numerical modeling of wave development is also often conducted in temporal evolution mode. Experimental data of duration-limited growth, however, are rare and do not cover a wide range of the wave development stage. As a result, theorists and modelers have to rely on fetch-limited evolution data, converting them into duration-limited conditions on the basis of some assumptions. During one of the field experiments of wind wave measurements, a dataset was obtained that is almost ideal for duration-limited wave growth analysis. The dataset extends the range of coverage in dimensionless time of the existing database by about one order of magnitude. The results of analysis provide strong support for the relation of space–time conversion for rendering the fetch-limited growth functions to duration-limited growth functions. Quantitative discussions on the development rate of fetch and duration growth are presented.
Abstract
The issue of duration-limited growth of wind-generated waves is of importance to wave studies. Most analytical solutions for wind waves are given in time rather than fetch domain. Numerical modeling of wave development is also often conducted in temporal evolution mode. Experimental data of duration-limited growth, however, are rare and do not cover a wide range of the wave development stage. As a result, theorists and modelers have to rely on fetch-limited evolution data, converting them into duration-limited conditions on the basis of some assumptions. During one of the field experiments of wind wave measurements, a dataset was obtained that is almost ideal for duration-limited wave growth analysis. The dataset extends the range of coverage in dimensionless time of the existing database by about one order of magnitude. The results of analysis provide strong support for the relation of space–time conversion for rendering the fetch-limited growth functions to duration-limited growth functions. Quantitative discussions on the development rate of fetch and duration growth are presented.
Abstract
A methodology for quantitative, directional validation of a long-term wave model hindcast is described and applied. Buoy observations are used as ground truth and the method does not require the application of a parametric model or data-adaptive method to the observations. Four frequency ranges, relative to the peak frequency, are considered. The validation of the hindcast does not suggest any systematic bias in predictions of directional spreading at or above the spectral peak. Idealized simulations are presented to aid in the interpretation of results.
Abstract
A methodology for quantitative, directional validation of a long-term wave model hindcast is described and applied. Buoy observations are used as ground truth and the method does not require the application of a parametric model or data-adaptive method to the observations. Four frequency ranges, relative to the peak frequency, are considered. The validation of the hindcast does not suggest any systematic bias in predictions of directional spreading at or above the spectral peak. Idealized simulations are presented to aid in the interpretation of results.