Journal of Waterway, Port, Coastal, and Ocean Engineering

A time-dependent model for obliquely incident nonlinear waves is developed and applied to predict longshore current and sediment transport. The wave model is based on the Boussinesq equations for breaking and nonbreaking waves. Wave breaking is introduced by adopting the surface roller concept. Longshore current velocity is calculated using the time-averaged alongshore momentum equation, including the effects of the cross-shore circulation on the dispersion of momentum. The wave module provides the longshore current and the sediment transport modules with all required hydrodynamic information such as radiation stress, bottom velocity, undertow velocity, and eddy viscosity coefficient. The Dibajnia and Watanabe formula is adopted to predict sheet-flow transport; whereas, for the suspended load, an energetics approach is used. Model results are compared with experimental data as well as with the Kamphuis and Costal Engineering Research Center formulas for the total alongshore sediment transport rate.

Landslide generated impulse waves were investigated in a two-dimensional physical laboratory model based on the generalized Froude similarity. The recorded wave profiles were extremely unsteady and nonlinear. Four wave types were determined: weakly nonlinear oscillatory wave, non-linear transition wave, solitary-like wave and dissipative transient bore. Most of the generated impulse waves were located in the intermediate water depth wave regime. Nevertheless the propagation velocity of the leading wave crest closely followed the theoretical approximations for a solitary wave. Between 4 and 50% of the kinetic slide impact energy propagated outward in the impulse wave train. The applicability ranges of the classical nonlinear wave theories to landslide generated impulse waves were determined. The main wave characteristics were related to the landslide parameters driving the entire wave generation process. The slide Froude number was identified as the dominant parameter. The physical model results were compared to the giant rockslide generated impulse wave which struck the shores of the Lituya Bay, Alaska, in 1958.

A methodology is presented for predicting the impacts of nearshore dredging on shoreline change, using numerical models for waves, currents, and shoreline change. The dredge pit, outside of the surf zone, was assumed to be steady state, but alters waves reaching the surf zone. The methodology accounts for the transformation of directional wave energy spectra due to spatial gradients in both mean currents and bathymetry, wind inputs, and energy dissipation via several mechanisms. The wave transformation model SWAN (Simulating Waves Nearshore) was used to simulate waves at Folly Island, S.C., which served as the test case. Wave model results were validated using short-term field measurements at three locations. Longer-term (1 year) simulations were driven using hindcast wave and wind data. Velocity fields and tidal stages were simulated using the hydrodynamic model Environmental Fluid Dynamics Code in two-dimensional mode, after calibrating with field measurements. Wave model results were used to estimate longshore sediment transport and shoreline change due to longshore gradients of the longshore sediment transport rate. Empirical sediment transport coefficients were chosen to match model results to measured beach volume changes derived from beach profile data. The calibrated model was then used to predict impacts of dredging on long-term shoreline change. Results reproduced observed trends of erosion and accretion along approximately 90% of the shoreline.

Recent advances in the Boussinesq modeling of nearshore hydrodynamics offer a platform for the study of wind effects on wave transformation and breaking-generated nearshore circulation. The paper documents: (1) the new parameterization of the momentum flux transferred from the wind to surface gravity waves in the coastal region on the basis of the field observations; (2) the implementation of the parameterized wind stress into phase-resolving Boussinesq wave models; (3) the development of empirical breaking criteria with the wind effect based on the existing laboratory data; and (4) the tests of the extended Boussinesq model against field observations and empirical results with respect to wind drag coefficients over shoaling waves and wave growth on a shallow lake. Fairly good agreement between the model results and measurements is observed. The methodology for the parameterization of the wind stress as a function of wave steepness and wind speed as well as the extended Boussinesq model incorporating the wind forcing can be used as a tool to improve our understanding of wind effects on nearshore wave propagation and horizontal circulation.

Six one-dimensional-vertical wave bottom boundary layer models are analyzed based on different methods for estimating the turbulent eddy viscosity: Laminar, linear, parabolic, $k$—one equation turbulence closure, $k-\epsilon$—two equation turbulence closure, and $k-\omega$—two equation turbulence closure. Resultant velocity profiles, bed shear stresses, and turbulent kinetic energy are compared to laboratory data of oscillatory flow over smooth and rough beds. Bed shear stress estimates for the smooth bed case were most closely predicted by the $k-\omega$ model. Normalized errors between model predictions and measurements of velocity profiles over the entire computational domain collected at 15° intervals for one-half a wave cycle show that overall the linear model was most accurate. The least accurate were the laminar and $k-\epsilon$ models. Normalized errors between model predictions and turbulence kinetic energy profiles showed that the $k-\omega$ model was most accurate. Based on these findings, when the smallest overall velocity profile prediction error is required, the processing requirements and error analysis suggest that the linear eddy viscosity model is adequate. However, if accurate estimates of bed shear stress and TKE are required then, of the models tested, the $k-\omega$ model should be used.