Journal of Waterway, Port, Coastal, and Ocean Engineering

In this study, a three-dimensional numerical model was developed based on a modal analysis and substructure method. The devised model was employed for an evaluation of the dynamic response of a bottom-mounted platform with a pile-soil foundation system that was subjected to random waves and currents. A two-dimensional numerical scheme with the Newmark $\beta$ method was extended to three dimensions. The equation of motion was solved by using cylindrical beam elements to analyze the response of the superstructure, and the impedance function method was employed to model the pile-soil foundation system. The displacement and bending stress at selected nodal points of the structure were computed using various input parameters, such as the shear-wave velocity of the soil and the mean wave height and period. The effect of the current on the structural response was also evaluated. Using the reliability index obtained from the Monte Carlo simulation method, the reliability of the dynamic response at critical structural members, which contained uncertainties due to dynamic forces and structural properties, was assessed.

Landslide-generated tsunami predictions are commonly based on two-dimensional (2D) wave channel or three-dimensional (3D) wave basin experiments with considerably different outcomes. It is not fully understood which idealized water body geometry applies best to a specific prototype. Hence, a physical small-scale model study has been conducted that, for the first time, systematically investigates the effect of geometry on landslide-generated tsunami height, amplitude, period, and celerity. A rigid slide generated tsunamis propagating in various geometries characterized by the basin side angle $\theta$. Considered were 2D ($\theta =0°$), 3D ($\theta =90°$), and six intermediate geometries. The differences between 2D and 3D wave heights were found to be about 20% at a distance of five times the water depth from the slide impact zone, but increased with increasing distance. It is shown that the 3D case applies on a much wider prototype range than the 2D case because it approximates the wave features on the slide axis for all investigated geometries with $\theta >30°$. The energy flux conservation based on the given 2D results can predict wave heights for the remaining geometries with $\theta \le 30°$. The implications of the present results in practice are discussed and an example illustrates how the results support tsunami hazard assessment despite significant scale effects.

Water oscillations observed in a 10-m-diameter, 2-m-deep hole excavated on the foreshore just above the low-tide line on an ocean beach were consistent with theory. When swashes first filled the initially circular hole on the rising tide, the dominant mode observed in the cross-shore velocity was consistent with a zero-order Bessel function solution (sloshing back and forth). As the tide rose and swash transported sediment, the hole’s diameter decreased, the water depth inside the hole remained approximately constant, and the frequency of the sloshing mode increased according to theory. About 1 h after the swashes first reached the hole it had evolved from a closed circle to a semicircle, open to the ocean. When the hole was nearly semicircular, the observed cross-shore velocity had two spectral peaks, one associated with the sloshing of a closed circle, the other associated with a quarter-wavelength mode in an open semicircle, both consistent with theory. As the hole evolved further toward a fully semicircular shape, the circular sloshing mode decreased, while the quarter-wavelength mode became dominant.

Hydraulic models differing from prototype scale are affected by model effects. Despite these effects, which are often negligible, they may become important in very small scales involving, for instance, small water depths. A scale family is a convenient method to detect these effects in a new test setup to demonstrate the equipment applicability. The scale family investigated here involves a previously installed test setup concerning the impulse wave run-up and overland flow. Solitary waves as a typical wave type of slide-generated impulse waves were analyzed using a pneumatic piston-type wave generator. Free surface profiles were measured to investigate the maximum flow depths and the flow front velocities. The observed scale effects led to the definition of the minimum Reynolds and Weber numbers for the considered test setup, above which scale effects can be neglected.

A two-dimensional hydrodynamic model of the macrotidal Gironde estuary (France) is developed and shown to predict the tide propagation with an accuracy of $<10\mathrm{cm}$ on water level. The 170-km-long computational domain represents, in detail, the central part of the estuary, characterized by a complex multichannel geometry. This model provides an efficient and accurate operational tool by using an optimized finite-element numerical scheme. Particular attention is paid to the offshore boundary conditions and to the calibration procedure. On the offshore boundary, the tidal signal is decomposed into harmonics whose main amplitudes and phases are obtained from a regional tidal model covering the North East Atlantic. Sensitivity studies are conducted to determine the influence of the harmonics number and harmonic constants on tidal prediction at the mouth. Differences up to $20\sim 30\mathrm{cm}$ in water level at the estuary mouth are observed between the scenarios for the tidal forcing. The calibration of the friction coefficient is thus linked to the tidal forcing scenario. A calibration procedure for the friction coefficient is also presented that can be applied to estuarine conditions. The van Rijn method is applied to predict a first set of values for the friction coefficients as a function of grain size and flow parameters. These predicted mean (time-averaged) coefficients are then adjusted with a maximum of 15% to reproduce in a best way water level and velocity measurements.

Sediment transport in wave-dominated environments is of great interest for dredged material placement, contaminated sediments, habitat protection, and other issues. The shear stress at the sediment-water interface during a wave event is an important parameter in determining erosion and transport for both experimental and model simulation applications. Sandia National Laboratories has developed a laboratory and field device called the sediment erosion actuated by wave oscillations and linear flow (SEAWOLF) flume in which high-resolution, particle-image velocimetry (PIV) has been applied to investigate turbulent flow shear stresses for a variety of flow conditions. The results of the PIV analysis for a wave cycle demonstrate a fully developed turbulent flow, relaminarization, and an explosive transition back to turbulence. In many cases, the results of the flume tests did not show good agreement with previously reported computational fluid dynamic results and existing theories, such as Blasius, for wave-current interactions, which raises the question of whether similar phenomena are present in real environments. These results implore more studies to be conducted with similarly high-resolution field measurements and modeling efforts to determine shear stress time history for oscillatory flows and the subsequent effects on erosion and sediment transport in wave-dominated environments.