Coastal Engineering 1990

New objective criteria are established for rejecting unfitting distribution functions to a sample of extreme wave data. Another criterion is also introduced to select the best fitting function among the eligible ones. Application of these criteria to storm wave data around Japan indicates a possibility of identifying the parent distributions for the regional population of storm wave data.

A model is presented for the generation of low-frequency waves propagating normal to the beach. The input parameters are the bottom profile and the variance spectrum of the incoming wind waves. The model predicts the variance spectrum of the low-frequency waves. Also correlations between e.g. wind wave envelope and low-frequency surface elevation can be calculated. Results of the model are compared to field data.

Predicted runup elevations for given waves and nearshore profile are confirmed as accurate by large tank and field studies with over 400 published measurements, the majority exceeding 1 m above static water level. Predictions are provided by a public-domain computer code incorporating detailed empirical guidance for smooth slopes developed by Stoa (1978). This model examines the geometrical match with specified situations, applies the composite-slope method of Saville (1958) where necessary, treats barrier texture using standard runup-reduction coefficients, and executes suitable interpolation and iteration for a fully consistent runup estimate. With irregular wave action, basic empirical guidance for uniform waves gives the mean runup elevation from the mean wave description. There is definite agreement between predictions and measurements for smooth or rough barriers with uniform waves, for controlled irregular waves, and for field situations.

The predictions of linear and nonlinear (Boussinesq) shoaling wave models for nonbreaking unidirectional surface gravity waves are compared to field observations, with particular emphasis on quantities that may be important for cross-shore sediment transport. The extensive data sets were obtained on two natural beaches, span water depths between 1-10 m, and include incident wave power spectra with narrow, broad, and bimodal shapes. Significant wave heights varied between approximately 30 and 100 cm and peak periods between approximately 8 and 18 seconds. Only the nonlinear theory predicts the increasingly asymmetric sea-surface elevations and horizontal velocities (pitched-forward wave shape) and the weaker variation of skewness (difference between crest and trough profiles) which are observed to occur during shoaling. The nonlinear theory also models qualitatively well the large skewed accelerations which occur during the passage of asymmetric waves.

The surfzone associated with a wave breaking on a plane slope may be subdivided into three regions: the transition zone, the inner zone and the swash zone. The transition zone is the region just shoreward of the point of wave breaking and is characterised by rapid wave decay and also by constant wave setdown (and thus constant radiation stress). In this paper, an empirical expression for the width of this zone is developed from monochromatic wave data. Two techniques are proposed for the consideration of this phenomenon in the numerical modelling of surfzone hydrodynamics for random waves. The implications of a zone of nearly constant radiation stress inside the breakpoint are examined with respect to wave-induced current generation - both cross-shore and longshore - and sediment transport prediction. Comparisons are made to both field and laboratory data from plane and undulating profiles.

A review concerning the methods of studying and describing wave groups is presented in this paper. After analysing 73 field records collected in Shijiu Port, China, the measured parameters of wave groups and some factors describing wave groupiness and their variations are given. Moreover, these results are compared with that of theory.

To understand the actual condition of the abnormal rise of sea water on the beach in a storm, the field observation has been carried out at Hazaki Oceanographical Research Facility for about one year under the comprehensive conditions. The relation between the sea level rising and the external forces such as waves, deviations of atmospheric pressure and the winds are examined. The contribution of the wave set-up on the sea level rising is large, while those of the atmospheric pressure and the wind are small. The observed wave set-up near the shoreline agrees well with the predicted value by the Goda's theory. There are other factors which slightly affect the sea level changing, that is to say, a time lag behind the storm, the ocean currents, and the wave steepness.

This paper describes the development of a model for the shoaling and refraction of an incident directional spectrum over beach topography. The present model is limited to topography which varies only in the on- offshore direction, but no restriction is made on allowed angles of incidence with respect to the shore-normal direction. The model is verified in comparison to laboratory data for Mach reflection of cnoidal waves from a vertical plane wall. It is shown that the model provides a more accurate representation of the evolving wave field than does an earlier parabolic approximation, using the same laboratory data.

A model is presented to describe accurately the energy transfer under breaking waves. In the model, the energy of organized large vortexes as well as those of wave-induced motion and turbulence is taken into account. The model allows to estimate the dissipation rate and distribution of energy, and then the cross-shore two-dimensional distribution of an undertow. The applicability of the model is confirmed by laboratory experiments.

The Euler equation of fluid motion is integrated in the vertical direction by assuming a hyperbolic cosine distribution of pressure, as a result, quasi-linear hyperbolic equations governing the wave motion above a solid bed are obtained. The equations are then applied to describing the wave transformation over a submerged plate.

An algorithm based on the method of characteristics is developed in the numerical computation. The wave reflection and transmission coefficients computed under various conditions are compared with the experimental data, and the over-all agreement is found to be fairly satisfactory.

A two layer undertow model is developed which consists of surface and inner layer. The surface layer defines breaking wave dynamics and the inner layer defines the mean flow(circulation) and turbulence fields. The interface between two layers is determined by time and depth averaging of the mean water level and wave height in the surf zone (interface model), in which Reynolds stresses are taken into consideration as well as radiation stresses. The system of equations in the inner layer is derived by time averaging the mass and momentum equations over one wave period. Time and space averaging of these equations in the surface layer defines the surface boundary conditions of the mean flow field in the inner layer. Turbulence in the inner layer is discribed by the standard $k-\epsilon$ model.

The numerical calculation method is also discussed and model calibration is performed by comparing with the experiments by Stive and Wind (1985).

A numerical study of water waves breaking against a vertical wall is described. The exact equations for free-surface potential flow are solved. Solutions for a simple overturning wave meeting a vertical wall placed at several positions give unexpectedly violent water motion. The face of the incident wave can converge towards a point on the wall causing very high pressure, acceleration and velocity. It seems likely that this particular type of flow corresponds to the most extreme conditions encountered in laboratories and on coasts despite the lack of any direct water impact. Details of the flow reveal that compressibility of the water is unlikely to be relevant for wave forces, but that scaling from laboratory to prototype should allow for wall roughness and for small waves riding on the incident wave.

This paper examines the following properties of observed wave groups: (1) the change of wave grouping from offshore to shallow water region including the surf zone; (2) the relation between wave groupiness and wave height distributions; and (3) the distribution of run lengths and time series of wave heights of extreme wave groups which contain the maximum wave height in each wave record. Main results are summarized as follows: (1) wave groups become flattened as they propagate landward; (2) no clear maximum in mean length of runs is seen around the position where the significant wave height is the maximum, in contrast to the laboratory data, while the tendency of the change of GF is the same as the experimental data; (3) when wave grouping becomes pronounced, the distribution of wave heights becomes wider; (4) extreme wave groups accompany more large waves than ordinary wave groups; (5) for the case of narrow spectra, the time series of wave heights of extreme wave groups is represented fairly well by that of the wave group formed by the side-band instability, and the distribution of three high waves is also in qualitative agreement with that of envelope soliton, and (6) for wide spectra, which may occur during the peak of storms, wave heights in front and rear of the maximum wave are half as large as the maximum wave height, implying that a maximum wave tends to appear without accompanying other large waves.

Two main items are treated in this paper. The first item is measurement of wave induced velocities in a large scale model and the second one is a verification of the 1-D numerical model, developed by Kobayashi and Wurjanto (1989).

Wave induced velocities on a smooth slope were measured in Delft Hydraulics large Delta flume. The run-up and run-down velocities were compared with theoretically derived upper bounds and formulas for these velocities were derived on the basis of these upper bounds.

Computations were performed with the numerical model of Kobayashi and the results (velocities, pressures and run-up and run-down levels) were compared with measurements, partly from the investigation mentioned above. The results for run-down velocities and run-up levels were acceptable, the results for run-up velocities were a little worse and the results for pressures and run-down levels were bad.

This paper presents two models for current-depth refraction of directional spectra due to time-varying currents and water depth, in which a piecewise ray method and a full ray method are used for numerical computations respectively. The results computed by the piecewise ray model in the cases of time-varying one-dimensional and two-dimensional currents are in reasonable agreement with those by the full ray model, which gives the almost exact solution in the case of linear current and water depth variations, except for a slight difference during the phase of strong opposing currents. The computations indicate that time-dependency of currents gives rise to a phase difference between wave height and current variations, and reduction in the amplification effect of directional spectrum due to opposing currents in comparison with steady current case.

We discuss the application of the third generation WAM wave model to the Adriatic Sea. We focus in particular on one of the extreme storms that produced also heavy flooding in Venice. We discuss the problem of a correct description of the wind fields as a crucial input information to the wave model. After hindcasting the wave conditions during the storm, we use them as input for an estimate of the wave set-up towards the shore. We show that its consideration is essential for a proper evaluation of the flood level in the town.

Field measurements of cross-shore velocity and water surface elevation from a natural barred surf zone during two storm events confirm the presence of undertow.

A model that predicts the depth variations of wave generated currents within the framework of a 2-D depth-integrated surf-zone model is described. The equations are formulated and solved for the case of a long, straight coast. The model is used to demonstrate that the effects of current refraction are weak and to determine the variation of the 3-D spiral shaped current profiles across the surf-zone. Finally, we show that the mixing coefficient required to get realistic variations of the longshore current is much larger than what the turbulence measurements can justify.

A robust algorithm is presented which is capable of embedding a deterministic sequence of waves into a randomly generated wave train without changing the stochastic properties of the random waves.

A three-dimensional, physical model study of the effects of nonlinearity and wave directionality (i.e. directional distribution of energy or the directional spread of energy about a peak wave direction) on wave transformation in the nearshore region is presented. The model consisted of a 1:30 fixed bed beach with plane-parallel contours. Fifty-four irregular wave conditions, typical of unimodal and bimodal directional spectra, were simulated. Twenty capacitance wave gages formed offshore and nearshore linear arrays to measure directional distributions and a cross-shore array to study transformation of the frequency spectra. Both linear and nonlinear processes were observed in the transformed directional spectra. The primary effect of wave directionality was reduction in the strength of the nonlinear phase coupling between modes. Directional spreading also reduced the formation of subharmonics (i.e. long period energy) in the transformed spectra.

In studying beach changes a law for non-uniform longshore sediment transport is generally needed, which concerns the non-uniform longshore currents. Therefore, experimental study on the similarity of velocity profiles in non-uniform longshore currents was carried out on a plane beach. It was concluded that, the velocity profiles are in similitude alongshore by comparing the nomial velocity at the breaker point. The boundary layer approximation can therefore be applied to the nearshore current equations to derive a single integral equation of non-uniform longshore currents. The theoretical solution of the equation is shown to compare with experimental results. The comparison between the theoretical and experimental non-uniform longshore current velocities are in satisfaction.

Classical time and frequency domain analyses give results with limited physical meaning when applied to shallow water cases. Lower heights and periods than suggested by the actual data are obtained via these analyses. It appears that slope effects make waves less non-linear for steeper than for flatter slopes. One can obtain useful trends by making use of higher-order analyses, utilising higher-order, asymmetrical waves of the covocoidal type as an example.

A wave equation is presented for predicting reflected and transmitted waves for a permeable submerged breakwater. This equation includes the mild slope equation derived by Berkhoff (1972), which is a vertically integrated refraction-diffraction equation. Therefore, the equation derived can also predict the combined effects of refraction and diffraction. Numerical calculations with a dissipation term due to breaking are performed to obtain reflection and transmission coefficients as well as wave height distributions. Through the comparisons with the experimental results, the validity of the model is confirmed.

In this report, a quasi-3D model for nearshore circulation is presented. The aim of the model is an economic simulation of surf-zone flow features.

This model relies on the efficient integration of three main modules: i) Wave propagation, ii) Depth-uniform currents, and iii) 1DV model for current-profiles calculation.

The basic equations are vertically integrated up to the trough level (z_tr), assuming at this level the existence of a rigid-lid, which permits to replace the free surface elevation by an equivalent pressure.

This device allows a continuous description of the flux below z_tr, making possible the obtention of the flux components in the same vertical solution domain.

The model is still under development. The results obtained, although insufficient to validate the code, serve to explore its capabilities.

The wave climate along the northeastern tropical coastline of Australia is controlled by The Great Barrier Reef (GBR). However, the processes by which the GBR attenuates and transforms waves are little understood. As the first part of an on-going study of the interaction between waves and coral reefs, a field experiment was conducted to study the processes that occur as waves break on an offshore reef and proceed across the reef flat into the lagoon. Eighteen wave, water level, and current measuring instruments were deployed and data for a wide range of tide and wave conditions were collected. Preliminary results for wave attenuation are presented. Results for wave attenuation across the reef show that wave heights on the reef flat and in its lagoon are controlled by the depth of water over the windward reef flat. As the waves travel across the reef flat, the ratio of significant wave height to water depth reduces to a value of 0.40, and the ratio of maximum wave height to water depth reduces to a value of 0.6 to 0.8. In the deeper water in the middle of the reef lagoon both the ratios of significant wave height to the depth over the reef flat and maximum wave height to the depth over the reef flat remain in the above ranges. However, at the mid-lagoon position these results are less general as wave heights inside a lagoon are also dependant on wind speed, direction, and fetch length inside the lagoon.

The purpose of this paper is to analyze the behaviour of the lognormal distribution function in wave studies, when visual observations are used to predict the long term wave height distribution.

We are focused our attention on the problem of visual biased samples. For this kind of data the lognormal model does not fit accurately the sample leading to wrong extrapolations.

One possible solution to avoid this problem is to consider the sample as truncated. According to this procedure the distribution function is estimated based only on data higher than the point of truncation.

Estimates of wave conditions within a storm event must recognize the variability of wave conditions during the storm and not just the peak conditions. A methodology is presented for the estimation of the probability distribution of the maximum wave in a storm event and the wave count in the same storm event. The predictions are compared with field observations of wave conditions in tropical cyclones from Australia's North West Shelf.

The characteristics of irregular wave run-up on coastal structures are discussed related with spectrum shape and grouping parameters of incident waves using the results of 200 cases of hydraulic model experiments. It is clarified that wave run-up is affected by the effects of the irregularity of incident waves and the slope configurations of structures.

A statistical model to estimate the run-up height from the incident wave properties is proposed considering 90 % confidence intervals. Also, the simple probability distribution model of irregular wave run-up height is developed and proved to be useful to the design of coastal structures.

In the present study a third-generation numerical wave model is used to study effects of a straight Gulf Stream and a Gulf Stream ring, on ocean waves along the coast in swell and storm conditions. The model accounts for all relevant processes of propagation, generation and dissipation of the waves (including current effects) without imposing a priori restraints on the spectral development of the waves. The dominating mechanism affecting the waves appears to be current induced refraction even though the short-crestedness of the incoming waves tends to mask its effects. Depending on wind and wave conditions, refraction may trap locally generated waves in the straight Gulf Stream or it may reflect wave energy back to the open ocean. In the Gulf Stream ring, refraction induces a considerable variation in significant wave height and short-crestedness but it hardly affects the mean wave direction. In storm conditions the processes of generation and dissipation are considerably enhanced in counter-current situations and reduced in following-current situations.

Nonlinear wave theories was improved to describe the asymmetry in time histories of the wave elevation and associated particle velocities under relatively steep nonbreaking waves. The improvement was based on the idea that the asymmetry causes both the amplitude changes of the primary and harmonic Fourier components, and the relative phase shifts between the primary and harmonic components.

Relationships between the asymmetry and local wave parameters were prerequisite to the improved calculation, and were derived by using experimental data. The improved calculation proposed was made for a specified local wave condition, without aids of direct numerical simulation techniques. Good agreements of the improved calculation with measurements confirmed applicability of the calculation to the practical use.

The main features of a model proposed recently by the authors for the propagation of long waves at grazing angle to a rubble-mound jetty are reviewed. Model results for a single jetty and for a jettied channel are shown.

This paper presents a summary of carefully conducted laboratory experiments on the attenuation of simulated spectral waves propagating over an 18-m-long bottom section covered by 10-cm-thick layers of 0.12- and 0.2-mm-diameter uniform quartz sands. The measured attenuation is used in conjunction with the theory for spectral wave attenuation developed by Madsen et al. (1988) to determine values of the movable bed friction factor for spectral waves. When the values of movable bed friction factors, f_wr, are plotted against a fluid-sediment interaction parameter, the ratio of skin friction Shields parameter for an equivalent monochromatic wave, ${\psi }_{mr}\prime$ , and the critical value of Shields parameter for initiation of sediment motion, ${\psi }_c$ , a well-defined relationship between f_wr and ${\psi }_{mr}\prime /{\psi }_c$ emerges. This relationship is independent of spectral shape (both finite-depth Jonswap and Neumann spectra were simulated) and is the same for the two sediment types tested. Although limited by the range of experimental conditions achieved in the laboratory study the results suggest a simple methodology for the evaluation of spectral wave attenuation from knowledge of sediment and wave characteristics.

Model tests have been carried out in a wave channel with and without consideration of 2nd order wave components in the control signal for the wave paddle. The wave channel was equipped with a 1:30 sloped bottom starting 57.00 m in front of the paddle. The water depth was 1.00 m during all tests.

The Self-Correcting System was proved to be able to suppress unwanted free higher and lower harmonics during the generation process.

On the slope measured amplitudes of the 2nd order waves were lower than predicted by 2nd order theory. Commonly used wave height parameters are not influenced significantly by the type of control signal.

The paper describes the principle and the computational procedure for the calculation of coastal wind-driven circulation by means of the response function technique. Response functions for a specific coastal seas are derived by use of a three-dimensional model of the area. This method computes the dynamic features associated with the coastal circulation via numerical convolution. Consequently, it is several orders of magnitude more efficient than using the direct integration of the three-dimensional coastal model itself.

Velocity measurements have been made with a laser doppler anemometer in a steady flow flume with an oscillating bed. The velocity profiles in these tests, for which the oscillation was at right angles to the steady current, were found to be qualitatively similar to those observed in previous measurements with currents collinear with the direction of oscillation. Measured time-mean velocity profiles show moderate agreement with the predictions of the models of Grant & Madsen (1979) and Christoffersen & Jonsson (1985) in the outer layer (the 'current boundary layer') but poor agreement close to the bed (the 'wave boundary layer').

A new technique has been described for the optimal fitting of JONSWAP model spectra to measured sea states. This technique, validated by numerical simulations, can provide enhanced estimations of the peak frequency, the peak enhancement factor and the significant wave height of natural seas.

For the laboratory study of many random-wave induced phenomena, wave-board control signals according to linear wave theory are not sufficient: apart from bound sub-and superharmonic waves, spurious free waves are generated. Wave generation according to a higher-order wave theory is necessary in order to suppress these spurious waves. The perturbation method of multiple scales is used to derive formulas for the second-order wave-board control signal, assuming the carrier-wave spectrum is narrow banded. The resulting algorithm is much faster than the one based on the full second-order theory. Furthermore, the applicability of the narrow-band approximation is indicated for carrier-wave spectra frequently used in coastal and ocean engineering.

This paper describes a series of observations within a combined wave-current flume. A two dimensional progressive wave train was superimposed upon a co-flowing current profile. The direction of this current was reversible, thereby allowing the formation of both a "favourable" current velocity (one in which the current is in the same direction as the wave celerity), and an "adverse" current velocity. The combined flow field was measured using laser Doppler anemometry. The nature of the current profile was modified so as to allow an investigation of the interaction resulting from both a uniform current and a sheared current profile.

An extensive series of laboratory experiments have been carried out with regular and irregular waves travelling over a horizontal bottom. Velocity measurements of the flow were obtained at numerous vertical positions with major emphasis given to the surface zone. The flow measurements were made possible by use of a custom designed LDV having the special characteristic of being operative very near the free surface. Presented is an overview of the study and some of its majors results.

A continued increase in the popularity of surfing, coupled with the potential for adverse impacts by coastal construction projects on favored surfbreaks, has recently confronted engineers with the problem of defining and quantifying a 'good' surfbreak. A rudimentary analysis of surfing mechanics demonstrates that conditions can be examined in terms of the joint statistical climate of 1) breaker peel rate, and 2) attainable board speed as characterized by the Irribarren Number. Assuming a planar beach and using linear wave theory, two theoretical models for the joint probability density of peel rate and Irribarren Number can be used to appraise short-term surfing conditions. Results indicate that the shortcrested character of real waves plays a primary role in enhancing surfbreak. For practical calculations of longterm surfing climate at specific sites, offshore wave gage data can be transformed across a measured beach profile to predict breaker conditions in the outer surf zone. Application of such a numerical algorithm to the beach at Duck, North Carolina indicates that the waves are suitable for surfing roughly 25% of the time.

A verification study of the numerical 2D wave propagation model HISWA is presented for a complex coastal area. Model predictions are confronted with field measurements. Also the effects of specific model possibilities and limitations are examined. Results have been fair: mean errors in wave height of 12% and in wave period of 4%, at a distance of 4 km from the incoming boundary over which measured wave heights reduced 40% and periods 21%.

This paper presents the experimental results on Eulerian mass transport which is induced by waves propagating on a sloping bottom. The measurements are carried out by laser Doppler velocimeter (LDV) in wave flume. A particular installation of measuring system is required to detect the boundary layer flow extremely close to the point of 0.15mm above the sloping bottom. From the experimental results, it is clear to see that the off-shore mass transport takes place inside the region from the proximity of bottom up to wave trough. Around the elevation of wave trough, the mass transport velocity changes its direction to on-shore raidly, and it reaches the maximun value at still water level, then it decreases to zero at crest. Further, the vertical distribution and similarity profile of Eulerian mass transport inside boundary layer are also obtained and discussed in this paper.

Ocean surface gravity waves breaking on gently sloping beaches generate substantial turbulent velocity fluctuations, both from overturning at the surface bore and from shear stresses at the bottom. We have used measurements made with laboratory-style hotfilm anemometers in the surf and swash on a natural beach to determine the relevant length and velocity scales. Battjes (1975) has pointed out the importance of determining the turbulence scales in the surf zone. Modelers, such as Svendsen and Madsen (1984), for example, rely on length and velocity scale estimates to parameterize and solve the complicated equations that govern surf zone flows. We find that turbulence length scales depend essentially on the bore height, and therefore on the local depth, but may decrease sharply under the bore. We also determine that at least the horizontal velocities approach isotropy at frequencies of 2 to 3 Hz, which turn out to also correspond to length scales on the order of the local depth.

For separating incident and reflected waves, three methods based on linear long-wave theory, small-amplitude wave theory, and quasi-nonlinear long-wave theory were applied to field data recorded on a steeply sloping beach and on a gently sloping beach. The method based on the quasi-nonlinear long-wave theory gave the best results and described well the observed, incident, and reflected wave field.

Measurements made during one day of the 1986 SUPERDUCK experiment at Duck, North Carolina are used in order to investigate the model of Bowen & Holman (1989). This model explains low-frequency oscillations in the longshore current, which were observed during that experiment, in terms of a shear instability in that flow. The model is extended to include dissipation (in the form of bottom friction), and it is found that there is good agreement between it and observation.

In many design studies for coastal harbours mathematical models are used to define wave conditions in the vicinity of their entrance. There is a wide variety of models available for this purpose, each with their own characteristics. This paper compares the performance of three models which are in use for this type of study. The advantages and drawbacks of each of the models is considered by comparison with results from a physical model of a typical harbour approach bathymetry.

Observations of pressure fluctuations at the sea floor in 13 m depth are compared to an existing theory for weakly nonlinear surface gravity waves (Hasselmann, 1962). In this depth, free surface waves (obeying the dispersion relation) at sea and swell frequencies (0.05-0.3 Hz) are weakly attenuated, but free waves at frequencies higher than about 0.4 Hz do not reach the sea floor. However, nonlinear interactions between (primary) free waves of about the same frequency, propagating in nearly opposing directions, theoretically excite long wavelength, double frequency (secondary) forced waves that are only weakly attenuated at the sea floor. Bottom pressure spectra observed in 13 m depth show an (O(10²)) increase in high frequency (0.35 - 0.6 Hz) forced wave energy levels in only a few hours after a sudden large veering in wind direction. Estimates of the free wave frequency-directional spectrum show a correspondingly rapid change with undirectional seas when forced wave energy levels are low before the wind veering, and two nearly directionally opposing seas (0.15 - 0.3 Hz) when forced wave energy levels are high, consistent with the theoretical generation mechanism. The observed forced wave energy levels are in good agreement with theoretical predictions based on the directional spectrum estimates. Estimates of average wave numbers verify that sea and swell obey the dispersion relation, and show that double sea frequency forced waves have longer wavelengths than the interacting free waves, consistent with theory. Observed third order statistics confirm that the double frequency forced waves are coupled to the directionally opposing seas with a phase relationship predicted by theory.

DOLPHIN-I is a new hybrid parametric deep water wave prediction model developed by the authors. This paper examines the applicability of DOLPHIN-I, on the basis of the comparison of the computation with the data of significant waves obtained in five sea areas with different horizontal scales around Japan during monsoons and typhoons. In particular, the wave data in Osaka Bay includes frequency and directional spectra. It is verified that the model practicality is excellent, because it reproduces well the wave data observed in any concerned sea area.

A micro shear gauge was used to measured the bottom shear stress under the shallow water wave on the smooth bed, the data show that bottom shear stress have a asymmertic form similar to the surface water elevation due to the non-linearity of wave. The bottom friction factor under wave creat and wave trough were very different with that defined by Jonsson's and the results were contrary when the bottom particle velocity were defined by linear and non-linear weve theory respectively. Empirical formula were obtained for practical use.

A standard directional spectrum is determined by a finite number of parameters included in its mathematical expression. In this paper, a theory is developed to estimate the parameters from a given data set on the basis of the maximum likelihood method. The theory is applied to estimating the parameters in the Mitsuyasu-type standard directional spectrum from data obtained by a three-component array. Main results of analysis of field data are 1) the peak wave direction agrees well with the mean direction defined at each wave frequency, 2) the peakedness parameter can be approximated by that estimated from the long-crestedness parameter at each wave frequency, and 3) the peakedness parameter takes its maximum at the frequency slightly lower than the peak frequency.

The results of extensive laboratory investigations on the oscillatory flow over ripple models were presented. The ripple models were installed in an oscillatory flow flume. A two-dimensional lasser doppler velocimeter was utilized to measure the fine structure of flows in the vicinity of the whole length of a ripple model which was induced by various types of oscillatory flow. The results of measurements were displayed in the form of distributions of mean fluid velocity, stationary velocity, fluid turbulence intensity and Reynolds stress. By using these data the characteristics of oscillatory flow, such as the formation and decay of vortices, the generation of stationary velocity and the variation of kinematic eddy viscosity were discussed.

Lateral variations in wave height have been observed in directional wave basins when generating monochromatic plane waves. Two numerical schemes are presented to examine wave height sensitivity to perturbations in stroke length between paddles and gaps between either adjacent paddles or the sidewall of the basin. A non-uniform wavemaker front generates both the desired wave and can generate waves at an angle and at the same frequency as the main wave.

The aim of the work presented here is to propagate random waves from deep or intermediate water depth to the nearshore region with a Boundary Integral Equation Method (BIEM) which is able to handle the nonlinear effects that occur in the process. A two-dimensional (x,z) mild nonlinear model to propagate waves over an uneven bottom is presented here. It takes into account 2nd order nonlinear effects of the wave transformation entering into shallow water. If energy dissipation is neglected, the flow field generated by the wave propagation can be described by a velocity potential formulation which is governed by the Laplace's equation in the domain. The results obtained from the model are time-dependent. The model is tested with solitary and irregular waves and compared with analytical and experimental results.

Spatial radar signals received by a conventional marine radar are digitized as a 8-bit word by means of an A/D converter. The numerical procedure is discussed to estimate the two-dimensional radar spectrum from a digitized radar image. Directional spectra of sea waves are also observed by a 3-component wave array. A comparison of the mean wave direction and period estimated from the radar spectrum with the results from the array spectrum shows that the developed data acquisition system and numerical procedure represent a reliable and very cost-effective means to observe the properties of sea waves.

The phase variation of turbulent intensity in the surf zone during one wave period is discussed. The existing data show a peak near the wave crest phase and a following gentle decrease. The peak delays downwards. The model by Deigaard et al.(1986) which uses the one equation model for the turbulent kinematic energy transportation explains the above mentioned trend roughly. The vertical convection term has no significant effect on the phase variation. The production term given by Deigaard et al. overestimates the energy production at the water surface by wave breaking.

Forcing of the nearshore circulation system by incident wave groups is examined. Attention is focused on the generation of migrating rip currents and developement of low frequency motion. Analysis of experimental results categorises the migrating rip currents as very low frequency motion in the FIG energy band. It is suggested that migrating rip currents, in addition to shear waves, are a valid generation mechanism for very low frequency motion in the nearshore region.

Reef coasts, because the wave height of incoming stormy high waves are extremely dampened by coral reef, have been believed as calm sea areas. However, many coastal structures have actually been damaged by stormy waves dampened by the reef. Surf beat with large wave height was discovered through field observations. The surf beat has been named as "Bore-like Surf Beat". The wave height and the wave velocity of the Bore-like surf beat is larger than that of individual waves. Numerical simulation taking into account of non-linear effect of surf beat phenomenon on the reef shows good agreement with laboratory data.

The main reason of the disasters of coastal structures on the reef coasts is existence of the Bore-like surf beat resonantly excited incoming wave groups.

We consider the extension of previous theories for Bragg reflection of surface waves by parallel bars to the case of artificial bars placed discretely on the seabed. The case of non-resonant, weak reflection is considered first, followed by a consideration of the application of resonant interaction theory to the dominant Fourier mode of the bar field. Both theories are compared to numerical results, and discrepancies are seen in both cases. Finally, experimental results are compared to theory.

The conditions under which abnormal waves occur in the western part of the Japan Sea and the northwestern part of the Pacific Ocean are examined. A numerical study of the abnormal waves was performed for a selected monsoon condition in winter. The mechanism of abnormal wave development is clarified using a wave spectrum model. It was found from the directional spectra that the effect of wind duration plays a key role in abnormal wave generation. For typhoon conditions, the relation between a standard project typhoon and the abnormal waves induced, is considered and the simulation method of the typhoon model is improved by introducing the typhoon stagnation effect. Finally, it was shown that abnormal waves generated by typhoons are due to long typhoon stagnation.

Two data sets of field measurements of wave-induced nearshore currents are analyzed and discussed with respect to cross-shore transport phenonema and the effectiveness of beach groynes. The results are discussed in order to improve future artificial beach nourishments.

The paper deals with a method for the simulation of irregular, directional waves over large coastal areas using a linear model based on a energetically equalized discretization of wave spectrum and a parabolic approximation to the Berkhoff equation for wave evolution. A sensitivity analysis based on a comparison with physical model tests gives some ideas about the applicability of the method.

This study derives a practical equation that can describe the weakly nonlinear waves having an arbitrary spectral form and propagating unidirectionally from deep to shallow water, and carries out the numerical simulation using the equation. Based on the simulated results, some investigations are made on the characteristics of spatial variation of wave group and representative wave-heights in the propagation process.

This paper presents the results of analyzing six months of simultaneous records of water levels, currents, wind and atmospheric pressure that have been collected outside the surf zone at three locations in the eastern coast of Gran Canaria (Canary Islands, Spain). The results shown in this paper concern mainly one of the coastal stations, placed at the port of Las Palmas city. Although tides and currents use to be astronomically dominated, a non-permanent term appears in certain time intervals, identified as a long wave -rotary type- with a frequency very close to the inertial one. Another interesting result is the appearance of harmonic tidal periodicites in the atmospheric variables, which suggest a higher importance of the meteorological tides or/and a strong interaction sea-air.

Solitary wave forces acting on a large vertical cylinder are calculated using a time-stepping method which improves upon the conventional boundary element approaches. The effect of this improvement on the numerical solution is a more realistic solitary wave profile. Nonlinear effects on the solitary wave forces are investigated by comparing the numerical results with the first approximations given by Isaacson. The first approximation for overturning moment is underestimated remarkably when the incident wave height is relatively high.

The turbulent boundary-layer flow over flat rough beds due to a wave or a combined wave-current interaction is studied by using a simplified numerical second-order turbulence model. The model results are compared with many sets of experimental data. Excellent predictions for ensemble-averaged velocities and favourable predictions for turbulence quantities are obtained. Variations of kinematic and dynamic characteristics of boundary-layer flow with wave, current and bed roughness parameters are determined. The model is also modified to simulate the oscillatory turbulent flow over rippled beds. The mean velocity field and the distribution of time-averaged turbulence quantities are calculated. The validity of the model is verified through comparison with experimental results. The performance and the limitation of the model are discussed.

The three related phenomena of wave runup, wave setup and the dynamics of the coastal watertable are considered and their inter-relations are investigated via a comprehensive field study. The measured runup distributions confirm the expectation that the Rayleigh distribution is a reasonable model and that the vertical scale of the distributions is proportional to √(H_ormsL_o) as in Hunt's (1959) formula but that the proportionality to the beach slope, also prescribed by this formula, only applies for fairly steep beaches. The measured setup profiles are quite different from the ones predicted on the basis of H=γh and linear wave theory in the surf zone. The measured profiles are flatter in the outer surf zone and steeper close to the beach. The shoreline setup is generally about 0.4 H_orms' which is somewhat higher than the previously suggested values. This is because the previous values of the shoreline setup were rarely measured but generally extrapolated with insufficient recognition of the steepening of the profile near the shoreline. The watertable data show that the inclined beach face acts as a strongly non-linear filter which makes the watertable variation at a point inside the beach far from sinusoidal (when the tide is approximately sinusoidal) and elevates the average position of the watertable considerably compared to the mean sea level.

The Second-order Mode Coupling Equation (SMCE) is applied to the 3-dimensional evolution of a solitary wave the crest-line of which is not straight but bent initially. It is found that the bent solitary wave recovers a straight crest-line spontaneously by producing a nonuniform distribution of waveheight and hence that of propagation speed along the crest-line.

Wave run-up phenomena due to breaking waves acting on sea dykes are focussed in this paper. Field measurements at two locations of the North Sea Coast are reviewed and compared with full-scale investigations in the Large Wave Channel, Hannover. Commonly used design approaches on wave run-up at sea dykes are discussed. For some naturally shaped shallow wave spectra (Pierson-Moskowitz type) estimations are proposed for wave run-up, run-down and water front velocity in relation to the wave height and to the breaker number.

This paper presents the analytical derivation of probability density functions applicable for tidal (high frequency components), residual (low frequency components), and total (the vector sum of tidal and residual) current velocities. The theoretical probability density functions are compared with histograms constructed from measured data. A method to estimate the extreme current velocities in a specified time period is presented by applying extreme value statistics.

A fully nonlinear potential flow theory is solved numerically but almost exactiy for a soritary wave passing over a submerged obstacle by using BIM. Based on the numerical solutions, the refrection characteristics and the breaker type and criterion are made clear for a solitary wave up to breaking caused by a step.

A report is given on the NATO Advanced Research Workshop on "Water Wave Kinematics" held in Molde, Norway, 22 - 25 May 1989. The reports from the five working groups set up during the workshop are included in full.

A method to determine the dimensions of rip current systems (distance between rip currents, and width and depth of the rip channels) is presented for the case of a coast with longshore bars. The method is based on an overall sediment balance in the nearshore region. The balance is determined by application of a model for the wave-induced flow combined with a sediment transport model. The influence of the wave height, the wave period, and the direction of wave propagation, as well as of the sediment size is analyzed.

A breaking wave model, which is partly physical and partly analytical, is proposed. This model is based on observations that up to a certain moment the wave presents a long, smooth, horizontal, cylindrical edge, which then segments due to surface tension effects. A disturbance on a cylindrical surface, withdrawn from the influence of gravity, becomes unstable when its wavelength exceeds the circumference of the cylinder. The rate of growth of the instability, is a function of the radius of the cylinder and the wavelength of the disturbance. Using the theory describing the evolution of the assumed hyperbolic shape of the tip of a breaking wave, the radius of the cylindrical edge is approximated to the radius of curvature of the hyperbola. The model describes the three-dimensional evolution of the curling wave crest. Scale effects are then derived which show good agreement with experimental results.

The practical wave hindcast methods namely, SMB (Sverndrup-Munk-Bretchneider), PM (Pierson-Moskowitz) and SPM 1984 (Shore Protection Manual) were evaluated and aceesed for their accuracy using the one year recorded of the available oceanographic data at the lower Gulf of Thailand in 1987. It was found that all models show the comparable accuracies of significant wave height (H_s) and significant wave period (T_s) but SPM (1984) rather gave the lowest root mean square error when H_s exceeds 1 m. The empirical formulae of land-wind velocity relationship was simply established in a exponential form. The value of constants in the expression between the 5 meteorological stations on land in the southern part of Thailand and the offshore platform were obtained as well. Using the land-wind velocity empirical formulae and a selected wave hindcast method to predict H_s and T_s' the better result was obtained by the low RMSE value.

When applied to real engineering problems, mathematical models of combined wave-current flows usually assume that both wave and current properties remain in a quasi-steady state. Shear stresses and other flow parameters are calculated for instantaneous values of wave height and current strength, regardless of the changes being induced by tides, winds, or wave grouping.

This paper describes tests carried out in a wave flume to investigate velocity profiles, bottom shear stresses and water surface characteristics under gradually developing flow conditions, such as those created when a group of waves is superimposed onto a pre-existing current or when a steady current is added to a train of regular waves. Results indicate that, at laboratory scale, the water surface characteristics adapt within a few wave periods to changes in superimposed waves or current, but that mean velocity profiles and shear stresses may take considerably longer to respond to the new flow conditions.

The consistency of the mathematical formulation of a 3D hydrostatic current model for coastal areas is discussed. At various points, widely accepted concepts and formulations are shown to lead to inconsistencies, and more consistent alternatives are proposed. Besides, some essential lacunae in our physical knowledge of 3D wave-driven currents are indicated.

Interior flow fields under drastically transforming solitary waves were visualized and recorded using a high-speed video system. Flow velocities were estimated by tracing fluid particles through spatial correlation analysis of video images. A Lagrangian equation system was successfully employed to numerically simulate two-dimensional wave transformations in a vertical plane.

The reflection of small-amplitude surface waves by a number of vertical porous plates is investigated. The plates are fixed in a semi-infinitely long channel of constant depth. Analytic solutions in dimensionless closed from are presented for the surface wave profiles in the several regions separated by these plates. And from them the reflection coefficient is obtained. Three cases, one to three porous plates, are evaluated. The results show that for incident waves in the range of 1/20 < water depth/wave length < 1/2 , it is appropriate to maintain the spacing between every two adjoining plates, and between the end wall and its nearst plate at a value of 0.88 times the water depth. As two or three porous plates are used for damping wave, an arrangement would be more effective if the front porous plate has a larger porosity and the back one has a smaller porosity. Several experimental tests for these porous-plate wave absorbers are conducted and the reflection coefficients are measured by using Goda's method. It is found that the theoretical solutions agree fairly well with the experimental results.

A mathematical model of wave transformation over a submerged permeable breakwater is developed on the basis of the equations for waves on a porous layer which are newly derived under the mild-slope assumption. The model equation is given as a two-dimensional elliptic equation analogous to the mild-slope equation on an impermeable bed. The model of wave breaking on a submerged permeable breakwater is proposed on the bais of the modified mild-slope equation. The validity of the model is confirmed through comparison with the experiments for a trapezoidal breakwater and with strict solution for a rectangular breakwater.

To cater for the effect of wind-driven nearshore processes on the prediction of nearshore onshore/offshore sediment transport a series of field experiments were carried out to establish a reference framework to be used as basis for the modelling of wind-driven cross-shore currents. The results indicate a clear relationship between bottom return current, wind speed and the water depth. This implies that a mathematical framework can be established using the conservation of mass and momentum and the similarity criterion.

In this paper, a numerical model based on an extended form of the Boussinesq wave equations which takes into account the influence of an ambient current on waves is used to investigate irregular waves propagating and refracting on an ambient current. After presenting the equations, a comparison based upon linear (small amplitude) wave theory is conducted to illustrate the range of application. The numerical model was used to simulate irregular waves with a current in a flume and in a basin. The results were in good agreement with the solution of Longuet-Higgins/Steward.

In this paper the transformations of sea waves, reproduced by a JONSWAP spectrum in a laboratory channel, have been studied both in frequency and time domain. In frequency domain the occurrence of non-linearities producing energy transfers to high and low frequencies was studied. The reported experimental results enabled us to comparate the two approaches and to define the limits within which they give the same results.

Data from an alongshore array of bidirectional current meters located in the surf zone were analyzed to provide estimates of the behavior of three bands of wave energy during the passage of a moderate storm. The variance in the incident wave band is found to be saturated, as is expected of depth-limited waves. The infragravity band variance is found to depend most strongly on the offshore wave height. Shear wave variance is found to be correlated with the magnitude of the measured longshore current. During a period of nearly constant high waves, the composition of the low frequency band motions (f < 0.05 Hz) changed dramatically. This change is thought to be related to the directional characteristics of the incident band wind waves.

Comparison is made between a deterministic infragravity-wave model and existing laboratory experiments. The theoretical model considers incident bichromatic waves including the effects of the accompanying incident long, bound wave, an oscillating position of the break point, and the intrusion of short-wave grouping into the surf zone, whereas frictional effects are neglected. A measure of the infragravity wave activity is the amplitude of a seaward progressing free, long wave. For this amplitude the qualitative agreement between theoretical results and experiments is excellent. Quantitatively the theory overestimates the infragravity waves by typically 50-100%. This may in part be attributed to the neglect of frictional effects.

An economic model is developed to calculate the optimal height of sea defences in case of sealevel rise. The optimal amount and the optimal period of heightening are found, for the single period as well as for the multi-period case. Finally the optimal strategy of heightening of sea defences in case of an uncertain sealevel rise is formulated.

A theoretical and computer simulation has been conducted on near field tsunamis generated by three dimensional bed motion. Linear dispersive wave theory has been used. Nonlinear effect has been found to be negligible for nearfield tsunami generation problem, as nonlinear effect is significant primarily for propagation over long distances. The specified bed deformation takes into account many possible dipole dislocation patterns with different aspect ratio. Three dimensional pictures of the water surface elevation were constructed from the array of computer data.

For the nearfield tsunamis problem, the distance between the generation area and the shoreline is relatively short, thus the main generated wave quickly interact with the reflected leading wave from the coastal boundary to create an even more complicated wave profile. The coastal boundary can either focus on diffuse these waves, resulting in large variations in wave amplitude in the coastal region.

A detailed knowledge of the velocity field is often needed, e.g. for environmental impact assessment and the presence of a continuously stratification often necessitate the use of three-dimensional models. The present paper describes such a three-dimensional numerical model. Its ability to model situations involving stratification is illustrated through two classical hydraulic problems and a practical application. The problems of boundary data in the three dimensions are discussed and a possible way of overcoming these problems is devised.

A fully nonlinear model for free surface potential flows is used to analyze the transformation of solitary waves above a mild slope, from intermediate to shallow water (shoaling, overturning), and to study the interaction of these waves with coastal structures located in the shallow area. Computations include wave runup, overturning and reflection form steep slopes or vertical wall, and from a combination of a slope and a submerged breakwater. Results are compared with other numerical, analytical and experimental results. Effects of the submerged breakwater, of making horizontal velocities more uniform over depth and of reducing wave runup on the slope, are further detailed.

The influence of tide- and surge-induced currents and surface level variations on wind wave generation and propagation in the North Sea is investigated. The main interest is focussed on the shift of the absolute frequency due to time varying currents. For several North Sea storms calculations have been performed with a third-generation wave model, which includes all relevant effects of wave-current interactions and wave generation. Results of several hindcasts have been compared with data. The study shows that tides and surges result in relatively small but distinct modulations of mean wave parameters (5 – 10%), and large variations of spectral energy densities (50%). The tide-induced modulations of in particular wave periods are strongly influenced by the unsteadiness of depth and current. Observed modulations of mean wave parameters with periods comparable to that of the tide appear to be wind-induced for a significant part.

The results of two-dimensional hydraulic experiments are presented on the decomposition of regular and irregular surface waves by a fixed, submerged horizontal plate with a finite length. Nonlinear interactions, containing the near-resonant interaction as well as the self- and mutual-interactions, among the waves affected by the submerged plate are found to induce the generation of higher harmonic components which propagate independently in the lee side, and to result in appreciable disintegration of wave form. This wave decomposition may produce considerable modification of the spectrum of a transmitted-wave train through the nonlinear transfer of energy from primary wave components to higher frequency components and consequently reduce its significant wave period. With the above mechanism, thus, the submerged horizontal plate may serve as an effective interception device against waves, especially of long wave length.

One of the important results from the recent SUPERDUCK experiment was the discovery of very low frequency motions (10² – 10³ second periods) with anomalously short longshore wavelengths (10²meters). Subsequent analysis showed the motions to be shear waves, unstable perturbations of the mean longshore current that act to conserve potential vorticity. A specific prediction of the model is that these waves have negligible elevation signal (they are non-divergent), in contrast to gravity waves. Examination of simultaneous run-up data confirms the non-divergent nature of the solutions. While edge wave modes were detected, no significant swash energy was seen in the low frequency, high wavenumber region where shear wave energy was found in the current meter data. This apparently negative result adds support to the interpretation of these motions as shear waves.

This paper describes the history and present state of the art with respect to safety philosophy for flood defence systems in the Netherlands. It is explained that present day philosophy is based on a so called dike-ring approach and on a probabilistic treatment of all load and strength parameters. Both target safety levels as well as verification procedures are discussed. Simplification rules for application in every day practice are given.

A method is presented for predicting primary stability of armor units on a breakwater exposed to solitary waves. A 2-D flow model is developed based on a boundary element method for simulating run-up of a nonbreaking solitary wave on an impermeable, smooth slope. Since Laplace equation subject to exact boundary conditions is solved by the model, the 'mild slope' restriction in using a 1-D long wave model is eliminated. Schemes including inductive wave generation, Lagrangian shoreline motion and free surface regridding are proposed. The run-up flow model is then coupled with an armor stability model to predict a stability number for armor units as function of time and location on the slope. Aspects such as applicability of Morison-type approach and selection of lift coefficient are examined. Results of computed wave run-up and armor stability are compared with experimental data.

This paper examines the relationship between the off shore bathymetry, resulting wave climate, shore boundary conditions (i.e., seawalled versus dune/beach sections) and the shoreline response as represented by long-term shoreline change rates over 120 years for the southern Virginia ocean coastline in the United States. Along the tourist area of the City of Virginia Beach, the data supports the conclusion that a seawall's presence for over 50 years has produced no significant increase in the recession rate. The highest recession rate (3m/yr) occurs at the Sandbridge sector further south. Some now claim that the beach width is narrowing at Sandbridge as a result of recent seawall construction. This allegation completely ignores the root cause of the problem which is shown in this paper to be the steep offshore bathymetry and high wave energy in this region. If we neglect the offshore boundary conditions when making field studies of "hardfacing" versus dune/beach sections, we can reach completely erroneous results.

The numerical model that was shown to be in fair agreement with six test runs of available data on the stability of rock units under irregular wave attack is used to examine the critical incident wave profile associated with the minimum rock stability for each run. The minimum rock stability computed for the runs with dominant plunging waves on gentler slopes is caused by the large wave with the maximum crest elevation during its uprush on the slope. The minimum rock stability computed for the runs with dominant surging waves on steeper slopes is caused by the downrushing water with high velocities resulted from a large zero-upcrossing wave with a high crest followed by a deep trough. In addition, a simplified model is proposed to predict the eroded area due to the movement and dislodgement of rock units using the probability of armor movement computed by the numerical model. This model is shown to be in qualitative agreement with the empirical formula of Van der Meer (1988).

This paper presents a numerical model for computation of floating breakwater mooring forces. The model is based on time domain mooring analysis techniques which permit simulation of: (1) nonlinear mooring line load characteristics and (2) mooring line loads associated with second order wave drift forces. Numerical model results are compared to physical model tests (Torum, 1989) and prototype measurements (Nelson and Broderick, 1986). These comparisons demonstrate that the numerical model provides good estimates of floating breakwater mooring line forces. Accordingly, it is concluded that the numerical model serves as a useful engineering tool for analysis and design of floating breakwater mooring systems.

Experimental results from large scale tests are presented for the distribution of wave-induced pore pressure within rubble mound breakwaters, for the reflection coefficients and the damping of wave motion through the armour and under-layer. The results are then discussed with respect to application and further research.

This paper describes some of the results obtained from a series of model tests to measure wave forces and overtopping in a vertical breakwater located in a surf zone. The aim of the tests was to better design conditions using submerged structures. With the sections proposed in this study, it is possible to achieve a considerable reduction in wave forces and overtopping on vertical breakwaters.

The possibility for the construction of "sand mound breakwater", which is a caisson type breakwater on a artificial sand mound, has been studied. First the wave force characteristics acting on the composite sand mound breakwater model are described, to identify the conditions where impulsive breaking wave pressures are generated. Second, the scouring of the toes of rubble mounds is discussed. Finally there is an outline of field experiments, scouring of sand mounds, and the status of the deposits. The results of experiments provide useful information for the design of breakwater constructed on the artificial sand mound or on a similar double-sectioned sea bottom.

Low-crested structures are designed for some or even severe overtopping. The stability of these structures is sometimes higher than the non-overtopped structures, due to the fact that wave energy can pass over the crest, giving lower wave forces on the armour layer of the seaward slope.

Low-crested structures can be classified into three categories: dynamically stable reef breakwaters, statically stable low-crested structures with the crest above swl and statically stable submerged structures. Well described investigations at various institutes (in total about 275 tests) were re-analysed and this has led to practical design formulas and graphs for each of the three classes mentioned above.

A boundary integral equation method combined with a non-linear time stepping procedure for the free water surface is developed for simulations of the interaction between highly non-linear water waves and fixed submerged horizontal cylinders.

The wave forces on the cylinders are computed and a good correspondence is found with other computed results for low Keulegan-Carpenter numbers.

A new method for tracing the orbits of water particles in the fluid domain is developed, and the influence from submerged structures on the orbits is visualized through some computational examples.

The subject of this present research is to study probabilistic design of caissons of composite breakwaters, rubble-mounds and foundations protected with armor blocks. Concerning caissons, sliding, rocking and over-turning motions are considered to be the failure modes. Wave forces acting on a caisson can be calculated using Goda's formulas(1974). The stability of the rubble-mound is estimated by means of Bishop's method while geotechnical problems employ a simplified equation as a reliability function in accordance with the Japanese Standards for Coastal Structures.

Since distribution for occurrence for each failure mode is unknown, Monte Carlo simulation was applied to calculate the risk of each failure. The probability of geotechnical failure was greatest among the failure modes but the rubble-mound was not in danger of collapse.

Failures of rubble mound breakwaters armoured with complex types of unreinforced concrete armour units are often due to breakage. This happens when the stresses exceed the material strength. Sufficient parametric studies of the stresses are not yet available to produce design diagrams for structural integrity.

The paper presents the results and the analyses of model tests with 200 kg and 200 g load-cell instrumented Dolosse. Static stresses and wave generated stresses were studied as well as model and scale effects. A preliminary design diagram for Dolosse is presented as well.

The research into the physical processes in concrete block revetments under wave attack resulted in three separated design methods. A resume is given of the numerical, analytical and black-box methods and their respective fields of application have been indicated. Special attention is given to the verification of the analytical model with large scale model tests.

A series of hydraulic model tests of armour rock stability have been carried out on a 1:2 slope armoured with rock of each of 5 different shape types. Results of these tests have been compared with damage predicted by van der Meer's stability formulae. Coefficients are suggested to describe the influence of armour shape and layer thickness in revised stability formulae.

Waves reflected from breakwaters and sea walls may cause navigation or mooring problems in or near harbours, and may increase beach erosion or local scour.

This paper gives results of hydraulic model tests on the reflection characteristics of idealised rock armoured slopes. Examples of the performance of a number of practical structures are also discussed.

This paper discusses a mathematical model of the large, short-lived pressures brought about by waves breaking against coastal structures. The idea of pressure impulse, P (the integral of pressure with respect to time from the start to the finish of the impact) is used to simplify the equations of ideal incompressible fluid notion. P satisfies Laplace's equation in a domain which is the mean position of the wave during the very short time of impact. We solve analytically a two-dimensional boundary-value problem, which models an idealized wave striking a vertical wall. Expressions are derived for the impulse on the wall, the peak pressure distribution, and the change in fluid velocity due to impact. The results are insensitive to the shape of the wave far from the wall. The results agree with some experimental measurements, from the literature.

This paper discusses an analytical model for solitary wave overtopping at sea dikes. The model is based on the "weir flow analogy" of Kikkawa et al.(1968). Results of laboratory test on smooth and impermeable model dikes with angles of 45°, 60° and 90° are utilized in constructing the model. Results for solitary waves are compared with those for regular oscillatory waves for the case of vertical dike.

Identification of the Morison equation force coefficients C_m & C_d depends on the condition of the wave force/kinematic data. For simple harmonic kinematics, the data are equally conditioned for identifying C_m & C_d at the critical value of the period parameter (K = U_mT/D) defined by Keulegan and Carpenter (1958). Critical values for K are = 11.4 from geometric interpretations and = 13.16 from numerical interpretations. Stable transverse lift forces may also be shown to correlate with the critical values of K from the physical interpretation of the period parameter given by Keulegan and Carpenter (1958). Data from three different types of physical experiments are used to demonstrate the importance of the condition of the data.

New criteria for the design of filters under revetments are introduced and verification tests are described. These criteria are based on a hydrodynamic limit instead of the usual geometric limit, assuming a similarity between the flow in open channels and flow in the pores at an interface between filter and base. First a theory was developed for a granular filter only, and later this theory was extended to include the influence of a geotextile. The verification experiments are outlined, and the applicability of the derived formulae is discussed.

The authors have carried out a research project for investigation of the wave forces acting on breakwater armour units. The project included a literature study, hydraulic model testing and analysis of the test results. The hydraulic model testing was made in a wave flume on a 2-dimensional idealised breakwater structure with an armour layer consisting of two rows of horizontal pipes.

The stability of the armor layer of rubble mound breakwaters has been shown to be highly influenced by some wave grouping characteristics. The measured damage functions are dependent on the Envelope Exceedance Coefficient, α, and also on the Groupiness Factor defined as: ${\text{GF=σ[H}}^{\text{2}}{\text{(t)]/8m}}_0\text{.}$

The paper describes the monitoring program that has been set up for the hydraulic loads on the Eastern Scheldt Barrier. The strategy for the measurements will be discussed in detail. The results of the measurements, that have been performed yet, will be presented. These results are discussed only briefly, because the evaluation of the results has not been completed yet.

This paper intends to review the experimental research work performed at the University of Hannover on the structural stability of tetrapods.

In this paper, various damping action on submerged breakwater test results will be analyzed, in order to improve the methodology for future analysis. Experimental data from the authors an other researchers will be analized.

The movement of a block lying loose in a placed block revetment under wave loading was simulated numerically. The results have been compared with the results of large scale model tests in which the motion of a block and the pressure distribution in the filter layer underneath was measured. It appeared that the mechanism concerning the pore pressure distribution and the block movement are described well in the numerical program.

The block movement as calculated in the program is higher than measured. The reason for this is described in this paper.

An experimental investigation was conducted to examine the change in fluid pressure through a breakwater armour layer as a function of the incident wave characteristics, breakwater geometry, armour unit type, number of layers of armour and core permeability. Values of internal pressure were found to decrease with the number of layers of armour. The largest percentage decrease occurred within the first few layers; although this trend was somewhat influenced by armour type.

A comprehensive examination of the structural response of breakwater armor units to static and wave-induced loads has been completed using a dolos armor unit load cell in a physical breakwater model. Three extensive test series were completed to assess the influence of a wide range of breakwater and sea state parameters, and data reduction and analyses techniques were developed to estimate the structural response as a function of the critical independent parameters.

Using these data, improved design criteria, which consider both the hydraulic and structural aspects of breakwater design, have been developed for the dolos unit. Perhaps even more significant is the fact that the basic instrumentation scheme, model testing program, and data reduction and analysis procedures could be utilized to develop improved design procedures for other armor units, or to assist in the development of a new armor unit.

This study aims to establish the effective construction scheme of the rubble mounds which are constructed by discharging a large amount of rubble from hopper barges. The numerical simulation technique is developed to estimate the spatial geometry of individual rubble mound discharged from a hopper barge. The effective distance between the discharge sites of the barge which makes the uneven property of the mound surface a minimum is discussed in connection with the change in the spatial geometry of the discharged rubble, the water depth and the direction of barge allocation.

The Dutch dikes along the North Sea and the tidal inlets need protection against erosion by wave attack. In 1983 two tests on full scale were carried out, concerning the resistance of large grass sods taken from Dutch sea-dikes, against wave attack. For both tests in two different ways grass sods of 7500 to 9000 kg were dug from se-dike slopes, transported and placed in test facilities. In Test 1. the hydraulic load was performed by irregular waves up to Hs = 1.85 m, in Test 2. the hydraulic load was a simulated regular wave run-up with velocities until 4 m/s. Test 1. affirmed the possibility of a grassmat to withstand a design storm surge on a dike slope of 1:8 without severe damage. Test 2. showed the limits of erosion resistance and affirmed that the erosion resistance of a grassmat on sandy clay is concentrated in the upper layer of soil and roots.

Using cylindrical caisson for a large-scale breakwater is not popular so far in the world. Never-the-less with the completion of Port of Kaohsiung's 2nd Harbor Entrance Project, the engineers in Taiwan has encouraged by the successful experience in this field due to its simple construction method and low cost consideration.

This paper reviews knowledge acquired in the Alaskan Beaufort Sea pertaining to three methods of Arctic slope protection: (1) sacrificial beaches, (2) “soft” (geotextile based) armor, and (3) linked concrete mats. Design considerations and performance evaluations are presented for each method, along with recommendations for future use. It is concluded that the existing technology is capable of providing reliable protection for structures located in coastal and nearshore areas, and that more durable systems will be required to withstand the extreme wave and ice loads which can occur at exposed sites in deeper waters.

The feasibility of a new type of breakwater is explored through a combined program of theorteical analysis, laboratory experiments and a prototype field test. The breakwater consists of a series of low height, shore-parallel bars which are placed just outside the surf-zone. The incident wave field is strongly reflected when the incident wave length is equal to twice the distance between adjacent bars. The breakwater acts to shelter the beach against storm wave attack and to build a tombolo sand deposit behind the breakwater. Theoretical and laboratory studies indicated that the Bragg reflection breakwater concept may have considerable merit, however, a small scale field test served to demonstrate the many practical difficulties in implementing the concept.

This paper describes the scale effect in experiments on the stability of armor units from the point of view of the wave forces. The relationship between the drag coefficient C_D and the stability coefficient K_D of Hudson formula is theoretically derived to be that K_D ∞ C_D^-3 on the condition that the inertia force is negligible. Three kinds experiments were performed by using various sizes of Tetrapods ranging from 16 g to 6800 g according to the Froude law for scaling, 1) to measure wave force on an armor unit placed in an armor layer of a breakwater, 2) to determine the drag and inertia coefficients in wave fields, and 3) to determine the drag coefficient in a steady flow. It is found that the wave force in the small-scale experiments is relatively larger than that in the large-scale experiments. As the wave height increases, the drag force becomes predominant in comparison with the inertia force. It is concluded that the scale effect of the wave force on armor units is mainly due to the change of the relative drag force number as a function of Reynolds number.

The reshaping of dynamically stable breakwaters was studied subject to variations in armour stone, (gradation and shape), wave characteristics and duration of wave attack from head on waves in a two dimensional wave flume. Tests were undertaken at the Coastal Engineering Research Laboratory of Queen's University, Kingston, Canada using irregular waves. Profiles of the structure during the various stages of reshaping were measured using a semi-automatic profiler developed for this study.

The volume of stones and the initial berm width required for development of a stable profile along with the extent to which the toe of the structure progressed seaward were chosen as the characteristic parameters of the reshaped breakwater. The results indicated that the toe width formed as a result of reshaping and the area of stones required for reshaping were dependent on the wave height, gradation of the armour stone and duration of the storm. The initial berm width required for reshaping was also found to be dependent on the wave height, armour stone gradation, percentage of rounded stones in the armour and the duration of the storm.

The force exerted on a vertical wall due to obliquely-incident waves has been analyzed in this paper from the calculation of short-crested waves system by a numerical scheme. The numerical model preserves the wave elevation in an implicit function form for the dynamic and kinematic boundary equations on water surface, the accuracy is then improved. It is found that the maximum loading in onshore direction may be caused by obliquely-incident waves as the water is greater than intermediate depth. However, the onshore loading does not necessarily be the greatest force for such instance. The greatest force could be produced by the offshore loading under steeper normal standing waves. While the wall is in the shallow water region, the greatest force is occurred in onshore loading at crest by normal standing waves. This research also shows good comparisons from some experimental results.

During the harbour extension works at Zeebrugge a new armour unit has been developed : the HARO. Tests in wave flume and wave tank showed a very good hydraulic stability, comparable to that of the dolos. Static and dynamic loading tests in the laboratory and pendulum tests on site confirm the HARO's excellent structural performance, comparable to that of the cube. Based on the results of exhaustive investigations one can state that the HARO is a safe and economical solution for the protection of maritime structures.

In the wave flume of the Institute of Hydroengineering in Gdansk experiments were performed with a vertical rigid cylinder supported by elastic springs at the bottom and top in the inline and transverse directions (four degrees of freedom). The displacements at the positions of the springs were measured. A Kalman filter of the type as presented by Wilde and Kozakiewicz (1984) and by Wilde and Romanczyk (1989) were used to decompose the measurements in components with frequencies equal to the multiple of the wave dominant frequency. The paper discusses the analytical procedure and experimental results.

The characteristics of the “Bore-like surf beat”, one of long wave phenomena occurred on the reef coasts, are clarified by the field observations and experiments. And also, wave forces on breakwaters in the reef coasts are verified through the experiments.

A new pressure formula, taking into account of the wave force due to “Bore-like (Tsunami-like) surf beat”, is suggested, and it is shown that the calculations by the formula show good agreement with results of experiments.

Laboratory experiments were performed in a wave basin to reproduce the local sea bed scour around large-scale maritime structures. Experiments suggested that the mass transport velocity and wave-induced steady currents were essential in modeling sand transport. Based on experimental observations, a numerical model was presented for the local scour around a large circular cylinder. The applicability of the model was confirmed with experiments. Although some discrepancies are still present between experiments and computation, the model seems to provide a basic framework for the local scour around large-scale structures. Future studies are expected on the three-dimensional boundary layer flow as well as on the wave nonlinearity.

Model tests were done to investigate the effect of waist thickness on the stability of dolosse under irregular wave attack. A comprehensive test programme resulted in a better understanding of dolos behaviour under a wide range of different conditions.

Rubble mound breakwaters are structures with both random loading and random boundary conditions, such that a single set of stress measurements cannot characterize the loading. This paper presents a method to determine a design stress for wave induced loads based on the design wave height, a designer specified exceedance probability, and a site specific parameter, ks, which can be empirically determined from physical model measurements. The methodology was developed from analysis of many field measurements of wave loads on 38-tonne dolosse armor units at Crescent City, California.

The design of breakwater in deep water requires a reliable assessment of the forces, it will be subjected to and full understanding of the behaviour of the structure. Wave forces on a vertical wall caisson of composite breakwater vary with the configuration of the rubble mound foundation as well as the condition of incident waves. If the conditions are very unfavorable, powerful impact breaking wave forces may act on the vertical wall caisson. It is an important subject to clarify the generation conditions of impact breaking wave forces, so that the breakwater under design would not be exposed to a very strong breaking wave force.

The concept of a stone revetment where rock movement is expected is relatively new. For people familiar with gravel and shingle beaches, the idea is an easily accepted extension of the traditional statically stable revetment. However, for engineers familiar with classic breakwater design, the idea that a dynamic structure can provide suitable shoreline protection raises many doubts. Recent research in the Netherlands by van der Meer (1988) and colleagues has gone a long way toward answering difficult questions regarding the development of sound design criteria for dynamic revetments. This paper discusses progress in quantifying the problem and defining solutions. Laboratory model tests were conducted at the Coastal Engineering Research Center (CERC) to confirm and extend Dutch research. These tests are described and important findings presented, including a simple quantitative method to predict the amount of stone required to protect a bulkhead, i.e. critical mass. The method to protect a bulkhead appears to have more general implications which could lead to extensive application to a wide range of shoreline erosion problems. Analysis of data is continuing.

A statistical model of storm duration has been developped and successfully applied to wave records on various French sites. Distribution of storm durations at various significant wave-height levels are fitted to Weibull curves. Extrapolation of these distributions to rarer storm events is presented, and enables the estimation of return period of any storm as a function of the exceeded wave height and the duration of this exceedance.

These results on storm duration have been applied to breakwater design, with the help of laboratory tests displaying the influence of duration on breakwater damages.

Coastal defence is an important issue in the Netherlands since more than half of its territory is situated below sea-level. Coastal erosion causes a permanent threat for this largely natural system. As a consequence flood-protection of the polderland may not everywhere be sufficiently guaranteed in the future. Besides dune area with valuable interests may be lost. These problems will worsen due to sea-level rise.

In 1987 the minister of public works instructed to perform a study to identify these problems and to develope alternatives for an overall coastal defence strategy in the Netherlands and to determine the effects of implementation of the alternatives over a time interval of several decades. The impacts of 4 alternatives have been determined with a policy analysis model, which has been especially developed for this purpose.

Model calculations indicate that the more (structural) measures for coastal defence are taken the less are the losses of dune area and the higher the costs. As time proceeds (up to 2090) the differences in costs between the alternatives decrease, while losses for some of the alternatives rise highly. Accelleration of sea-level rise results in an increase of costs and losses of dune area. A remarkable increase of impacts shows up when a change in wind climate is introduced.

In 1986 the TOW Coastal Engineering Research Programme was merged with the Coastal Genesis Research Programme. Rijkswaterstaat, the financer and initiator of these programmes, concluded that the growing interest in the longer term development of the coast in relation to sea-level rise and other climatic changes made it necessary to approach the research questions from a wider perspective than just the coastal engineering discipline. Geology, historical geography and physical geography were included as disciplines, and a multidisciplinary approach to a holistic understanding of the Dutch coast was undertaken. The results of the first phase of the Coastal Genesis Programme became available in 1987, comprising a reconstruction of the Holocene evolution of the Dutch coast, and a view on the available knowledge regarding the physical processes responsible for this evolution. A summary of the reported conclusions is given here.

This paper presents the first results of a study on the wind and wave climate along the Dutch coast of the North Sea. The analysis of time series of wind force, wave height, wind direction, etc, identifies the occurrence of long term changes in wind and (wind)wave climate. In the Dutch Coastal Defence Study, the results of these climate analyses were used to develop the so called “unfavourable hydro-meteorological scenario” with which an evaluation was made of the extreme impact of the shore line retreat of the Dutch coast. The wind and wave climate study is part of a project that examines the relation between changes in the regional wind and wave climate and the long term and large scale morphological development of the Dutch coast.

This paper is concerned with the sand balance of the Dutch coast over the years 1965-1986. A method for calculating the balance based on yearly measured profiles is described. Besides the more general results for the Dutch coast a more detailed example of the Schouwen peninsula is given.

Yearly averaged transport vectors have been computed along a number of rays, perpendicular to the coast, on the shoreface of the Holland coast. Variation in time and space of wave conditions, tidal flow and long-term average near-bottom flows are considered in fine detail. Special attention is paid to various correlations, both between different velocity components and between velocity and water level, as these correlations play a major role in determining direction and magnitude of net transports. Calibration is performed against both long-term and short-term observations of developments. The resulting transport pattern helps to explain some observed large-scale developments of the Dutch coast.

In order to predict the amount of dune erosion due to a storm surge, a model for cross-shore transport during storm surges has been developed. During a storm with intensive wave breaking, suspended sediment transport is predominant. In the model the nett local transport is computed from the time averaged sediment concentration and the time averaged (secundary) flow profiles. Both profiles are related to the local hydraulic conditions. Finally the computed erosion profile development and the amount of dune erosion are well predicted by the model.

The Coastal Defence Study, as executed in the Netherlands, involved a 100-year prediction for the coastline evolution of specific parts of the Holland coast. In order to create this prediction a mathematical two-line model was used. In this model the effects of wave driven as well as tidal transports are incorporated, also cross-shore transport is implemented. The model is calibrated using sand balances, derived from measurements of coastal profiles. Qualitatively correspondence is reached, but quantitative differences between model and nature remain.

This paper is part of a series of papers on a study of the Dutch Coast and describes the method used to determine the sand losses of the North Sea coast to the tidal basins of the Dutch Wadden Sea in the North and the estuaries in the South of The Netherlands. The sand losses can be caused by sea level rise, sand mining, closure works, natural land accretion in the basins or bottom subsidence due to gas extraction. The results of this study were relevant boundary conditions for the study on the behaviour of the North Sea coast in the vicinity of the tidal inlets (paper No. 9 of the Dutch Coast of Stive, Roelvink and De Vriend of these proceedings).

Using the coastal evolution of the Netherlands in the Holocene up to the present as an example and a test case, a coastal evolution concept is proposed and materialized with which shoreline position changes for different sea level rise scenarios are predicted. The (more generally applicable) model applies to quasi-uniform coastal stretches. It accounts for morphodynamic processes from the shelf to the first dune-row, and integrates over coastal units of approximately 10 km alongshore length. The added value, compared to earlier published concepts or models, lies in the full inclusion of cross-shore and alongshore processes, and in the distinction between a – with respect to sea level rise – instantaneously responding active zone and a noninstantaneously responding central shoreface zone. Relevant differences have been found to exist between coastal cells on the closed and the interrupted coast. An important conclusion is that the cross-shore effective Bruun-effect is only of limited importance. This is especially true in the case of the interrupted coast. Longshore sand transport gradients are very large there. This is mainly related to the sand demand which is placed on coastal stretches adjacent to estuary mouths of those estuaries which tend to follow the sea level rise.

Large scale coastal behaviour (LSCB) regards the development of a certain coastal stretch (order tens of km's) in time (order decades). Knowledge of LSCB is still very limited. Three approaches for analysis of LSCB are distinguished with increasing complexity and necessary level of knowledge, viz. the geostatistical, the phenomenological and the modelling approach. Every approach has serious shortcomings to such an extent that predictions of LSCB are as yet impossible. Nevertheless it is necessary to develop a research strategy which may serve future studies on LSCB.

In the Netherlands up to the fifties of this century, dike and groyne building were the most important coastal defence measures. Because of the increased awareness of the importance of the sandy coast the demand for a more environmentally friendly coastal protection measure increased. However, artificial beach nourishment was very expensive. In the past two decades the costs of nourishment have decreased as a result of technical developments in the dredging industry, and erosion control by means of sandfills has become in general cheaper than the construction of groynes and dikes. In that period also large volumes of sand have become available from harbour extension and maintenance dredging. These have been the conditions that led to a new coastal defence method. Up till now nearly 50 beach nourishment projects have been carried out, with a total amount of 60 million m³.

This paper provides a review of the projects carried out in the last 4 decades, with a description of the most important aspects.

Because of the increasing need of sand for use on land and coastal protection, offshore sand extraction has become commonplace. On the other hand, the practice of beach nourishment has become a common method to combat beach erosion. Here, economic and other interests may be served by dumping the sand not directly onto the beaches, but nearshore. Within the framework of the Coastal Defence Study for the Dutch coast and the preparation of a Marine Aggregate Extraction Plan model studies were performed in order to obtain indications and quantitative estimates of the morphological effects of offshore sand extraction and nearshore profile nourishment. The major results of the model computations are:

- offshore sand extraction landward of the 16 m isobath may affect the coastline within a century.

- nearshore nourishments landward of the 7 m isobath have positive effects on the coastline.

The model computations could not be validated in the field. Due to the relatively large margins of inaccuracy the political implication of this study is that sand extraction landward of the 20 m. isobath is not licensed.

Some parts of the Dutch coast are facing extreme erosion as well as very poor reserves and for those locations the application of a somewhat different approach can be profitable or (in time) even necessary. This approach was roughly investigated as the "Seaward Coastal Defence" strategy. The effectiveness of this approach has to be compared with maintenance by beach-nourishment as such as well as the effects it will have elsewhere on the morphological system. A pilot-study has been defined at the isle of Texel. The north-point of this island suffers severe erosion. Several beach-nourishments have been carried out here. Unfortunately the effectiveness was not always as planned. This paper contains a brief description of the seaward strategy as such and more specific the preliminary results of the 'Texel' pilot-study as obtained so far.

The design, construction and performance of a shore-parallel nearshore breakwater at Ocean City, NJ (USA) are described. The breakwater, which is connected to a long, high, impermeable groin at 9th Street, was constructed from armor stone removed from the groin when the groin cross-section was lowered. Constructed over a two month period in late 1987 and early 1988, the breakwater caused accretion immediately behind it and erosion along downdrift beaches. A beach fill of 40,000 cu yd of sand was recommended as a part of the original project; however, the beachfill was not placed until the spring of 1990, more than two years after the breakwater/spur was constructed. The project, in the absence of the recommended beach fill, is performing as would be expected. The sand within the groin compartment has been redistributed with a net gain behind the breakwater just downdrift of the 9th Street groin and a loss updrift of the groin at 11th Street.

Criteria are presented for predicting the long-term shoreline response to a single detached breakwater. The criteria, expressed as two algebraic equations, distinguish tombolo formation, salient formation, and limited shoreline response, and were developed by calculation-intensive application of the shoreline change numerical model GENESIS for a large number of wave conditions and structure configurations. A unique feature of the modeling effort is explicit incorporation of wave transmission at the structure. The two other major non-dimensional parameters in the criteria are the length of the structure divided by the average wavelength at the structure and the average deep-water significant wave height divided by the depth at the structure. Prediction of the generalized criteria agree with available field data.

Hysteresis in the relationship between suspended sediment concentration and the flow velocity is shown to be influenced by sediment-induced flow stratification in high energy coastal environment. A 1-D numerical approach combined with field observation from Hangzhou Bay, China is used to highlight the effects of concentration dependent settling velocity, buoyancy stabilized mass diffusion and bottom sediment fluxes on hysteresis. Typical formulations for the bottom fluxes are believed to have limited utility in high concentration environments.

This paper reports on measurements taken on a beach at the end of a large bay of restricted width. Large waves can only be generated along the long axis of the system, so that the angles of wave approach relative to the beach tend to be small (0 – 15°). Attention is therefore focussed primarily on the on-offshore structure of the flow. The field measurements and study results are discussed.

Breakers in the surf zone are saturated, that is, the wave height at any point is limited by the local water depth. The larger waves in a storm break further offshore making the surf zone wider but leaving the wave height in the inner surf zone the same. Why the beach will erode during the storm? To answer this question, there has recently been considerable interest in the long period waves of one to several minutes in period. The field observation has been carried out for more than one year to acquire the field evidences of berm erosion due to the long period waves. Based on the data obtained, two typical evidences of berm erosion will be shown. A critical level of berm erosion will be discussed, which can be predicted with the mean sea level and the height of long period waves at the shoreline.

The set-up and results of a field experiment on sand-dike breach erosion are described. It is found that the breaching process for the 2.2 m high sand-dike is similar to that in Visser's (1988) laboratory experiments with a 0.6 m high sand-dike.

Confrontation of Visser's (1988) breach erosion model with the field data shows reasonable agreement for the first stages of the breaching process. As yet the model is not applicable to the final phase of the sand-dike breach erosion. If applied then it would fairly overestimate the breach growth in this final stage.

Empirical orthogonal functions in one-and two-dimensional formulation are employed to identify the variability of longshore and cross-shore features in different scales, ranging from days to years. The prototype topographic data was measured in coastal zones of the Baltic and Black Seas. Conspicuous features such as berms, bars, salients and cusps can be identified through e_2x. Temporal variation of cross-shore transport is reflected in c_2t and e_3x while linkage of c_3t to coastal factors remains unspecified.

The beach processes due to regular waves, grouping waves and random waves are studied. The beach models which consist of coarse or fine sand are set to consider bed load and suspended load sediment, respectively. To obtain the equivalent beach topographies between regular wave and random wave conditions, the concept of same energy flux and same representative wave height are applied.

A QUASI-3D model for suspended sediment transport based on an asymptotic solution of the convection-diffusion equation is developed for currents and waves. The influence of waves on the sediment concentration is included through the diffusion analogy and the bed boundary condition. Validity analysis provides conditions that permit the check of the model applicability. The presence of waves enhances the validity area of the model considerably. The model compares favourably with a 3D model and experimental data.

A computational model based on sediment conservation and transport equations is presented. The bottom shear stress associated with breaking waves is the primary forcing function for the model. A breaking wave model is used to calculate distributions of wave height and mean water level setup across the surf zone and the profile elevations are updated at the end of every time step for each grid across the profile. Examples of the model prediction using large wave tank data are presented. The results of the tests are encouraging for the prediction of bar formation and migration.

A quasi three-dimensional numerical model for detailed sediment transport calculations on a beach is developed for a micro-computer. It includes a complete description of the wave climate and wave induced currents over the study area. Vertical concentration profiles of suspended sediments, calculated locally to include the effects of shear stress on the bottom and the intensity of turbulence due to breaking waves, are superposed on local velocity profiles to yield sediment transport rates. Sediment transport is calibrated with laboratory experiments performed on a straight sand beach.

This paper describes a numerical model of three-dimensional beach change as produced by breaking waves and wave-induced currents. Cross-shore lines spaced at intervals alongshore form the basic calculation element, and cross-shore and longshore transport rates are calculated independently on the lines. Transport rates are coupled indirectly through the mass conservation equation and a contouring routine that determines local depth contour orientation for calculating wave transformation. The model has reduced calculation time compared to fully gridded hydrodynamic and beach change models, yet allows representation of complex boundary conditions, coastal structures, and movement of longshore bars and berms. Results of two test series are presented, one for the movement and protective functioning of a linear mound constructed of dredged material, and the other for impoundment at a jetty.

Experimental visualizations have been performed of the oscillatory flow and of the sediment motion produced by a sea wave propagating over a rippled bed. In particular the dynamics of the vortex structures generated by flow separation at the ripple crest is investigated along with the mechanism through which sediments are lifted up from the bed and carried in suspension. A numerical model is formulated and the results obtained are successfully compared with the experimental data.

Wave-orbital velocities at 20 cm above the bed and continuous vertical profiles of near-bed suspended-sediment concentration were measured seaward of the surfzone over a rippled bed. Sand was intermittently entrained from the bed and ejected upwards in discrete clouds. The typical sequence of events in the entrainment cycle, with sand being rolled up in a thin near-bed layer and ejected into the flow on alternating wave strokes, is attributed to the action of vortices shed from the bed ripples by the reversing flow. The orderly cycle of events causes a time-domain correlation between the wave fluctuations of velocity and concentration, and the resulting sediment flux is comparable in size, and at some elevations directed opposite to, the mean sediment flux.

A study of hydrodynamics and multicomponent sedimentary processes associated to unbroken water waves propagating in coastal areas is presented. For the specific case in which the waves are normally incident, weakly non-linear and dispersive, and the foreshore profile displayed a gentle slope, a model of wave-induced heterogeneous sand transport and resulting topographical and granulometric changes is developed. The general conclusion of this study is that the model presented describes some of the observed features of the sedimentology of longshore bars systems.

The Canadian Coastal Sediment Transport Programme (CCOAST) is designed to increase our knowledge of the dynamics of non-cohesive shorefaces through field measurement of sediment transport and the response of the bottom boundary under combined waves and currents. Suspended sediment concentrations are measured using Optical Backscatterance Suspended Solids Sensors (0BS-1P – D & A Associates) and a new multi-frequency, Remote Acoustic Sediment Transport system (RASTRAN). Near-bed velocities are measured using electromagnetic and eleqtroacoustic flowmeters; cross-products of velocity and concentration provide estimates of sediment flux. The bottom boundary response is determined at the roughness (bedform) scale using a High Resolution Remote Tracking Sonar (HRRTS II) and at the shoreface scale using Depth-of-Activity rods and standard survey. Suspended sediment flux is extremely episodic, responding to individual waves, wave groups and low frequency motions, including mean flows. A significant timedependent transport is revealed in the vertical structure. Net transport includes contributions from wind waves, low frequency waves and mean flows. A sediment transport balance controlling local slopes can therefore be achieved by spatial and/or temporal divergence of these components. Reservations must exist with respect to transport models which use depth and/or time averaged terras and which fail to recognize the distinct components of fluid motion contributing to the total sediment transport.

The southern Gold Coast beaches have suffered serious erosion (approximately 7M m3) since the construction of the Tweed River training walls to the south of the Gold Coast in 1962-1964. As it was considered that nearshore nourishment would be the likely method chosen to solve this problem, analytical studies were undertaken by the Water Research Laboratory for the Gold Coast Council. This was done in order to provide a qualitative assessment of the impact on coastal processes of the dredging and nearshore nourishment at various depths and to examine the applicability of the modeling techniques for predicting future nearshore nourishment behavior and options. Forward ray wave refraction computer modeling techniques were employed to examine the sensitivity of the shoreline wave climate to offshore dredging. Model studies showed that, in general, dredging in areas seaward of the 20m depth contour is not likely to have any significant effect on shoreline conditions. Despite adverse conditions, the works carried out in 1988 appear, as predicted by the model investigations, to have been effective in protecting the foreshore and allowing accretion of a usable beach.

In recent years, two-dimensional numerical models have been applied successfully not only to large scale tidal computations, but also to small scale detailed flow computations (Bosselaar et al., 1987, Leendertse et al., 1989). However, the application of three-dimensional models for fine-grid computations is very limited due to the very high requirements for computer times, caused by the explicit numerical algorithms which are being used. This problem is now alleviated by use of an alternating direction implicit computation method for three-dimensional flow- and transport-modeling. With this method, presented in this paper, the effective computation time is about an order of magnitude smaller then with previous models with similar accuracy.

Gold Coast beach profile changes have been analysed using a dynamical model describing wave-bed interactions. The model explains the formation and dynamics of prominent longshore bars which are characteristic features of the studied nearshore. Model predictions are compared with field observations and measurements in several locations. Obtained results justify protection measures (e.g., nourishment of offshore bars) applied in the study area by the Gold Coast City Council.

Mapping sea bottom topography with imaging radars or optical sensors can add valuable information to traditional bathymetric surveys. This paper describes a series of (already performed or planned) experiments in which the Dutch Bottom Topography Group is involved. It is shown how radar and optical imagery can be used to map the sea floor.

This study was designed to evaluate the equilibrium beach concept for conditions of ‘rapidly’ varying water level, experienced on the North American Great Lakes. It was determined that the mathematical expression h(x) = Ax^2/3 is appropriate for describing sandy beach and nearshore profiles of the Great Lakes. In addition, a correlation was found between lake-level change and the shape factor A, that indicates a phase lag in beach and nearshore response to "rapidly" changing water level. Results from this study also raise some questions about the reliability of determining the shape factor A directly from sediment size. Reliability in the determination of A may be related to the stability shape of the profile relative to its equilibrium shape.

Estimating dune erosion during severe storm events continues to be a major coastal engineering problem. Most beach replenishment projects include storm protection berm and dune systems that must be sized to survive some design storm event. Given this and other requirements, methods of predicting dune erosion must continue to evolve and improve. In this paper, revisions to the Kriebel and Dean model are reviewed that remove some of its previous limitations. These modifications, primarily in the onshore boundary region, enable more realistic simulation of a variety of beach and storm conditions while retaining the general simplicity of the original model.

Two series of experiments were conducted in a new Large Oscillating Water Tunnel of Delft Hydraulics with the main objectives to study the behaviour of bedforms and the transport of sand in controlled wave conditions simulated at full scale (1:1). The main results are:

i) Bedform dimensions are strongly influenced by the type of oscillatory flow. The transition from rippled bed to plane bed conditions – as predicted by Nielsen's (1979) empirical relations – was not observed for sinusoidal waves but did occur for (ir)regular asymmetric waves.

ii) The net sediment transports as measured during (ir)regular asymmetric wave experiments in plane or almost plane bed (sheet flow) conditions show a consistent linear relation with the third-order moment of the velocity <U³> and show a good agreement (within a factor 2) with the energetics total load sediment transport model of Bailard (1981). A calibrated formula of the form q_s = A · <U³> shows a reasonable agreement (within a factor 1.5-3)s with available existing datasets over a wide range of measured values (factor 1000).

In order to get the data about bed-load transport rate and bottom topography changes under storm wave conditions, field equipments were newly developed and tested at the T-shaped Observation Pier(TOP) on the Ogata coast facing the Japan Sea. The surf sled with a pressure senser can precisely detect the spatial water depth changes along the survey lines. Based on the compiled data, the 3D bottom topography changes for a winter season connect with the appearance of a dip and shoal system on the Ogata coast. The bed-load sampler has also developed to operate in rough sea state in a plane bed conditions. The data were well fitted with the law of cross-shore bed-load transport proposed by Kawata(1989a).

As a first step to develop “numerical mobile bed” as a promising tool for the study of sheet-flow problem, a computational scheme based on a single-phase fluid model has been constructed by formulating the vertical transport of mass and momentum within the sheet-flow layer. The numerical results have shown the satisfactory agreements with the experimental results in velocity profile, sediment transport rate, and others. Further the difference in the period of oscillation was found to have appreciable influence on the temporal evolution of the sheet-flow thickness, indicating the importance of unsteadiness in the sheet-flow dynamics.

Field measurements of sand concentration in the sheetflow layer have been carried out in the swash zone on a beach on the island of Norderney. Simultaneous registration of both sediment concentration at three levels and the velocity of wave induced water movement in the boundary layer gives an impression of the phenomenon sheetflow for prototype conditions.

Under stormy waves on a sandy beach, a movable flat bed may appear when ripples have been washed out. Studies of the sheet flow have received much attention recently because a large amount of sand is transported under such conditions. However, most of the existing studies on the sediment transport are dealing with the phenomena under small tractive forces because this is easily reproduced in a small wave tank. Until now, very little understanding has been obtained on the sheet flow phenomena. The present paper develops a numerical model to analyze the oscillatory sheet flow. The mathematical formulation based on fluid-sediment two-phase flow is performed by expanding models from a unidirectional flow into an oscillatory flow. The developed model predicts the velocity field, properties of friction, and sediment transport rate. The model is shown to describe existing experimental data for oscillatory sheet flow.

Tanah Lot Temple is situated in Tabanan Regency-Bali, on the coast of the Indonesian Ocean. Due to continuous wave attack, wind force, and weathering of the rock bank where the Temple stands, abrasion has occurred which is more and more threatening the existence of the Temple. Considering that Tanah Lot Temple is a sacred place for the Hindu Balinese people and a place of high cultural value, steps to save the Temple are imperative. Measures have been undertaken to protect both the seaside and landside banks of the Temple rock bank. This paper discusses countermeasures undertaken for the seaside bank of the Temple.

This article is an overview of the behaviour of artificial beaches based more particularly on three important examples designed by SOGREAH: La Croisette (Cannes), Larvotto (Monaco), Prado (Marseille)

The main conclusions are:

• a man-made beach in an urban area is generally very well received by the general public,

• the general bahaviour of beaches has proved to be satisfactory and as forecast on the basis of sedimentological scale model studies from the following points of view: stability, water quality, safety, maintenance.

This paper describes a study of littoral sediment transport rate based on three-dimensional hydraulic model experiments performed with regular and irregular waves. Deep water wave conditions, wave heights through the surf zone, wave breaking angles, longshore current velocity distribution and the bed load and suspended load sediment transport distributions were measured simultaneously to form a coherent data set. An expression is developed linking sediment transport rate to wave steepness, beach slope, relative grain size and breaking angle. The expression compares well with published field data.

The Port of Saldanha, situated 120 km north-west of Cape Town, South Africa, is protected from the dominant south-south-westerly swell conditions by a sand breakwater which is presently eroding. This paper summarises the results of the monitoring of the breakwater, which consisted of aerial and beach photographs as well as hydrographic surveys. Another part of the investigation was to examine the reasons for the erosion (found by the monitoring to be increasing) on the basis of the prevailing environmental conditions such as wind, waves and currents. To do this the wind and nearshore wave climates were determined. Subsequently, the optimum alignment of the sand breakwater was calculated by three different methods. The equilibrium slope of the breakwater, which was also determined, indicated that the design slope was too steep. Of the remedial measures considered, namely, the use of sand or rock, the latter was chosen.

To calculate the sediment movement under waves and currents, a simulating system called CW system has been developed by the authors. It can determine the effect of the dredged spoil disposal.

This paper presents a mathematical model of waves, currents and beach change with wave-current interaction. The wave model is based on the time-dependent mild-slope equations extended to a wave-current coexisting field, and is applicable to the computation of wave deformation due to combined effects of shoaling, refraction, diffraction, reflection, breaking and currents. Some examples of numerical computation are shown, and effects of the wave-current interaction on the nearshore waves, currents and beach evolution is discussed. In addition, a simple treatment of obliquely incident wave condition as well as improvement of sediment transport rate formulas are presented.

A 1-to-7.5 scale (midscale) movable-bed physical model was used to validate model scaling criteria selected as most appropriate for turbulence-dominated, erosion of sediment by waves. Two-dimensional flume tests successfully reproduced profile evolution observed in prototype-scale wave flume tests conducted in Germany under both regular and irregular wave conditions. For the case of regular waves, a sloping concrete revetment was exposed, thus validating the scaling guidance for use in studying scour at coastal structures. Comparisons between regular wave and irregular wave profile evolution indicated that best correspondence is achieved when the significant wave height equals the monochromatic wave height, although irregular wave profile evolution takes about twice as long.

The long-term bottom topography changes around the entrance of a fishery harbour were simulated by using the three-dimensional beach evolution model. The practical applicability of the model was demonstrated through comparisons with the actual topographical changes. And the equilibrium state of beach evolution due to construction of a coastal structure could be also predicted by repeating the calculation of the wave and current field and that of the bottom topography change.

Laboratory experiments were conducted to obtain data on sand transport processes and transport rates under combined action of waves and currents using an oscillatory-steady flow tank. The sand transport processes were classified into seven types. Four kinds of non-dimensional flow power parameters were defined, and their linear relations with the non-dimensional net transport rates were found for all the transport types.

Two independent numerical models for prediction of sand wave heights and migration rates and for calculation of pipeline/cable seabed interaction have been developed. Both models are fully dynamic and respond to detailed and comprehensive input describing environmental conditions, seabed and soil characteristics and pipe/cable data. The models are based on state-of-the-art knowledge and include theoretical developments, model tests, field data and advanced sediment transport formulations. The models have been validated and calibrated against field data of sand waves and buried pipelines.

Measured shoreline changes up- and downdrift of several tidal inlets on the eastern coast of Florida are decomposed into even and odd components. The odd component of shoreline change is compared to both analytical and numerical predictions of shoreline response to a shore-normal littoral barrier. A best-fit iterative scheme is used for application of the analytical solution. The numerical solution improves upon the analytical approach by accounting for refraction, diffraction, and wave energy dissipation on the ebb tidal shoal. Predicted and measured changes agree reasonably well for several cases, and future refinements and improved long-term wave data are expected to allow more accurate prediction of smaller-scale features.

Rather than endeavouring to protect a shoreline under severe waves attack, on the northern coast of FRANCE, a new solution a retreat landwards up to 800 m. With a slight retreat (180 m), the orientation of the new shoreline is drawn according to the dominating wave climate and enables the stabilization of the beach. A larger retreat (800 m) could allow the re-estuarisation of a valley. These solutions are tested on a physical movable bed model.

Results on the conditions of initiation of first sediment movement on a flat bed under the action of monochromatic short waves are presented in this paper from a laboratory study. The results indicate the existence of a correlation between the Ursell number and the parameter a. The correlation is of the a=λ Uⁿ type. The values of n and λ, as defined from the experiments, were not unique and were different from the values found in an earlier study by Sunamura. Therefore it was considered necessary to cross-check the values of n using data from real-world beaches.

The present paper describes the different behaviour of the foreshore of a sandy beach faced with stormy events. Two storms have been selected, since they present a fairly distinct wave approaching direction between them, according with wave forecast maps.

For the first stormy situation waves came for west, and the effect was an accretion on the north end of the beach while an important erosion happened on the south side. With regard to the second storm in which waves came from northwest, an erosive trend was observed all along the beach.

Sandy movements are quantified as net volume changes as well as by means of the erosion and accretion percentages, defined from the sedimentary variability indice.

In 1987 Anglian Water (now the National Rivers Authority, Anglian Region) initiated a coastal management study covering 750 km of coastline, the largest and most comprehensive study of its kind in the UK. Stage I involved the selection of the best technical approach to the problem. In Stage II the successful consultants adopted a philosophy of maximising the use of existing data sources through a geographic information system (GIS) incorporating a database and powerful enquiry facilities. In all 19 main variables were included in the GIS database, each with complex data structures in order to provide a comprehensive description of any particular variable. The main variables were selected on the basis that they either provided information on the direct influences or responses of the coast or on their implications with respect to any protection strategy that may be implemented.

Analysis of the data through the GIS led to a number of significant findings with respect to dominant processes on the coast. This work also highlighted the areas in which further data collection and study would be beneficial to formulating the most appropriate long term management strategy. As a result Stage III of the study has been initiated in order to develop a better understanding of the physical processes responsible for the mechanisms identified.

The function of offshore detached breakwater of controlling longshore current and resulting sediment transport under oblique wave incidence is examined mainly through experiments. Although the offshore detached breakwater can effectively reduce incident wave energy, it has little influence on longshore current. More than 50% of the volume of sediment transported in the longshore direction on the natural beach is trapped behind the breakwater. The total longshore sediment transport rate around the breakwater can be predicted more precisely by integrating the local longshore sediment transport rate estimated from the flux model than by the empirical formula of Savage type.

Numerical results show that the beach face of a tidal beach could be considered erosive due to the action of the surface flow formed by the escape of groundwater unless the action of moderate accretive waves, and that the effect will be intensified when the mean level of water table is higher than the mean sea level. Experimental results show that waves act to conceal the effect on the whole. These results suggest that the action of accretive waves are indispensable to maintain a tidal beach in a equiliblium condition. They also suggest that the cause of erosions of Surigahama beach and Iso beach was accumulative workings of the surface flow and insufficient wave action for the restoration during high tides.

A method has been developed to determine the strength of a dike expressed as the mean return period for the critical load.

The method is based on tests carried out on prototype dikes.

This paper is a continuation of the papers of Bakker & Van Kesteren (1986) and Bakker et al. (1988). In this paper the results are presented of ring shear tests on mixture of oil and spheres with the same density (880 kg/m³). In order to study in detail the behaviour of a sand water mixture at high concentrations in a Couette like shearflow in the viscous regime the graindiameter and viscosity are enlarged by a factor 100: respectively 15 mm and 10^-4 m²/s. The test are executed in the so called grain carrousel with shear rates up to 75 s^-1 (Bagnoldnumber N = 240). The experimental set up is such that only constant volume test could be performed.

The test results indicate that at concentration below 30% (vol.) no dispersive stresses are generated. At higher concentrations the generated dispersive stresses increases with concentration and shearrate. From videofilms it was observed that the grains are moving in layers with only small mutual exchange (grainswop).

This paper analyzes and compares radioactive sand tracer experiments with beach profile measurements and calculations of littoral sand transport rates performed at Pointe Sapin, N.B. in the Gulf of St. Lawrence, Canada. The three methods give comparable results. In theory, all sediment tracers should be dealt with the same way, but in practice the nature of the tracers has to be taken into account. Tracers do fulfill the purpose for which they are used and they can solve sediment transport problems when used with knowledge and care. The wide variation among littoral sediment transport formulas should not be attributed to the lack of accuracy of tracer experiments but rather to the fact that expressions are over-simplistic and important factors such as the influence of beach slope and grain size are not taken into account.

This paper is to investigate experimentally the beach erosion control function of a submerged tension-moored breakwater. Two- and three-dimensional laboratory experiments are performed.

The single-peaked, double-peaked and triple-peaked cuspate spits are formed by placing the submerged tension-moored breakwater on an accreted-type or intermediate-type beach. When the submerged tension-moored breakwater is placed on an eroded-type beach, the shoreline denudation is stopped and the shoreline is restored to the original location.

The submerged tension-moored breakwater is concluded to be possibly one of effective measure works against beach erosion in an inner gulf area in which huge waves do not attack.

In order to find the optimum sand fence porosity and the best position for placing a sand fence in front of an embankment, a series of experiments were conducted in a large wind tunnel. The results obtained were that a fence with 40 % or 50 % porosity should be placed at a distance between 15 H and 20 H in front of the embankment, where H is the fence height.

The effects of seawall on the adjacent beach is examined by the three dimensional physical model test. The experiment results are analyzed by means of the volumetric change analysis and the shoreline change analysis. The results show that the groin effects are dominant but they are localized within a region spanning three or four times of seawall length for the cases tested.

A numerical simulation method is described to estimate damage and associated benefits to oceanfront (and "backrow") properties without-, with- and adjacent to shore protection project areas. Damage to structures, land, and structure contents is described — as is the effect of shoreline armoring. Input data include property attributes and estimates of beach recession for various intensity storms. Projects are described by their geometry, life expectency, construction intervals, and effect upon the chronic and storm-induced recession. Projects can include nourishment, armoring, or other types which have a quantifiable effect upon chronic and storm-induced recession. Damage is computed based upon (1) the encroachment of the eroded bluffline to the property by chronic (historical) erosion stress and by storms, and (2) functions which depend upon structure type or land use. Spatial and temporal resolution is adaptable to the user's needs.

As a result of severe coastal storm damage in recent years along the California coast and the continuation of development and redevelopment in hazard prone oceanfront areas, large numbers of coastal protection structures have been built. This same trend has been observed on the Atlantic and Gulf coasts as well. At present, fully 12%, or 130 miles of California's 1100 miles of shoreline have been armored. As the number of structures and their coastal frontage has increased, concern along the California coast and elsewhere has arisen in regard to the impacts of these protective structures on the adjacent beaches. Three Atlantic coast states (Maine, New Jersey, and North Carolina) have responded to this concern by establishing state-level policy which prohibits construction of any new "hard" protective structures.

In the North Sea off the Dutch shore annually tens of millions cubic meters dredged material are dumped. Recent progress in assessing and predicting the fate thereof is described. Field studies and modelling activities are both required in the considered relatively complex situations.

Time series of semidiurnal single values of Tidal High Water and Tidal Low Water and their occurrence times from certain tide gauges in the German Bight were evaluated. Detailed analyses were performed to detect distict periodicities and changes in the trends not only of water levels but also of the occurrence times by means of mathematical-statistical methods.

This paper describes the structural and environmental aspects considered in the installation of a 45 metre × 16 metre steel pontoon equipped as a day visitor centre and installed at Wistari Reef, a southern section of the Great Barrier Reef, Australia.

The data of several field measurements performed in different types of estuaries along the Polish Baltic Coast are presented here. The results of measurement showed how unsteady and changeable the flow conditions and the motion of salt water in tideless estuaries are. The salinity and velocity distributions in the tested estuaries demonstrate that the change of current direction occurs at different depth and different times. It has been stated that the dynamics of the flow in tideless estuaries is similar to that of the tidal estuaries. A method of determination of longitudinal dispersion coefficient for tideless estuary is presented. The results from calculations for tidless estuary using a threedimensional estuary model previously developed for tidal estuaries are presented.

The problem of contaminated marine sediments has emerged as an environmental issue of national importance. Harbor areas in particular have been found to contain high levels of contaminants in bottom sediments due to wastes from municipal, industrial and riverine sources. The paper examines the extent and significance of marine sediment contamination in the United States; reviews the state-of-the-art of contaminated sediment cleanup and identifies research and development needs.

In this paper, stability characteristics of tidal channels of the Western Scheldt and of the Wadden Sea are compared.

The Western Scheldt is a major estuary in the southern part of the country and is the only remaining open sea-arm in that area after construction of the Delta Project. The Wadden Sea is situated in the northern part of the country and the inlets form the connection between the North Sea and the Wadden Sea.

Stability parameters used for the comparison are the maximum discharge, Qmax, the Flood Volume FV (of Tidal Prism) and the mean tidal velocity.

It appears that linear trends can be established between the cross-section and the various stability parameters, for each of the two systems, but that the regression relationship is not the same.

Attempts have been made to develop a universal relationship by introducing a dimensionless stability parameter, in which the dominating hydraulic characteristics are combined. The results are interesting but further research is required to refine this method. The established morphological relationships can be used to predict the changes in channel equilibrium when external changes in the estuary system are introduced. For the Western Scheldt the one-dimensional numerical tidal model IMPLIC will be combined with distinct morphological relationships for the estuary to be used as a predictive tool for the new equilibrium situation created by dredging works.

In numerous navigation channels, the bed consists of muds. The navigability threshold adopted by many ports is currently that which corresponds to a specific gravity of 1.2. This virtual standard does not fully take into account the rheological characteristics of the muds, which may differ from one site to another.

Investigations were carried out to determine the rheological properties of the Gironde estuary muds. Analyses were carried out in the laboratory on samples taken from different sites and on different dates and in situ, using the SR 10 probe developed by SOGREAH and ultrasonic sedimentometric probe developed by the Bordeaux Port Authority. The results demonstrated the occurrence of major spatial and temporal variations.

The two principal phenomena revealed by the research were that with the same dry sediment content:

A proper understanding of the spatial and temporal variations of rigidity is therefore essential to determine and adjust the navigability threshold.

The paper gives an overview of the results of the evaluation studies of the hydraulic aspects of the construction of the Eastern Scheldt Storm Surge Barrier. Subjects discussed are: methods used, design filosophy, results of flow modelling, stability of rubble stone structures, local scour and morphology.

The improvement of the Yangtze Estuary has attracted much attention in China. The fluvial processes of the estuary are rather complicated. The frequent move of sandbanks in the estuarine channels and the bar forming at the mouth give much trouble to navigation.

Investigation on the guiding principles and schemes for the improvement of the Estuary has been carried out since 1958. Comprehensive field survey, laboratory experiments and numerical modeling as well as planning work have been done in the past 30 years with some preliminary achievement. A scale model study with horizontal scale 1:1600 and vertical scale 1:120 was conducted by the Nanjing Hydraulic Research Institute.

The temporal evolution of consolidation and associated rheological properties of mud deposits has been studied by means of laboratory experiments for more than 40 sites throughout the world.

Standard procedures have been used for all these experiments.

The present paper describes the experimental procedures and gives a summary of the results obtained.

The geomorphological and ecological response of the ebb tidal deltas of the SW Netherlands to implementation of the Delta Project during the last decades, has been the subject of an integrated research project. General observed geomorphological trends are: erosion of the delta fronts and of the relic tidal bars in the landward parts of the deltas, and sedimentation in the tidal channels and in longshore bars at the edges of the deltas. Ecologically, a general trend towards richer benthic fauna communities is observable.

The integrated approach in coastal research combining geomorphological analysis of field data with numerical modelling, and integrating hydraulic, geomorphological, sedimentological and biological data, has appeared very promising. Progress in this field would be stimulated by further research of the driving processes in the interaction of hydrodynamics - sediment - biota. The level of detail (spatial and temporal) in geomorphological contributions to such research will have to be much higher than usual in current coastal morphological studies.

A multi-layered model is developed for the prediction of mud transport in an estuary and ports to take into account the effects of submerged dikes and to reproduce the vertical distribution of suspended mud concentrations. A most remarkable result in the field observation in Kumamoto Port was the effect of a submerged dike on reducing the amount of deposition rate within the test trench.

It is predicted from the calculation that submerged dikes 1 m in height from the bottom will prevent siltation in the access channel and anchorages of Kumamoto Port by about 30% of the total deposition without submerged dikes.

A depth-averaged numerical model for the erosion, sedimentation and transport of cohesive sediment is applied to a mesotidal estuary. The transport model is coupled to a hydrodynamic and a wave model. First results are presented as part of the aim to predict year-averaged quantities. Comparisons with measured data of net sedimentation and suspended concentration are made. As the bed of the estuary consists mainly of non-cohesive sediments, the erosion formula is adapted to account for the local (non-) cohesive fraction in the bed. The effect of the spatial variation in this property of the bed is significant. Furthermore the influence of surface waves is investigated and showed to be of great importance in a qualitative and quantitative sense. The model can take into account the dumping of dredged material and gives reasonable results in case of the spreading and sedimentation of mud from one source, as shown by a simulation of the transport of fluvial mud.

The Great Belt Link crosses the Great Belt (Denmark) which is the major strait connecting the Baltic Sea with the North Sea. A law passed by the Danish Parliament imposes an environmental design criteria of 'zero far field effect' on the Link. This paper describes how the challenge of zero effect is met by compensation dredging. The compensation dredging is designed by the use of highly sophisticated numerical models.

In order to calculate mud transport rate, two forms of transport should be considered, which are (1) mud mass transport in mud layer, and (2) suspended mud transport in water layer. The first type of transport (mass transport in mud layer) is greater than the second one in quantity under soft mud or fluid mud condition, therefore the first type of transport is mainly considered. The previously proposed methods to calculate the first type of transport are summarized and discussed. Then a new transport model called visco-elastic-plastic model is derived and some of the results of the model are shown. The model is based on the assumption that the fluid mud layer can be assumed to be visco-elastic fluid and when the magnitude of stress in mud layer exceeds the value of yield stress, the mud layer is modeled as visco-plastic fluid. Results of the new numerical model are compared with laboratory results and some discussions are given.

The tidal characteristics, and consequently the morphology, of an estuary can be altered substantially by engineering works. A method is suggested for estimating the magnitude and the time-scale of the resulting morphological changes. This method is then applied to the Welsh River Usk which will be affected by the construction of a planned tidal power barrage across the nearby Severn Estuary.

An approach to predict flow area of constricted tidal inlets due to a mean sea level rise is developed. Increase of bay surface area is found to be the most dominant factor to increase the flow area due to SLR. A case study is conducted for an entranse channel of Lake Notoro in the Okhotsk coast.

Knowledge of ship behaviour at berth is of fundamental importance for the design of harbours and marine terminals. To develop more precise data on acceptable ship movements than available in the literature, a joint research project was undertaken in cooperation between governmental and research institutions in the Nordic countries: Denmark, Faroe Islands, Iceland, Norway and Sweden. The purpose of the project was to establish criteria for acceptable ship movements in harbours for working and for safe mooring conditions.

The paper gives a brief description of the studies carried out in France during the last years, especially on scale models, in order to improve the knowledge of the behaviour of ships progressing in approach channels subject to siltation.

The strong increase in the capabilities and efficiency of modern marine simulators are briefly outlined as the background for demonstrating the advantages of a much closer coordination of hydraulic and maritime oriented design work in relation to harbour planning and alteration. The interaction of the two disciplines and the implications on the final configuration of two different harbours are described.

This paper describes methods used at Hydraulics Research to evaluate risks of ships grounding in navigation channels due to wave induced vertical motion. Risk assessments are then used as an aid to optimising channel depths with a view to minimising dredging. A mathematical model of ship response in waves in shallow water is presented and two case studies are considered to show how methods may be applied.

A numerical model for the computation of the nonlinear interaction of water waves and floating large-volume bodies is described. The model works in the time domain and is based on the boundary element method (BEM) with internal generation of the incident waves and absorption of all outgoing waves (internally generated as well as reflected). This procedure makes it possible to simulate long time series of nonlinear wave forces even in case of irregular waves. Drift forces are extracted from the time series by filtering. Steady and slowly varying drift forces are computed in 2D and first order wave forces are computed in 3D.

This paper describes the critical situation experienced in the northeast coast of Brazil by an oil and a gas pipeline, in their shoreline reach, due to the growth of a spit, and the remedial measures adopted, which a voided an environmental and economical disaster.

The "Sand bypassing to Playa de Castilla" project consists of a sand bypassing from the zone behind Huelva Harbour dike to "Playa de Castilla" beach, situated in the downdrift direction of littoral transport.

The objective of this project is to artificially nourish a 25 km eroded beach ("Playa de Castilla") as well as to check some coastal engineering theories.

The total volume of sand has been transferred to the updrift extreme of the beach, forming a 2 Km protrusion. The high rate of littoral transport guaranties its quick distribution along the whole beach, following the downdrift direction.

A physical modelling programme was undertaken to investigate the contamination of a berm breakwater armour stone layer with fines. For construction purposes, it became necessary to significantly increase the berm breakwater core crest width (construction roadway) beyond the design value. Thus, contamination of the armour material by fines, much beyond the specified tolerance, immediately adjacent to the core section resulted. As the original design did not consider the influence of this contamination on the armouring stability, the present modelling study was carried out, in response to the above construction limitation. The implications of a contaminated berm armour section was investigated. The results showed that for a high crested breakwater structure, the contamination of the armouring close to the core did not influence the breakwater stability and, profile reshaping was virtually the same for the contaminated and the original structure. This is due to the fact that the contaminated section was not placed in the active berm area. However, for a low crested structure, the contaminated section was less stable that the uncontaminated counterpart, due to the increased volume of water which overtopped the structure resulting in back slope instability.

This paper describes the development of an alternative design for the offshore breakwater of the Sergipe Marine Terminal (Terminal Portuario de Sergipe, TPS), in northeast Brazil. Earlier alternative designs for the breakwater are presented and the chosen solution, a berm breakwater design, is described. Details of the hydraulic model testing carried out during the designing stages are given and the construction methods are explained together with a brief overview of construction controls and organization. In order to accommodate the unfavourable subsoil conditions which were found to be worse than initially expected, a redesign of the offshore breakwater was carried out. The new redesign is presently under construction. The TPS breakwater is the first berm type structure to be constructed in Brazil and one of the few in the world to be constructed over a soft marine clay layer.

The St. Paul Phase 1 berm breakwater was destroyed by two storms within approximately two months of its completion. Investigations made after the failure demonstrate that the cause of the failure was that the breakwater was built from material that was vastly undersized as compared to the recommended minimum rock gradation, and heavily contaminated by fines and overburden from the quarry.

The harbour breakwaters at Durban in conjunction with the dredging of the harbour entrance cut off the sediment supply to the important bathing beaches of the city, necessitating a beach restoration scheme. This paper reviews this scheme. Before 1977, the measures to restore the beaches included an underwater mound, beachfill and rock groynes. After 1977 and following a previous review, it was recommended not to complete the mound, to place initial beachfill, regularly nourish the beaches and to build low-level groynes. Implementation started with a sand pumping scheme and 600 000 m³ of beachfill followed by the building of three multi-functional pier/groynes. Monitoring of the beach restoration scheme enabled valuable experience to be gained on aspects such as the loss rate of sand from the beaches, beach profile and scour levels alongside the pier/groynes and on the determination of a potential erosion line and building control zone. This paper concludes with some important lessons learned from the scheme.

A description of the pre-design investigation, design and construction of a pleasure craft harbour is given. The influence of the breakwaters on the sedimentation pattern in the direct vicinity of the harbour during construction and after completion of the harbour is described. Adaption of the design of some of the harbour components due to unforeseen circumstances revealed during the construction period is also discussed.

Norderney is one of the East Frisian Barrier Islands that extend along the western part of the German North Sea coastline (Fig. 1). Severe dune erosion jeopardized the health resort settlement on the west end of Norderney and necessitated the construction of solid coastal protection structures since 1857. These structures were successful in stopping the migration of the inlet and in preventing further dune erosion; but they were not able to maintain stable beaches. To protect the existing structures against damage artificial beach restoration was initiated on Norderney in 1951. Since then five further beach fills were necessary to affect an active protection by shifting the zone of wave attack seawards. The scouring from the groyne fields had been monitored. Based on these data the sane budget has been investigated. It has been proved, that on Norderney beach nourishment is an appropriate solution to protect the existing structures from failure during severe storm-floods. If we want to lower the amount of losses, we have to nourish the beach not higher than necessary to achieve the protection goals. Above a 'critical beach profile' in certain areas losses of nourished material increase considerably with height. Especially for the Norderney-case we should develop a 'low-cost-technique' for a 'low-rate-beach-nourishment'.

The Acosta Bridge is located in a bend at a natural constriction of the tidally influenced St. Johns River. This natural constriction has resulted over time in scour of the river bottom resulting in present water depths up to 80 feet. The Acosta Bridge is presently being replaced with a new bridge. The existing bridge is located immediately adjacent to a railroad bridge and both bridges are on caisson foundations. Divers have reported that penetration of the foundations for one of these existing bridges is as little as two feet. Construction of the replacement bridge will increase the scour potential. Therefore, scour protection had to be provided not only for the replacement bridge but the two existing bridges. The scour protection for the existing bridges was completed in 1990. The replacement bridge is under construction and scour protection is being installed as the piers are constructed.

Influence functions combine part of the strong body of coastal engineering knowledge, the one-line theory, with the most reliable sources of data: position of the shoreline and geological information about the evolution of the shoreline. Influence functions yield a model of the shoreline erosion or -accretion, which is based upon the charteristic length √4at. This model can be used to reproduce the present shoreline and predict the changes of the shoreline after human interference. Examples are given of the Petrace delta in Italy and the port construction near the Kelantan Delta in Malaysia.