Coastal Engineering 1982

Horizontal and vertical velocities are measured with a hot-film anemometer (HFA) and a two-component laser-doppler velocimeter(LDV) in surf zones on uniform slopes of about 1/30 in two wave tanks. The turbulence generated by wave breaking is detected from the records. Following three aspects of the turbulence are discussed : (1) the distribution of the turbulence intensity in the surf zone, (2) the variation of the vertical distribution of the turbulence during one wave period and (3) the variation of the Reynolds stress during one wave period. It is found that the pattern of the distribution of the turbulence in the surf zone depends on the breaker type. A model is proposed, by extending the turbulent wake theory, to explain the variation of the vertical distribution of the turbulence during one wave period.

This paper describes a mechanism of breaking waves over sloping bottoms in terms of changes in integral quantities of the waves. Systematic computations are made of wave profiles of shoaling waves up to the numerical unstable points by using the K-dV equation with variable coefficients and internal properties such as horizontal and vertical water particle velocities by a stream function method satisfying the conservation laws of mass and energy. Applicability of the numerical results is examined and a relation between numerical unstable points and actual breaker points is found. Characteristics of the integral quantities of shoaling waves are investigated in relation to the existence of the extremum of the energy of the shoaling waves and their breaking inception.

Initial results are described of precise observations of waves shoaling in the nearshore zone. The key technique of the experiments is a 16 mm memo-motion camera system by which long term measurements of waves can be made simultaneously at many locations. Six or seven pairs of synchronized cameras were mounted on a research pier crossing the surf zone. The cameras were focused on target poles mounted on sleds which were towed about 200 m outside the breaker line, and on a line of poles jetted into the sea bottom across the surf zone. Waves transforming in the nearshore zone were observed from about 400 m offshore to the shoreline. At present only the characteristics of the statistical waves, wave height distributions, wave period distributions, and the joint distributions of wave height and period are described as part of the initial analysis.

The main purpose of this paper is to propose a model for prediction of the spatial distributions of representative wave heights and the frequency distributions of wave heights of irregular waves in shallow-water including the surf zone. In order to examine the validity of the model, some experiments of irregular wave transformation have been made. In addition, an attempt has been made to clarify the spatial distribution of wave grouping experimentally. Especially the present paper focusses finding the effects of the bottom slope and the deep-water wave steepness on the wave height distribution and wave grouping.

This paper presents the results of an analysis carried out on wave spectra measured at three nearshore sites along the United States Florida coast when Hurricane DAVID passed these sites in 1979. Included are (a) the variability of the shapes of wave spectra during the stages of growth and decay of the hurricane-generated seas, (b) the presentation of spectra according to various spectral formulations, and (c) a comparison between spectra measured at the coastal sites and those measured in deep water for the same severity (significant wave height) of hurricane-generated seas.

A study was made, using all the available sources of deep–sea wave data around the South African coast, to establish a statistical representation of the wave climate in this region. Data obtained from clinometer observations, the Datawell Waverider and voluntary observing ships were found to be the most useful.

The most accurate source of height, period and spectral information came from the Waverider which has been in extensive use in the region since 1969. To optimise the information value of these data a sophisticated data qualification technique was developed for digitised Waverider data in order to exclude poor data. This technique was also found to be useful in optimising the operational efficiency of a Waverider station.

A clear pattern was obtained for the general wave climate around the South African coast; this is in accordance with the pattern that may be expected from the weather climate affecting this region. Of interest are the SE'ly winds generating increased wave heights in proportion to the distance offshore from the west coast as well as the effect of the Agulhas current, opposing the SW'ly waves along the east coast, also creating high-wave conditions.

A hindcasting methodology is described for the total water level and wave hydrographs at a coastal site during a hurricane. It accommodates phasing of the separate components of the sustained water level (astronomical tide, storm tide, breaking wave setup) , as well as storm variability and coastal bathymetry. Complete hindcast models are utilised, but an intermediate cost and precision is achieved by compromising the number of complete hindcast storms, rather than the precision of the hindcast model. A synthesis technique is developed to predict the response hydrographs of the remaining storms in the historical data set.

The paper refers to pressure gage wave measurements. First order transformation of the pressure spectrum into a surface level spectrum leads to hitherto unexplained discrepancies with prototype simultaneous pressure and level measurements. Use of second order gravity wave theory allows to draw the following conclusions. Second order effects appear to give a reasonable explanation of the observed discrepancies. A complete check would require specially made wave measurements and analyses. Second order corrections do not significantly affect mean values, such as significant height, if the manometer depth is not unduly large.

A field investigation program on waves was carried out in the Weser estuary, German Bight of the North Sea. Wave height and period distributions in this complicated wave climate can be approximated by a Rayleigh distribution. Empirical distributions of the wave heights characterise the different regions of the estuary. The presence of wave grouping as well as the group bounded long waves are shown in a few examples. The necessity of further investigations and analysis is highlighted.

Resonant interaction processes in the surf zone are marked by an anomalous dispersion property (dc/df > 0) as it is well known from resonance absorption processes of electromagnetic waves in dielectrics.

A new method of calculation of wave diffraction around islands, offshore structures, and of long wave oscillations within offshore or shore-connected harbours is presented. The method is a combination of the finite element technique with an analytical representation of the wave pattern in the far field. Examples of application are given, and results are compared with other theoretical and experimental investigations.

A system for indirect programming of shallow-water models is discussed, particular attention being given to facilitating successful operation by novice users. Pre-developed programs are used to check numerical coding of model layout, to execute the finite-difference computations required at each time step and to analyse the computed surface elevations and velocities. Many error-prone steps are thus eliminated and model development is speeded for modest increase in computation costs.

Low-frequency forcing of water currents over the continental shelf of Australia is quite strong and should be taken into account when the flow for durations greater than 1 day is important. In the case of the Queensland coast, the longshore wind generates barotropic continental shelf waves, raising or lowering the mean sea level by as much as 30 cm and generating longshore currents over the continental shelf, even very close to the coast, that are often larger than the tidal currents. These wind-driven currents can reverse sign, flowing alternately northward and southward, although the longshore wind stress, though fluctuating, does not change direction. To reproduce such phenomena in an analytical or computer model of wind-driven currents, it is necessary to extend the offshore boundaries of the model offshore from the continental shelf break.

Based on the linear non-dispersive theory, the reflection of a Converging Cylindrical long wave, of wave length L, onto a Cylindrical shelf, of radius r = a and positive or negative height Δ h relative to an otherwise flat bottom, is study analitically. It is found that these linear approximation agrees well with the existing non-linear numerical solution when the ratio a/L is large enough. It is also found that these two-dimensional reflection process is the contrary of the corresponding one-dimensional case, since the solution of these problem gives a negative reflected wave for a positive step and a positive reflected wave for a negative step.

The South African wave energy program has been underway for several years and has included an analysis of the temporal and spatial distribution of wave energy along the full coast-line, determination of energy attenuation perpendicular to the coast-line at a site on the south western coast, and the development of a wave energy converter which is most suited to local conditions and requirements.

The resource analysis has shown that the inshore power levels occurring along the south western coast are as promising as any elsewhere in the world. A bottom mounted, V-shaped wave energy conversion device driving an air turbine has been found to be most suited to prevailing conditions.

The conversion characteristics of the device are presented, based on 1:100 scale three dimensional and 1:50 scale two dimensional model studies. Preliminary design studies of the proposed conversion system have underlined its potential viability as a cost effective supplementary source of power.

A depression in mid North Atlantic moved at a speed such that it generated high waves of unusually long period and large wavelength. These waves travelled in precisely the right direction to carry them into the English Channel, where they arrived during a time of spring tides and when low barometric pressure contributed to long waves impinging high, on the foreshore, causing appreciable damage to sea defences and property.

This paper deals with the results of hydraulic model investigations of orbital velocities and pressures in irregular waves. Different simulation methods in the time and frequency domain were checked or developed, and the theoretical results compared with measurements. Using simulation methods based on linear wave theory, results with good correlation are obtained, at locations near the water surface, however, a tendency towards over- or unterestimation exists.

The results are reported of velocity measurements in oscillatory flow over rippled beds. Velocities were measured with a laserdoppler anemometer in both an oscillating tray rig and an oscillatory flow U-tube. Both self-formed and artificial ripples were examined.

The results obtained in the two test rigs were compared and it was concluded that the flow fields obtained in the two cases were dynamically similar.

Measurements of the flow field clearly showed the formation, growth and ejection of vortices, as well as a strong surge of fluid over the ripples during and after flow reversal.

The concept of mass transport is theoretically discussed within the framework provided by Vocoidal theory. The Lagrangian mass transport is divided into two parts; firstly treating the fluid as being inviscid and secondly, incorporating viscosity by means of the free surface and bottom boundaries. Eulerian mass transport is defined and is shown to correspond, in deep water, to the net flow predicted by Stokes and others.

A new probability density function of wave heights off the western coast of Taiwan is submitted in this paper. According to the bathymetry of this area, waves from the central part of Taiwan Strait refract to the point of measurement and minor waves generated by local wind add the energy on the major ones; So an analytical solution is to be worked out by assuming that the wave energies are the linear sum of these two sources and convolution integral is adopted. The new model approaches reality better than Ray'leigh's.

The objective of this research is to study the prediction method of hurricane waves around this island, especially in the Taiwan Strait. The paper describes the prediction of hurricane waves used by Bretchneider's (1976) Method and finds out the predicted waves are different from measured waves, therefore the Bretchneider predicted model is modified by the authors and then the modified model is applied to predict waves again. It is found out that predicted waves match well with the measured waves. The results of the modified Bretchneider model are compared with those of the Ijima tracing method and find out the former is better than the latter.

The second part is to apply the modified model to predict the extreme value of wave heights and compute the worse hurricane wave condition of the surrounding sea area around island, within recent score year (1959-1978). The calculated sites are Chu-Wei, Nan-Liaw, Ta-Shih, Cheng-Kung, Pu-Tai, Tung-Kang, Nan-Wan as shown in Fig.1 and Shiau Liu- Chieu totally 8 stations. Then use the Gumbel Distribution TYPE 1 to predict the extreme wave height of each returned period.

A study of wind data collected from the Waglan Island anemometer (Hong Kong) during 39 tropical cyclones indicates that turbulence intensity values in excess of 20% at a reference height of 50 m are likely during extreme wind conditions in a tropical cyclone. The implied surface drag coefficient of approximately 0.01 in these extreme wind conditions is consistent with wind flow over a land surface roughness of trees and suburban housing, but is much higher (by a factor of five) than that predicted by the currently accepted formulae from the reviews of Garratt and Wu for wind flow over a fully developed sea in neutral atmospheric conditions. For wind loading design calculations in extreme wind in tropical cyclone conditions it is recommended that mean wind and turbulence intensity profiles should be calculated with a roughness length z₀ = 0.20 m in the Deaves and Harris wind model.

A non-linear wave refraction model was developed which allows for the combined refraction and shoaling of Vocoidal waves over an arbitrary sea bed. The effects of bed friction and percolation are also catered for. The method is based on the cirular arc technique which is widely used for linear wave refraction. The method was extensively tested against Vocoidal wave refraction results obtained previously for a plane beach. A comparison of Vocoidal and linear wave refraction over an arbitrary sea bed indicated that Vocoidal waves refract less than linear theory, thereby yielding higher wave heights and angles of incidence at the breaker line. This result is in line with results of non-linear refraction over parallel bed contours quoted for other non-linear wave theories in literature. Further work is required before caustics can be adequately treated. Future research should include wave spectrum transfer and the re-evaluation of empirical relationships in use in the shallow water region and which will use this new higher-order refraction technique.

The question of how waves and currents interact, especially in the near-bed region is of considerable importance in relation to sediment suspension and sediment transport. Whereas empirical relationships provide useful estimates and indications in relation to the data on which they are based, a more thorough understanding of the physical processes at work is necessary for interpreting sediment transport behaviour in a more generalized way. Clearly the conditions under which flow reversal occurs near the bed, and also the extent to which wave motion may modify the current induced turbulence in the boundary layer, are both of great interest, and these and other aspects have been included in the present study.

The research program was designed to look initially at the interaction between waves and currents in the absence of sediment, in order to define the mean velocity components, the structure of the turbulence, and the shear stresses. The study proceeded from experiments on waves alone, to waves propagating with the current and against the current. In all three cases the tests were carried out in the first instance with a smooth bed and subsequently with a rough bed consisting of two dimensional triangular slats. One of the main areas of interest was the height to which the water was disturbed above the bed when acted on by waves alone, and the comparable situation when a current was superimposed on the waves. Since the characteristics of the turbulent current were measured independently, it was possible to deduce whether there had been any interaction between the waves and the current, and also to infer what might happen to the distribution of the sediment which it was assumed would be put into suspension in the two cases. In the second stage of the research separate experiments were carried out in a standing wave channel and an oscillating water tunnel, using lightweight bed materials, in order to observe whether the inferences made from the clear water study were borne out by comparable changes in the distribution of the sediment in suspension.

The interaction of a gravity wave with a steady uniform current is described in this paper. Numerical calculations of the wave length change by different non-linear wave theories show that errors in the results computed by the linear wave theory are less than 10 percent within the range of 0.15 ≤ d/L_s ≤ 0.40, 0.01 ≤ H_s/L_s ≤ 0.07 and -0.15 ≤ U/C_s ≤ 0.30. Numerical calculations of wave height change employing different wave theories show that errors in the results obtained by the linear wave theory in comparison with the non-linear theories are greater when the opposing relative current and wave steepness become larger. However, within range of the following currents such errors will not be significant. These results were verified by model tests. Nomograms for the modification of wave length and wave height by the linear wave theory and Stokes’ third order theory are presented for a wide range of d/L_s, H_s/L_s and U/C. These nomograms provide the design engineer with a practical guide for estimating wave lengths and heights affected by currents.

The paper gives details of a study to predict numerically the background temperature rise in Poole Harbour and Holes Bay, in Dorset, England, caused by the siting of either a 700MW or a 350MW power station. The numerical model is based on the numerical integration of the flow equations which simulate the water movements of tides and upon numerical integration of the advective-diffusion equation representing the movement of heat.

The results of the study predominantly suggest that the temperature field is proportional to the station capacity, with the maximum and mean temperatures across the basin for the 700MW station capacity being almost exactly double the corresponding values predicted for the 350MW station capacity. Observations from the velocity fields and tidal prism ratios both tend to suggest that Poole Harbour and Holes Bay have poor flushing characteristics, which confirm the relatively high temperature predictions in that only a relatively small volumetric percentage of the heated water is flushed out of the basin during each flood tide.

A three month time series of wave height measurements taken every six hours for twenty minutes duration at two sites (one in 24 m deep water and the other in 15 m deep water) off the South African coast have been used to examine wave energy dissipation. The rocky nature of the bottom has ruled out the possibility of dissipation due to sediment motion and percolation and only dissipation due to bottom friction has been considered. The theoretical work of Hasselmann and Collins 1968 has provided the basic technique. Average friction factors obtained have been in the range 0.17 to 0.50 but have very large scatter. Such large friction factors were not considered possible until recently (Grant and Madsen 1982) when it has been shown that if the hydraulic roughness is of the same order as the bottom orbital excursion, the friction factor tends to a value of 0.23. An attempt will be made in future work to reduce the scatter of the results by careful selection from a longer subsection of the original time series.

In a field observation, water surface fluctuations were measured at many points on line from the shoreline to just outside the surf zone. The data were analyzed by an individual wave method, where the concept of primary individual wave is introduced in order to investigate irregular wave deformation. Primary individual waves are defined by applying the zero-down crossing method with a suitable band width at the zero level to the high-pass filtered water surface fluctuation. It is shown that a wave thus defined behaves like a regular wave with a fixed period in the nearshore zone. A deterministic model based on wave height change of monochromatic waves on non-uniform beaches is then introduced. The model is found to describe the observed deformation process expressed by the primary individual waves.

A nonlinear solution of wave profile equation directly derived from stream function is submitted. In stead of expressing by trigonometric series to approach the real solution, implicit function is adopted. In the era of electronic computer, Such an expression wi11 be convenient for practical utilization.

Equations in either deep water or in finite water depth are worked out. They are proved more reasonable in graphical shape of wave profile and consistant in the continuity of wave celerity in various depth in comparison with Stokes theory.

Determination of a design wave height at a coastal structure requires calculation of a shoaling coefficient or determination of the maximum probable breaking wave height at the point of interest. In shallow water over a sloping bottom, low steepness waves are not accurately predicted by linear shoaling coefficients. Empirical breaking indices are inconsistent with both linear and nonlinear wave theories. Nevertheless, the coastal engineer must select a design wave in order to responsibly design the structure. A graphical procedure is presented herein to relate the equivalent deepwater wave to a breaking wave as it transitions into shoaling water. The procedure provides the coastal engineer with a more consistent understanding of the shoaling process. The results furthermore identify regions of relative depth and steepness where discrepancies arise when using linear shoaling coefficients that may significantly alter engineering design and laboratory studies.

The results of experiments are described which show that surface waves may experience a resonant interaction with undulations on the seabed. This interaction manifests itself in a strong reflection of incident wave energy when the wavelength of the bottom undulation is about half that of the surface wave. It is shown that such a mechanism might enable a region of undulating seabed topography (eg sand bars or sandwaves) to extend in an up-wave direction, into a region of otherwise plane bed.

A new method for numerical simulation of tidal currents is presented. Based on a technique involving the splitting of operators, it allows an accurate calculation of momentum advection, and wave propagation for large Courant numbers. It also provides a satisfactory treatment of tidal flat uncovering and flooding, and permits the use of a curvilinear computational grid. The generation method for such grids is presented here, followed by an engineering application on cartesian grid.

The paper reports the current progress in developing a finite element method for the shallow water equations. The main feature of the method is the special care given to the advective and diffusive parts of the equations, so that it can be of interest to use it when dealing with flows strongly influenced by convective and boundary layer effects. The solution procedure has been chosen so as to allow a calculation with a big number of nodes.

Section 3 of the paper outlines the method. In section 4, is detailed the procedure for the advective terms, involving the determination of the characteristic curves. Section 5 is devoted to the diffusion and propagation terms. Finally numerical results are presented in section 6.

This report presents a three-dimensional numerical model, which calculates by a finite-difference method the vertical profile of horizontal velocities. The unsteady three-dimensional Navier-Stok.es equations with a free surface are governing this flow. We assume a hydrostatic pressure, and simulate the turbulent effects by the Prandtl's mixing-length hypothesis. The model is validated by experiments carried out in a laboratory flume with a prismatic channel inclined 45° over the flow. Then, the model is applied successfully in Gironde estuary and coastal areas in France for the computation of tidal and wind generated currents.

The Lagrangian long wave equations and the Lagrangian expressions for stress and strain are derived. Retaining the dominant terms the long wave equations are solved using an explicit finite difference method.

Using the numerical solution, particle paths are computed for the tidal motion in a basin connected to the ocean by a single inlet. At the open boundary particle displacements are described. Computations are carried out with and without the Coriolis force and for linear and nonlinear bottom friction.

The present paper deals with a new non-linear technique for generation of violent breaking freak waves (plunging breakers) at specified positions and times in wave basins. First, results concerning generation of non-linear wave trains, Stokes-waves and wave solitons in deep water are given. Then the technique for generation of non-linear wave transients are given with specified non-linear dispersion properties. Finally, the new techniques are used to obtain collisions between non-linear solitons coming both from the same direction (2-dimensional case), and from different directions (3-dimensional case) leading to generation of steep and violent plunging breakers.

The paper describes the experimental procedure used to produce a computor contour plot of the wave height distributions and wave directions in a model basin, using photogrammetric techniques. Only monochromatic waves are analysed. A technique is outlined to simulate and measure waves entering a basin of infinite extent, in other words to photograph the penetration of a wave train into a harbour basin before the pattern has been contaminated by reflections. Proof is offered that this infinite basin technique is a valid representation of the steady state situation of a continuous wave train entering an infinite basin.

An analytical model of an "episodic" wave as a function of time is given. The linear dispersion theory is applied to the transient to hindcast the wave train at a wave board which is a specified distance up-wave from a test site. First order wave board theory and linear system theory is used to convert the wave train at the wave board into a control signal. Conversion of the time discrete control signal into a smooth analog voltage signal by means of a real time digital program is described. Examples of numerical simulations and physical measurements are given.

The concept of wave asymmetry is reviewed and a prototype wave record is analysed. A three stage non-linear transformation, together with a Fourier transform substitution technique, is described. The method is tested by numerical simulation using statistical analysis procedures. Physical realizations are compared graphically.

A three-dimensional non-linear hydrodynamic model is developed for the determination of the mesoscale (tides, storm surges, ...) and macroscale (residuals, gyre, ...) circulations in the North Sea. The model consists of a hierarchy of submodels of different grid sizes, with interactive coupling at boundaries.

The model is used to study the impact of coastal engineering projects along the Belgian coast and in the Scheldt estuary.

The 8. 5 magnitude Alaskan earthquake of March 27, 1964,generated a great tsunami in the Pacific Ocean. This paper presents results of energy spectrum analyses conducted on a sampling of 24 of the marigrams from some 105 tide-stations which recorded the waves. In the numerical digital procedure used, two important properties of the evolving energy spectra are invoked; the frequency resolution and statistical confidence. High confidence is obtainable only at the expense of poor resolution, and vice versa. Excessive resolution may give rise to physically unreal spectral peaks, just as excessive restrictions for confidence may cause blunting of the spectrum and possible loss of physically real spectral components. A trial-anderror compromise between these properties was sought by testing the effects on the energy spectra, for Hilo, Hawaii, of the parameters that control the digital process. An optimum selection of spectrum parameters was finally made so that the primary and secondary wave forms of components, identified by the analysis for the marigram of San Francisco, Calif., most closely agree with the corresponding wave forms obtained by use of Chrystal's (1904) graphical method of residvation analysis. Using the parameters so determined, and by application of band-pass filters in the digital analysis, the most prominent peaks of spectral energy in the marigrams of the 24 stations are isolated and their wave-forms computer-plot - ed. In all cases a recognizable long-period wave component is found having a frequency between 0.50 and 0.65 cy/hr (average period 1.73 hrs). General similarity of these primary wave forms as to period and shape of beats suggests their common origin at the source of the earthquake disturbance. The secondary wave forms of higher frequency (periods generally less than 40 mins) are ascribed to local free oscillations forced by the primary wave trains at the receiving stations. In some, but not all of the marigrams there is evidence of wave systems averaging 3.2 hrs in period. These are thought to be an independent wave-train originating at source.

This paper describes results of field measurements on wave run-up caused by storm surge waves. The measurements have been done with newly developed run-up probes at two locations at the German Bight with different dyke profiles. It was found from the results that the wave runup, measured under real sea state conditions, have greater values than predicted by commonly used formulae. Furthermore the wave climate and the breaker type seem to have an influence on the magnitudes of wave run-up.

When a small amplitude wave climbing along an arbitrary sloping beach from deep water toward the shore, the variation of characteristics in the process of wave motion has been described in this paper. From the results of theoretical derivation, it is found out that the variation of water surface and amplitude are function of beach slopei(α) and dimensionless distance (kx) from the shore. And under the condition of the beach slope is α = 0 and α = ∞ that the solution will become a progressive wave and a standing wave respectively.

For a given site, deterministic models may be applied to predict the tide level with a rather good accuracy. However, a difference is observed between the observed and predicted tide level under storm condition. This difference is called storm surge. Two different storm surges prediction models are presented for the site of Le Havre ; an autoregressive model ; a model using wind and pressure local data.

The autoregressive model can be used for a prediction 5 hours in advance. The availability of accurate wind and pressure predictions by the Meteorological Service of Le Havre within 36 hours in advance makes the use of the second model of great interest because it provides the possibility of predictions within 39 hours in advance.

N.S.L. program is a finite-difference code for twodimensionnal flows with a free surface in a vertical plane. Basic equations are Navier-Stokes Equations with a simple simulation of turbulent effects by an eddy viscosity coefficient related to the mixing length and the mean velocity gradient. Theses equations are solved in a variable domain in time. The main features of the numerical method are presented. Some comparisons with theoretical solutions give a good validation of the code both in linear and non linear cases. Other examples of application are given.

This paper relates three applications of a numerical model of storm waves in shallow waters developed in LNH. The equations are recalled at first and then the applications performed are presented.

The numerical model has been used in the case of the port of Fecamp, on the English Channel coast, on which the results of a scale model were available. The computed results compare well with the scale model measurements.

The second case is the s imulation of a t sumami induced by a submarine landslide which appeared in 1979 near Nice ; the mode 1 has permitted the simulation of the rising of the wave.

The last applications consisted in simulating breaking waves by introducing a dispersion term in the equations. This simulation has been tested with a one-dimensional model at first. The results show that the numerical model reproduces the elevation of the mean sea surface due to the loss of energy in breakings. Then the longshore current induced by breaking waves coming obliquely over a rectilinear sloping shore has been reproduced with a two dimensional model.

The results show that the model is able to compute with a good accuracy re fraction, diffraction and reflection, and that it appears to be very interesting for longshore currents simulation.

In this paper , the authors paid attention to the non- linear interaction of two free wave trains crossing in intermediate water depth. From theoretical approaches , the velocity potential and water surface elevation have been expanded to the second-order by perturbation method. And wave height , velocity of fluid particles , pressure distribution , wave thrust and energy density of the common wave are also investigated. In order to verify the theoretical results , elaborated and numerous experiments have been performed. Some of the remarkable coincidence are obtained and the conclusions have been presented in this paper.

A comparison is made between two reference frames that can each be used to define "still water" for finite amplitude waves on water of finite depth. The reference frame characterized by zero mass flux due to the waves is used to find some exact relations between the wave integral properties. The averaged Lagranian (wave action) approach and the energy/momentum approach to the interaction of finite amplitude waves with slowly-varying currents are also derived in this reference frame. Results in many cases are simpler than those in the more commonly chosen reference frame characterized by zero mean horizontal velocity under the waves.

An application of the integral properties is made to Vocoidal wave theory, which is defined in the zero mass flux frame. It is shown that the rotation present in the orbital velocity field of Vocoidal waves is not always negligible.

A field study was conducted at Myrtle Beach, South Carolina (U.S.A.), to determine the response of natural beaches to artificial manipulation by sand scraping. Between March 1981 and May 1982, a total of 100,000 m³ of sand was shifted from the lower beach to the backbeach on three occasions over a 14-km length of shoreline. Fiftyfour profile stations were surveyed to the -1.0 meter contour as many as ten times during the study to determine the effect of scraping and fill along a stable-to-slightly erosional beach. The purpose of the scraping and fill was to provide temporary erosion relief, protect existing dunes and structures, and provide a wider recreational beach at high tide. It was found that scraping and fill had little adverse effect on the beach cycle in the northern zone of the project area, which is fronted by a natural dune system. Fill placed at a gentle slope along the seaward margin of the dunes remained in place for up to ten months before eroding to the prefill surface. In contrast, similar quantities of fill along shore protection structures in the southern zone eroded in several weeks to four months. The study found significant differences in the response of armored versus unarmored shorelines with higher erosion rates and slower recovery of the beach at armored stations.

The mechanism of onshore-offshore sediment transport and the process of beach profile evolution were studied through field investigations performed at Oarai Beach, Japan. The principal data set consists of twenty-four profile surveys taken at hourly intervals on each of two parallel lines spaced 10 m apart and of length 150 m. The lines extended from the backshore of the beach to a point on the sea bottom at a depth where no sand movement occurred. Simultaneous measurements of the waves and wave-induced currents were also made; the essential requirement of two-dimensionality was found to hold during the experiment. The profiles were used to calculate the net sediment transport rate on-offshore, and for an empirical eigenfunction analysis. A negative correlation was found between the transport direction (and resultant bottom change) and the mean sea level change. The sediment transport rate in the foreshore region was found to be proportional to the wave power, whereas outside the surf zone and near the breaker position it was proportional to the tractive force.

Aerial photograph and field studies in the southeastern Mediterranean, involving bathymetric mapping, and concurrent and antecedent wave measurements, have been used to delineate the sequential development of crescentic bars and associated dynamics. The bar sequence includes multiple parallel or wavy bars, ridge and runnels, oblique/transverse bars, single crescentic and double crescentic bars, and occurs during a calming down of wave activity from 2.5 to 0.5 m waves. The concomitant wave data, including wave directions, energy spectrum, significant wave height, and length of the calm period, showed strong correlation with the bar stages.

An increase in total bar occurrence during summer is related to a major wave energy decrease in the spring, when significant wave heights (H_s) ≤ 1 m sharply increase to 70-85% in April-May. Inner single crescentic and initial double-crescentic bars are largely restricted to the calmest wave months of May/April to October/November, which reflects their sensitivity to wave energy. The aseasonal occurrence is best shown by the mature double crescentic type, which apparently is the final stage in the crescentic bar development sequence.

Two bar developmental sequences were delineated: one shore-normal and the other initially oblique, but gradually rotating to shore-normal in the mature stage. Out of phase relationships between inner and outer bar systems resulted from the lag in response of the outer bars behind changes in wave direction. Among the inner crescentic bars and shore rhythms, phase-correlation was the rule.

Crescentic bars are well developed on this coast because of the dissipative conditions and the distinct wave climate. High waves in the winter remove the existing bars, and extended calms allow the full development of the crescentic bar sequence. Similar bar types occur on different coasts in different sequences and in different proportions of time. Thus, it is suggested that these differences are attributable to global differences in the occurrences of threshold wave height conditions.

The Nearshore Sediment Transport Study (NSTS), sponsored by the National Sea Grant Office included a field component to quantify the total longshore sediment transport relationship. This component was conducted at Santa Barbara, California and encompassed a period of eighteen months during which ten surveys were conducted. To date, eight of these surveys have been analyzed, yielding seven intersurvey periods. A total of 288,600 m³ of net sediment transport was documented by these eight surveys. The wave characteristics are based on one of two S_xy gages located in a water depth of 7 m. The most widely used correlation constant, K, in the relationship I_ℓ = KP_ℓs is 0.77. The values found from the data were 0.93 and 1.23 for linear and log best-fit values, respectively. The corresponding values of K_* relating I_ℓ and S_xy are 2.60 and 2.63 m/s, respectively.

Eight fluorescent sand tracer experiments were performed in energetic surf zones on natural beaches and on beaches near structures to measure the short-term longshore sand transport rate. Tracer of up to four distinct colors was injected on a line crossing the surf zone to investigate the on-offshore distributions of the longshore sand adveetion velocity and transport rate. The tracer advection velocity, v , and the depth of mixing into the bed, b, were determined from large numbers of cores taken in situ throughout the sampling area. The sand advection velocity and mixing depth were not constant across the surf zone, but usually exhibited a maximum either toward the shoreline or toward the breaker line, or in both regions. The local breaking wave height, H_b, and horizontal current velocity in the surf zone (yielding an average longshore current velocity $\overline{\text{V}}$ ) were also measured. The data were interpreted with simple dimensional arguments to give the following results: b₂ = 0.027 H_b, v_a = 0.014 $\overline{\text{V}}$, and the volumetric transport rate Q = 0.024 ${\text{H}}_{\text{b}}^2\overline{\text{V}}$. Agreement was also found between the measured total longshore sand transport rate and a predictive expression due to Bagnold involving the breaking wave power and average longshore current velocity. Although the results appear reasonable and consistent, a problem remains concerning the apparent decrease in tracer advection speed alongshore recorded in most experiments at the longer sampling times.

The coasts of Campello (fig. 1), just at the North of Alicante, have undergone important erosions in the last few years. A study has been developped for over one year to separate seasonal and permanent variations of the shoreline. The geomorphology and the recent generation of these coasts have "been established and the study and wind wave regimes have heen defined to evaluate the littoral transports, particularly the longshore ones. The seasonal transverse 1 profiling nay show the seasonal changes in the littoral zone and the sieve and mineralogical analysis of the different samples of the beaches have completed the data for the discussion.

Taking into account other studies of the authors in other parts of the East and Southeast coasts of Spain, an hypothesis on the littoral processes has been established, separating the seasonal and the permanent and degradative consequences. Some recommendations ti keep at least the present situation of the beaches and to protect the other stretches of the shore are presented as final conclusions.

The coastal morphology of Vrachati, which lays on the southeastern coast of the Corinthiakos Gulf, has considerably changed since the construction of two successive coastal structures. The present paper describes these changes and attempts to estimate the rate of longshore drifting sediment load. The description of the changes and the estimation of the rate are based on observations of aerial photographs of the area and on extensive bottom sediment sampling.

Moriches Inlet is located on the south shore of Long Island, New York, approximately 45 miles (72 kilometers) west of Montauk Point and 80 miles (130 kilometers) east of New York City. The inlet forms the primary outlet through the barrier island between Moriches Bay and the Atlantic Ocean. The inlet is protected and stabilized by two stone jetties approximately 800 feet (245 meters) apart.

During January, 1980, a severe northeast storm resulted in the breaching of the barrier island immediately to the east of the existing Moriches Inlet. By the fall of 1980, the breach had expanded to nominally 2900 feet (885 meters) in width with a maximum depth of around 10 feet (3 meters) and the U.S. Army Corps of Engineers was requested to affect its closure. The method adopted by the Corps consisted of the placement of beach fill in the opening to develop a cross-section with a center line elevation of +13.25 feet (4 meters) MLW and side slopes of 1V:25H. Initiated in October 1980, the closure operation was succesfully completed in February, 1981.

The formation of a significant breach immediately adjacent to the existing inlet and the artificial closure of the opening afforded a unique opportunity to study the dynamics of a tidal inlet under the influence of relatively rapid changes in tidal prism and cross sectional area. The purpose of this paper is to present the results of a field measurement program and subsequent analyses of the dynamics of the inlet/breach system. The analyses were based on data obtained before, during and after closure of the breach.

The lagoon of Venice (N. Adriatic) is the most important in Italy (Fig. 1). Between 1840 and 1933 the tifial entrances facing the lagoon were controlled by six large jetties, after the construction of which more evident erosion phenomena appeared in some tourist resort beaches near the inlets. Detailed historical, movable bed models, and field studies have recently been conducted with the aim of defining a proper defence system for these beaches. The results of these studies are reported here.

Inlet sediment bypassing, through the previously recognized mechanisms of stable inlet processes and ebb-tidal delta breaching, has been documented at six mixed energy (tide-dominated) coasts around the world including the coasts of: central South Carolina, Virginia, southern New Jersey, New England, the East Friesian Islands, and the Copper River Delta in Alaska. Regardless of the mechanism, the end product of the bypassing process is the formation of a large bar complex that migrates onshore and attaches to the downdrift inlet shoreline. Thus sediment bypassing is a discontinuous process at mixed energy tidal inlets.

The morphology of the bar complexes is highly variable with widths ranging from 40-300m and lengths from 300 to over 1500m. Generally, the size of the bar complexes increases as inlet size increases and as the rate of longshore sediment transport increases. The frequency of bar welding events at mixed energy inlets varies from 3-7 years. The location where the bars attach to the downdrift beach and length of shoreline that is affected by the bar welding process is dependent on inlet size, orientation of the main ebb channel and wave versus tide dominance of the shoreline.

A mobile-bed model study of Pointe Sapin Harbour, in the Gulf of St. Lawrence, resulted in construction of a detached breakwater and sand trap to prevent harbour entrance infilling. The sand trap is filling in at a faster rate than predicted from the model. This is partly due to incomplete modelling of the wave climate; partly to incorrect modelling of distribution of alongshore sand transport; and partly to a complex sand supply at the site. Nevertheless, the sand trap solution is performing well.

An efficient three-dimensional model of coastal currents and sediment transport has been developed. Simulations of tide- and wind-driven currents and sediment transport in the Mississippi Sound are presented. Results of a laboratory study on settling, resuspension, and deposition of sediment are briefly described. Wave effect on sediment resuspension is also addressed.

This paper presents a numerical model computing sea-bed evolution due to longshore current s, which has been developed at the Laboratoire National d'Hydraulique.

This model has been applied to a schematic semi-circular bay. The computation of sand sea-bed modifications reveals two main tendencies ; a marked accretion in the "up-stream" zone of the bay and a marked erosion in the "down-stream" zone of the bay, as far as longshore currents are concerned.

The blown sand transport rate and the vertical and shore-normal distributions of the wind speed were measured simultaneously on a windy beach. The sand transport rate was measured with conventional total quantity-type traps and with a large trap in the form of a trench. The vertical distribution of the wind speed was measured using an ultrasonic anemometer array consisting of six meters. The distribution of wind speed at a height of 1 m in a section normal to the shoreline was measured with five ultrasonic anemometers. A logarithmic law for the vertical distribution of the wind speed was satisfied, and the wind speed in the section normal to the shoreline was almost constant. The Kawamura and Bagnold formulae were found to predict well the sand transport rate. The trench trap and conventional traps gave empirical coefficients of 1.5 and 1.0, respectively, for the sand transport rate averaged over a section normal to the shoreline. The lower value determined with the conventional traps (1.0) is attributed to their inefficiency compared with the trench trap. In order to obtain data at high shear velocities, a wind tunnel experiment was carried out. This experiment showed that both the Kawamura and Bagnold formulae were valid in the range between 60 to 300 cm/s in the wind shear velocity. The empirical coefficient in the laboratory experiments was 1.0: the difference between the field result with the trench trap and the wind tunnel experiment is attributed to the fluctuations in natural wind.

Due to concerns of possible shoaling problems, an extensive field survey program was carried out at the site of the proposed Cerrejon coal port on the Caribbean coast of northeast Colombia. The program yielded considerable data on winds, waves, currents, and sediment factors. Techniques for the primary measurement of sedimentation-related tendency included dredged test pits, scour crosses, and suspended sediment samplers.

The port plan includes dredging a 4.6 kilometer long channel varying from 12 to 21 meters in depth. In order to assess the magnitude of maintenance dredging and related problems, a method was developed for incorporating the sediment response measurements into predictions of the areal and seasonal distributions of bed load and suspended sediment deposition.

Offshore test pits were monitored for rate of filling and character of the material being deposited. Suspended sediment samplers were similarly observed and also provided data on concentration versus depth.

The procedure which was developed for analysis and interpretation of the data included extrapolation of suspended sediment data to the seabed, investigation of correlations between wind activity and deposition rates, application of test pit data to a channel of larger dimensions, and testing of hypotheses regarding transport mechanisms. The procedure concluded that average annual shoaling would be approximately 300,000 m³ and predicted areal and seasonal variation of deposition rates.

A breach was opened in the barrier beach along the south shore of Long Island, New York during a winter storm in January, 1980. Due to public safety and environmental concerns, a contract to close the breach using sand fill was awarded to a local contractor. This paper discusses the evolution of the construction procedures developed to allow the contractor to successfully complete the project on schedule and on budget.

This paper describes the design and construction of a sand breakwater at Saldanha Bay exposed to the severe swell conditions of the south Atlantic.

The design of this spending beach breakwater was aimed to establish a seaward profile with sufficient "playroom" to incorporate profiles resulting from extreme wave conditions.

As far as the construction is concerned the emphasis is put on the sandlosses during execution of the works.

Finally the designed profile is compared with the actual established beach profile.

Carmel Submarine Canyon heads in shallow water near Monastery Beach at the southeast corner of Carmel Bay, California, U.S.A. Very coarse sand, shaped into large oscillation ripples, covers the narrow shelf between the beach and the canyon; when this sand enters the canyon head, it lies at angles as great as the angle of repose. In some areas, these sand slopes show evidence of active grain flows in the form of downslope-coarsening, inversely graded deposits.

The results of a dyed-sand tracer study adjacent to the canyon show that sand moved canyonward during the summer of 1979. Initially the dyed sand, which had been shaped into an oscillation ripple in the center of a 20-m by 60-m grid, moved offshore en ma sse. After a few days, though, the dyed sand dispersed with the center of mass moving canyonward.

As wave-transported sand accumulates along the canyon rim, the upper slopes oversteepen, thereby causing some of the sand to avalanche downslope. Systematic changes in sand levels along three lines of rods over 15 months document preferential deposition of sand along the upper slopes; the greatest change occurred at the top of the lines (12-15 m depth) and the least at the bottom (30-40 m). Greater accretion during the spring months than during the summer months probably reflects the more energetic springtime wave climate.

Between October 1981 and October 1982, 5.7 m³ of sand was deposited per meter alongslope on the middle line, which gives a calculated depositional rate of approximately 5 00 m³/yr in the study area. Although we have monitored this area for over a year, we have not yet documented any large-scale events capable of flushing sand out of the canyon head. The only erosive event we have observed was a small grain flow we generated while digging on the slope.

A field study was conducted on a non-tidal sand beach to measure the relationship between wave energy and the longshore transport of suspended sediment. The range of suspended sediment concentrations was similar to that reported elsewhere. The relation between longshore transport and longshore energy flux derived from this data set is about half that amount predicted by such widely quoted sources as Komar and Inman (1970).

This paper is an analysis of two sets of experimental results on littoral sand transport. A littoral sand transport expression is proposed, relating littoral transport rate to surf similarity parameter and hence to wave energy dissipation rate. The expression indicates that the 'constant' in the CERC formula is dependent on the mobile bed beach slope and on the breaker index. The expression is also compared with some of the few published field measurements.

The main purpose of this paper is to propose an ideal methodology for beach erosion control from the viewpoint of controlling the total rate of longshore sediment transport. For this, a new formulation of the rate of longshore sediment transport is made. The total rate is directly proportional to the longshore component of wave power in field coasts, but not in laboratory ones.

How to control the total rate of longshore sediment transport is considered. There are two ways applicable for practical purposes. The first is to decrease the breaker depth by changing the bottom topography, and the second to decrease the incident angle of breaking waves by changing either the bottom topography or the inclination of shoreline to the incidence of predominant waves. Two typical, but ideal examples are explained for beach erosion control by this methodology.

A series of laboratory experiments in an oscillatory tank was carried out to investigate the sheet flow of sediment. Objectives of the study were to determine the criterion for inception of sheet flow, and to evaluate the sediment transport rate under the sheet flow condition. In order to proceed with the investigation, it was necessary to develop devices appropriate for tracing the sediment particle movement, and for measuring the extremely dense sediment concentration in the moving layer of sheet flow. The chief results are: 1) the criteria for the inception of sheet flow given by Manohar C1955) and by Komar and Miller (1974) are both applicable to materials composed of spheroidal particles, and 2) the average rate of sediment transport for sheet flow is well described by an empirical relationship given by Madsen and Grant (1976) for the bed load transport rate on a plane bed in oscillatory flow.

A comparison is made between the measured infilling of two test pits off the coastline of San Francisco and predictions using a coastal bedload transport model. The model, based on the work of Madsen and Grant (1967), relates the bedload transport to the bottom shear stress through an empirical relationship based on laboratory experiments. The bottom shear stress is estimated from the bottom currents created by waves and low frequency currents. The model applies beyond the breaker zone in contrast to littoral transport.

The test pits, dredged as part of the Southwest Ocean Outfall Project for San Francisco, were located 1.6 km (1 mi) and 3.2 km (2 mi) offshore in 13' m (42 ft) and 16 m (53 ft) of water. The depth of the pits relative to the natural seabed was about 8.4 m (25 ft). The comparison was conducted for a period up to 2 months in the fall of 1978.

The paper discussed the quality and scope of available data required as input to the model and shows how regional wave data were trans formed to augment local measurements. Uncertainties in model results stemming from limitations in the input data are presented. With suitable adjustment of the scale of the gravitational term in the expression for the Shields parameter, overall agreement between computed and measured bedload was accomplished within the limits of accuracy of the bathymetric surveys. A sensitivity analysis of selected input conditions and coefficients was also conducted.

In order to investigate the response of beach profiles to incident waves, computations by the empirical eigenfunction analysis proposed by Winant et al. are performed. The analysis of the data obtained at Ajigaura Beach over three years from 1976 to 1979 indicates that beach profile changes due to longshore and onshore-offshore sediment transport are separable by the empirical eigenfunction method. The beach profile changes due to longshore sediment transport has a time lag of 12 weeks with respect to the change of wave direction at Ajigaura Beach. It was found theoretically that this time lag was due to the sand waves propagating in the longshore direction. Regarding as onshore-offshore sand transport, the second eigenfunction is associated with the beach changes due to onshore-offshore sand transport caused by the change of wave height.

Two-dimensional beach profile changes were investigated with a newly constructed prototype-scale wave flume. The flume is 205 m long, 3.4 m wide and 6 m deep. Sand of two grain sizes was used in the experiments. Analysis of the results was made through use of the parameter C, introduced by Sunamura and Horikawa (1974) to classify beaches as either erosional and accretionary. Beach profile changes obtained in the flume were similar to those in the prototype (field). Net sand transport rate distributions were classified into five types, two of which do not seem to have been observed in laboratory (smallscale) experiments. A simple model describing the five types was developed for evaluating two-dimensional beach profile changes.

In order to analyze beach profile changes due to longshore and onshore-offshore sand transport, here is proposed a new model named the "empirical predictive model of beach profile change", which is an application of the empirical eigenfunction method. The analysis of the profile data obtained at the Misawa fishery port in Ogawarako Coast over five years from 1973 to 1977 indicates that profile changes due to longshore transport and to onshore-offshore transport can be separated. The model is shown to be effective in the analysis of profile changes near coastal structures.

An energetics-based surfzone sediment transport model was evaluated in its ability to predict on-offshore sediment movements using current meter and beach volume measurements from the Nearshore Sediment Transport Study at Torrey Pines Beach, California. The magnitude of pertinent wave velocity moments were also evaluated from the same data set. These moments were found to be adequately represented by linear functions of the significant wave height. Because of the apparent noise in the beach volume measurements and the limited duration of the current meter records, the results of the model evaluation were inconclusive. However, a simplified version of the model, when coupled with estimated wave velocity moments, was found to mimic observed on-offshore sediment movements as a function of significant wave height.

Laboratory and field investigations were performed in order to formulate a predictive model of two-dimensional beach profile change. The observed transport was classified into six types, and transport formulas were deduced for each type based on a microscale description of sediment movement caused by wave action. A numerical model of two-dimensional beach transformation was then developed. Beach profile changes calculated with the model were then compared with the laboratory results. The model was found to give reasonable results except in the vicinity of the wave plunging point. The sediment transport calculation is based on a sinusoidal velocity profile. The model appeares to give good results as long as the wave motion can be reasonably approximated by linear wave theory.

The passage of Hurricane Allen over Padre Island in August 1980 presented an excellent opportunity to study the effects and controls of coastal morphology on hurricane deposits. In the Caribbean Sea, Hurricane Allen achieved a central pressure of 899 mb, making it the second strongest Atlantic hurricane ever recorded. Once in the Gulf of Mexico, the storm approached the Texas coast from the east-southeast, building a significant storm surge. Near Brownsville, the storm stalled, spending much of its energy offshore before making landfall early on August 10th near Brazos Santiago Pass at the southern tip of Padre Island.

Surge gauges show that peak recorded storm tides of about 3 m occurred at Port Mansfield, some 35 km north of landfall. Analysis of tide data indicates a time lag of some 14 hours on the rising storm tide between the Gulf and south Laguna Madre. This is due both to the limited tidal exchange across Padre Island and to set-down in the lagoon due to southward-directed cyclonic winds. By contrast, measurements taken in Corpus Christi Bay, some 180 km north of landfall, show that at that point Gulf and bay tides were in phase. The cross-barrier water level differential at South Padre Island (up to 1.5 m) greatly facilitated hurricane breaching of the island.

Oblique and vertical aerial photography show that Padre Island was breached in many places, with about 40 major hurricane channels still open several days after the storm. Surge heights were sufficient to inundate all of South Padre Island except for isolated "dune islands" resulting in broad and often coalescing washover deposits. The more continuous dune ridge on North Padre Island resulted in smaller, discrete washovers. Intensity, distribution, and morphology of washovers are functions of storm tide elevation, its phase relationships, island topography, and lagoonal water depth. The relationship is complex, yet precise enough to permit prediction of the island's response to the impact of a given storm.

Captain Sam's Inlet, a shallow unstable inlet, periodically migrates up to three kilometers (km) alongshore over a 30- to 40-year period. As the inlet migrates, sediment accumulates at the seaward terminus of the tidal inlet to form a wave-modified ebb-tidal delta that trends downdrift. Continued sediment accumulation on the tidal-delta shoal and subsequent lengthening of the tidal channel in a downdrift direction results in an unstable channel configuration. The ebb delta is eventually breached on its updrift side, releasing a sediment package for inlet bypassing.

The sediment-bypassing process was initiated at Captain Sam's Inlet after the final landfall of Hurricane DAVID in September 1979. Initially, the newly formed updrift hurricane channel scoured 1.1 m, migrated updrift, and became the predominant tidal channel at the inlet with eventual abandonment of the prehurricane, main ebb channel. These two channels outlined a portion of the ebb-tidal delta that was freed for bypassing. The initial sediment volume contained in the bypassing shoal, above the -0.6 m (-2 ft) MSL contour, was 47,000m³. The sediment volume of the bypassing shoal did not change significantly until final attachment to Seabrook Island. This channel dominance and the wave-induced migration of the bypassing sediment package aided the bypassing of sand at the inlet. Immediately after filling of the prehurricane, main ebb channel on the delta, the downdrift beaches began accreting. The accretion continued throughout the bypassing process. The tidal prism and cross-sectional area reflected little change during the bypass, showing evidence of the system's overall stability.

A channel dredged at the mouth of a tidal inlet is subject to rapid shoaling because of longshore transport, but this shoaling is slower than would be computed from simple trapping of all the moving littoral drift. The reduction in shoaling rate is due to the bypassing of littoral drift which occurs simultaneously with shoaling. This report presents a systematic method for computing the rate of shoaling in channels subject to shoaling with bypassing. The method also permits estimates of the effect of the dredged channel on the downdrift beaches.

Wave protection within a new harbour development has been achieved by shaped dredging in place of long breakwaters. The dredging shape was initially established through wave refraction calculations and then confirmed and optimized in a physical three dimensional model. The cost of additional dredging for the specified shape was A$100,000 compared to estimated breakwater costs of A$l,000,000.

Flume test data is presented for the depth of scour in deep water near a vertical cylinder placed in a uniform sand and a fine calcareous ocean sediment. The ratio of the depth of scour to the cylinder diameter at equilibrium is shown to depend only on the ratio of the shear velocity to the critical shear velocity at which bed motion is initiated.

Protection against scour by placing collars around the cylinder is shown to be of marginal value.

This paper is aimed at the establishment of a generally applicable criterion for the onset of grain motion under the influence of oscillatory flow. Data from previous studies are used in a dimensional analysis and an empirically derived relationship between the dimensionless parameters R* (shear Reynolds number) and D* (dimensionless grain parameter) is proposed as a criterion to be used in coastal engineering problems.

This study forms part of a larger programme by the Sediment Dynamics Division of the National Research Institute for Oceanology in Stellenbosch, RSA, which is aimed at the reevaluation and updating of the input parameters and relationships for the predictive equations for coastal sediment transport.

Surfieial layers of estuarial fine, cohesive sediment beds are deposited from flow and often are in a state of partial consolidation. A series of laboratory investigations were carried out to elucidate the erosional behavior of deposited cohesive sediment beds in flumes using kaolinite. A significant feature of such beds is that they are stratified with respect to the density and the cohesive shear stength. Under a given bed shear stress, erosion occurs at a continuously decreasing rate up to a depth at which the bed shear stress equals the shear strength. This bed shear stress is therefore also equal to the critical shear stress for erosion at that depth. An expression for the rate of erosion relating this rate to the difference between the bed shear stress and the critical shear stress has been obtained. The critical shear stress increases both with depth and with the bed consolidation time. The rate of erosion decreases with increasing consolidation time.

An experimental study is described in which the rate of transport of fine sand over a flat bed was measured as a function of the shear stress exerted on the bed by both laminar and turbulent flows, for a range of water temperatures. The results confirm the great sensitivity of the transport rate to changes in both shear stress and temperature over the test range. The hypothesis is propounded that in certain specified circumstances, the net rate of local bed load transport generated by an arbitrary unsteady flow, can be estimated by integration, using the measured steady laminar flow transport relationships in conjunction with the time history or probability distribution of the fluctuating bed shear stress. The concept has been tested using the measured data with encouraging results. It is suggested that this method of linking the characteristics of the near bed flow field to the local induced rates of sediment transport could be productively applied in a range of studies including the initiation and development of bed sand ripples by the action of various combinations of waves and steady currents.

Following a change in wave direction, the active contours in an idealized pocket beach respond by rotating such that they approach a perpendicular orientation relative to the incoming wave rays. Assuming that cross-shore sediment transport does not contribute to this contour rotation, and that the contours are in the early stages of this equilibration process, the amount of contour rotation can be interpreted as the cross-shore distribution of the longshore sediment transport.

As part of the Nearshore Sediment Transport Study, detailed nearshore profile measurements were conducted at Santa Barbara, California. Twenty-two of these profile lines were located on Leadbetter Beach, which is a quasi-pocket beach. To explore the concept described above, two of the nine intersurvey periods were selected due to their strong indications of wave direction change. Analysis of these data sets yielded two estimates of cross-shore distribution of longshore sediment transport which were compared with those presented by Komar, Fulford and Tsuchiya. Although these three distributions differ significantly, the effect of the tidal variations is to "smear" the differences in the inferred distributions as evident in the contour displacements. It was found that none of the relationships for longshore transport distribution predicted the amount of transport inferred in water depths greater than one meter. It is possible, especially for one of the intersurvey periods that the changes in contour locations were so extreme that substantial crossshore sediment transport was induced and would be interpreted as longshore transport occurring in water depths greater than had actually occurred. The method introduced here should be useful in other field and laboratory programs to investigate the cross-shore distribution of longshore sediment transport.

The accuracy of prediction of longshore sediment transport depends largely on the accuracy with which the wave-driven longshore currents within the breaker zone can be predicted. Longuet-Higgins (1970) developed a formulation for longshore transport which is widely used today. In the present paper the basic theory of Longuet-Higgins is reexamined. The effect of bed roughness on the magnitude of the longshore current is quantified with the aid of over 350 individual data sets and the theory is theoretically extended to include the effect of random waves, in a similar way to Battjes (1974), and higher-order waves. For this latter purpose the Vocoidal water wave theory of Swart (1978) is used. It is shown that the use of Vocoidal theory leads to a velocity distribution which is in closer correspondence to measured data than that predicted by using linear wave theory.

There is a growing need for generalized numerical models for longshore currents and nearshore circulation that solve the complete equations of motion, are flexible in the formulations chosen for various terms, and can be applied to field situations at a reasonable cost. The development and application of one such model is described in this paper. The model was first tested by comparing its results to known analytic solutions and experimental data. There was good agreement. It was next applied to a field situation near Oregon Inlet, North Carolina. The results appeared to be reasonable and the computational costs were modest.

A field experiment to measure the horizontal and vertical structure of shore-parallel, nearshore currents was conducted at Wendake Beach, Georgian Bay, Canada, in 1980. Waves and currents were measured with continuous resistance wire wave staffs and bi-directional, electromagnetic current meters, respectively. Substantial variations from theoretical horizontal velocity profiles were found as an influence of small amplitude nearshore bars. Data smoothing resulted in a Longuet- Higgins type mixing parameter estimate of P=0.1B. Vertical velocity profiles analysis suggests that an estimate of mean, surf zone roughness length is of the order of l×l0^–3m.

This paper consists of three parts. The first part presents a method for analyzing the forces exerted by breaking waves on a cirular pile. The force consists of two components, a slowly varying force and a much larger but very short duration quasi-impact force (probably of the order of 1/100 of a second). The second part is concerned with breaker characteristics, with emphasis being given to the few field data that have been measured. The third part consists of a presentation of some available data on surf zone bottom profile variations with time.

Information on all three of these parts is needed for the proper design of a pile supported structure in the surf zone. If the bottom along the site of a proposed pier is sand, an estimate of the variability with time of the profile must be made. The effect of bottom depth and configuration on the height of waves moving shoreward, and the effect of this, in turn, on the wave loading is important in the calculations of wave-induced moments about the bottom. The ability of the structure to withstand these horizontal loads depends in part upon the depth of penetration of the piles. If the bottom varies with time, then calculations of wave characteristics and wave-induced loads on the piles should be made for appropriate bottom configurations.

This paper describes the design and construction of a fishing harbour on a rocky coastline exposed to the prevailing south westerly swell of the South Atlantic and to severe westerly gales.

Because of economic pressure the first phase of the development was undertaken without adequate knowledge of the wave regime or the topography of the sea-bed in the area and resulted in a virtually unusable harbour.

The second phase of construction was therefore only embarked upon after extensive hydrographic surveys, wave recording and analyses, and probably the most exhaustive series of model tests ever undertaken for such a small project.

These investigations and the good co-operation between research staff, engineers and contractors resulted in the elimination of most of the initial problems and the creation of a functional fishing harbour.

The design scheme of the Zeebrugge Outer Harbour, Belgium, consists mainly of two breakwaters protruding into the sea as far as 1,750 m beyond the existing mole or 3.000 m out from the coastline. The west outer breakwater is 4,450 m long, the east breakwater runs 4,300 m out from the seafront. The east outer harbour will accommodate terminals for liquid bulk products such as LNG. The west outer harbour will provide space to install two harbour bassins to suit general cargo, hazardous cargo, container and ferry traffic.

In the paper emphasis is put on the environmental design conditions (wave height, wave period, water depth), the development of preliminary designs and the final design. Some design features such as dimensions, wave breaking carpet, armour units, workable limits in respect to rock grade, etc ... are discussed.

By developing the design the rubble-mound breakwater has been judged to be the only viable alternative versus the caisson type breakwater, taking into account costs, technical risks, construction problems and flexibility under changing environmental conditions.

The main feature of the breakwaters is that a huge concrete parapet will have been avoided. A so called flat semi-cube armour unit has been developed. The main advantage lies in the substantial economic benefit whilst maintaining the same stability performance as a concrete cube armour unit.

Gansbaai fishing and pleasure craft harbour is situated approximately 160km east of Cape Town. Modifications and extensions were required to the existing breakwater as well as construction of an internal access mound. An investigation was commissioned into alternative construction methods and construction sequences for the modifications and extensions.

The aim of the study forming the subject of the paper was to establish as accurately as possible the optimum sequence of construction for the various phases of the work and to establish the optimum construction methods to be adopted so that any damage to the work was minimised as construction proceeded. By means of a three dimensional hydraulic model, alternative construction methods and sequences were investigated under varying wave heights and sea conditions. Minimum wave heights and the sea direction causing damage or localised displacement of the elements of construction were studied.

A method of calculating nonlinear wave induced forces and moments on piles of variable diameter is presented. The method is based on the Morrison equation and the linear wave theory with correction parameters to account for convective inertial effects in the wave field. These corrections are based on the stream function wave theory by Dean (1974). The method permits one to take into account the added wave force due to marine growth in the intertidal zone or due to a protective jacket, and can also be used to calculate forces on braces and an array of piles.

This paper describes the performance of test panels of riprap on an offshore island in the Wash estuary, UK, for the first $2\raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{$2$}\right.$ years after their construction, by which time all but one had failed. It outlines the methods used in obtaining and analysing data on stone size, wind, tides and waves. The techniques used in surveying the test panels for damage and the reduction of the survey data to yield quantitative estimates of damage are described.

Comparisons are made between the damage to the riprap panels and what might have been estimated using laboratory data. Results do not support any scale effect causing riprap sized on laboratory data to be larger than necessary, and this conclusion is supported by the outcome of model tests carried out retrospectively by the Hydraulics Research Station (now HRS Ltd) at Wallingford, UK.

Slope protection is sensitive not only to wave height and stone size but also to construction methods and, bearing in mind possible departures from the desired specification, a cautious approach to the design of riprap protection is advisable.

Coastal protection devices include, but are not limited to, dune stabilization, seawalls, groynes, detached breakwaters, sand bypassing and beach nourishment. These are the "tools". The "rules" for their use comes from a knowledge of the beach processes which in many instances are site specific. Systems which work satisfactorily at one location may not necessarily work at another. A plan is made for authors to give more information on these processes when describing case histories.

The variability of the physical conditions and the non-linear nature of many of the coastal processes often dictates how coastal protection works will behave; a fact which is not always taken into account by the designer.

For coastal protection devices to be developed to their full potential requires improved instrumentation and data bases and a greater knowledge of the coastal processes than is available at the present time.

Wave diffraction and refraction in the surroundings of a new detached breakwater induce currents strong enough to cause substantial local sediment transport and consequently morphologic changes, the main feature of which are sand spits or tombolos.

As shown by field studies the morphologic changes cause an equilibrium state with some minor fluctuations due to changing wave climate. Model studies have also proved the existance of such an equilibrium state for any given geometry and sea state

The present paper treats mainly this equilibrium state, but also attempts to explain the mechanism of sand transport characterizing the initial and transitional states. According to observations by the authors in small scale models, the transporting mechanism in the transitional state involves sea bottom erosion, especially near the breakwater heads outside the protected area, and transport of sand towards the shore of the sheltered area in the form of small migrating sand bars, which finally join the shore line and widen it. Results of experimental investigations conducted by the authors, for the particular case of high impervious breakwaters attacked by waves of normal incidence, as well as results of field and model studies given by others, are used to define relationships among the factors determining the equilibrium state and to base a new hypothesis regarding the equilibrium state. This hypothesis states that a morphologic and sedimentologic equilibrium is reached behind a detached breakwater, when the shape of the contour lines becomes such, that along the sheltered beach the diffracted waves have components of momentum opposed to the gradients of the mean sea level induced by radiation stress due to non uniform wave heights along the wave fronts approaching from both sides of the breakwater. The significant parameters characterizing the dimensions of the spit or tombolo in the equilibrium state are shown to be the relative length of the breakwater (compared to its distance from the original shore line), the relative distance from the original shore line (compared to the position of the breakers' line) and the relative height of the breakwater crest (above M.S.L, compared to the incident wave height).

This paper summarizes a study recently concluded of a sandy coastal reach, 110 km long, on the Portuguese west coast. The final objective of that study was the design of a sea defense master plan covering all the stretch, to be built at a long term. That coastal reach has been under a strong erosion process for a long time. The history of the problem and its present situation were thoroughly analysed. Different techniques, related to different aspects of the general problem (wave clima te, littoral transport, sand bypassing, long term evolution under accretion or erosion conditions, groin field, spiral beaches, etc.lwere used. In what concerns littoral transport and sand bypassing the improvement of currently available methods was tried. The need of an accurately defined wave climate was felt; some rather disappointing results can surely be related to the lack of accurate wave data.

A non-linear numerical method for calculating wave forces on floating bodies has been developed by Isaacson (1981). The time stepping procedure is programmed for a computer solution, and an incident wave train is time stepped past a fixed two-dimensional rectangular breakwater. The influence of various input parameters on the'accuracy of results is investigated, and optimal values of the parameters are determined. The optimal numerical parameters are used to generate force and transmission coefficient results, which are compared to the results of other published studies. The method is shown to compare favorably with other results, with the non-linear nature of the method being clearly demonstrated by the different force curves produced by varying the wave height.

Model tests were done with dolosse having the same shape and volume, but with different block densities, to determine the effect of block density on stability and to check whether the theoretical third-power relationship between block volume and relative block density is valid.

From these tests it can be concluded that the higher the density of the block material the greater the stability becomes. Although the individual scatter is appreciable, the average results indicate that the stability of dolosse is inversely proportional to a power of about 2,3 of the relative density, which is significantly less than the theoretical power of 3.

Model tests were also carried out on dolosse having the same mass and volume but with different waist-to-height ratios to determine the effect of waist-to-height ratio on the stability of a dolos armour.

From the results of these tests it can be concluded that the stability of the armour decreases with increasing waist thickness, particularly for relatively high waist ratios.

Physical modelling is a technique which is commonly used in the design procedure of rubble-mound breakwaters. For reliable results it is necessary that the model tests accurately represent the prototype situation. In this paper, two significant improvements in modelling of breakwaters are presented. They are the generation of realistic sea states at the test site, and the simulation of the breakage of the armour units by using units which have mechanical properties properly scaled from the prototype units. These techniques have been used to study the recent breakwater failure at Riviere-au-Renard, Canada.

The Dolos, a type of armor unit, has been used widely for breakwater and shore protection works in the world. However, it has been reported that the armor layers of several breakwaters have been damaged by wave action, and it is probable that the breakage of Dolos has been the cause of that failure.

In this paper, static and dynamic tests using Dolosse units are described. 4t reinforced units and 4t, 0.4t and 0.04t unreinforced units were used.

In these tests, concrete surface and reinforcing bar stress of Dolos, and impact load were measured.

The results of these tests were as follows:

(1) From the both tests i.e. the static load test and the drop test, stress was greatest in the corner between the chamfer and the stem. Cracks occurred at this point.

(2) In the static load test, comparing the results of both units with reinforced and unreinforced chamfer, it became clear that the reinforcement of the chamfer could reduce the magnitude of the stress concentration.

(3) In the drop test, the drop height which made cracks was almost constantly independent of the weights of the units. And it could be considered that there was little influence of increasing the concrete strength as to the breakage of Dolos.

A numerical model has been developed to simulate breakwater response to wave impacts with special reference to armor unit behavior and breakwater stability. The model uses a finite difference hydrodynamic code to follow the wave impacts and determine wave forces upon the breakwater components. A discrete element code models the breakwater response and motions. The model rationale and numerical basis are followed by three examples used in this, the concept validation, stage of model development.

This paper proposes to improve the safety of breakwaters by two important changes in the philosophy of design. When hydraulic model testing is used as a design tool the authors propose to reduce the specific gravity of the model breakwater to introduce a factor of safety in the prototype. They also recommend that the concept of testing for stability with the once in 50 year or once in 100 year wave should be replaced by a more rigorous statistical analysis to determine a design wave which has a probability of exceedence of no more than 5% in the lifetime of the structure.

The objective of the research presented is to furnish design information for stone and dolos armor on non-overtopping breakwater trunks that are subjected to severe depth-limited breaking waves. Since it would be a mammoth task to comprehensively investigate all the different types of existing armor, this particular research effort concentrated on stone, which is a natural and economical protection when it is of sufficient size and quality to meet design constraints, and on the dolos, which according to nonbreaking wave data is the best hydraulically stable concrete armor unit.

Irregular wave tests have been conducted to research into the stability characteristics of armor units for a rubble foundation of a composite breakwater.

A cover layer to protect the rubble foundation from erosion had two layers of tetrapods. Waves higher than H_1/10 (the average height of the highest 10% of all waves) caused damage to armor units at the point of critical stability. This suggested that wave height changes in the surf zone should be taken into consideration for design purpose.

The stability of armor units for the rubble mound foundations of composite breakwaters has been investigated under the action of irregular waves. The tests establish that irregular waves are more destructive than regular waves, when the height of regular waves is set equal to the significant wave height.

The stability number, defined by Hudson, for quarry stones and concrete blocks with simple shapes is formulated on the basis of irregular wave tests. The stability number is expressed by two parameters of h′/H_1/3 and κ, where h′ is the crest depth of the rubble mound foundation, H_1/3 is the design significant wave height, and κ is a parameter for the combined effects of the relative water depth and the relative berm width of the rubble mound foundation to the wavelength.

The design mass of armor units can be calculated by the stability equation with the stability number. The application of the proposed method to the results of the irregular wave tests demonstrates that the damage percent for the quarry stones is at most 3.5% at the design condition and the damage progresses rather gradually for the action of higher waves. On the other hand, the damage of the concrete blocks almost jumps beyond the design wave height. In particular, the drastic damage is often caused in the case of high rubble mound foundations. The proposed method is confirmed, however, to be applicable for the ordinary low mound foundations with a sufficient safety.

A method of surveying the above-water slope of breakwaters by means of photography has been developed at the National Research Institute for Oceanology. The method is simple and inexpensive and yet capable of detecting movement or displacement of a single armour unit on a breakwater.

This poster paper describes the method and presents some examples of its application in the field.

Some early results are reported from an investigation of the forces exerted on horizontal cylinders by waves and currents. It appears that the values of the inertia coefficients decrease as U/u_o increases, and the drag coefficients decrease also, for values of U/u_o up to 0.7. Comparisons with measured data show that linear theory and the stream function theory satisfactorily describe the wave motion for the conditions investigated, and that velocity superposition can be used with either of these theories to describe conditions involving waves plus currents, with reasonable accuracy.

This paper describes the investigation of means of reducing wave action reaching the shoreward end of a power station cooling water outfall channel without resulting in significant head loss to the outflowing water. A variety of conceptual methods of reducing wave action in the outfall channel was examined. A physical model of the outfall was constructed. It was found that a rubble mound wave energy dissipator located in the outfall channel dramatically reduced wave action at the discharge seal pit.

In the design and construction of waterfront bulkhead systems, it is essential to consider the coastal effects of tides, waves, boat wakes, currents, bottom sediment movement and bottom scour. Many improperly designed bulkhead systems experience severe loss of backfill and toe materials with the bulkhead eventually failing if it is not corrected in time. Inadequate drainage, joint connections, and/or inadequate toe protection are typically the causes of failure.

This paper describes an investigation of a bulkhead system supporting a large waterfront development in southern California which was experiencing widespread sinkhole development in the bulkhead's backfill and was on the verge of losing toe material. The objective of this investigation was to determine the extent and cause of ongoing subsurface erosion, to evaluate its effect on the bulkhead stability, and to recommend and design mitigative measures. The cause of the erosion was determined to be piping of fine grained soils due to inadequate backfill drainage. A remedial drainage scheme was designed and field-tested, and several structural repair schemes were suggested for portions of the bulkhead where accumulated damage affected the integrity of the structure.

A survey of existing Corps breakwaters with concrete armor units and hydraulic model tests of rubble-mound breakwater trunk sections protected with dolosse were conducted to determine the effects of broken armor units on breakwater stability and to establish some criteria by which decisions can be made as to when maintenance and rehabilitation work should be initiated on damaged concrete armor unit cover layers. The survey revealed that where good engineering designs were used, prototype breakage has been random and has not exceeded about 3 percent of the total number of units placed. The model tests, conducted with both breaking and nonbreaking waves with no overtopping, revealed the percent breakage can be quite a bit higher than 3 percent before the overall functional integrity of dolos cover layers is affected.

The rising sea level issue affecting U.S. coastal policy is reviewed, and the U. S. Army Corps of Engineers coastal role is summarized. Three case studies are provided to demonstrate the successful application of several structural devices available to solve different coastal problems. The coastal engineer, and other scientists, should contribute significantly to the information upon which society bases its decisions on use of the shoreline.

In current usage, environmental aspects of coastal engineering design include aspects of ecology and aesthetics, as well as environment. In practice, the aspect of environment is a limited one, considering man's surroundings, with the works of man left out. The increased consideration of environmental aspects over the past 15 years has brought real benefits to the coastal engineering profession, as well as obvious problems. One problem is a mythology of coastal processes that has become widely accepted. Priorities in coastal engineering design remain a structure that will last a useful lifetime and perform its intended function without creating new problems. After satisfying these fundamental requirements, the structure should minimize ecological change, and fit pleasingly in its setting.

This poster paper describes a hydrological/hydraulic classification of the estuaries of Natal, based upon aerial photography. The modus operandi is outlined and although the study is in its early stages, some key factors (natural and man-made) are indicated together with the method by which they are quantified.

The authors describe the introduction and development of commercial yacht harbours in the United Kingdom. In the case of the south and west coasts where there is a high mean spring tidal range of up to 8 metres, this has resulted in the provision of enclosed basins, provided with either some form of gate or sill to retain the water. These are described, with typical examples. Recent developments in the design of sector gates are reviewed.

The paper presents the different stages of the environmental studies performed at the Laboratoire National d'Hydraulique concerning some of the French coastal nuclear power plants using a once – through cooling system, and the different means used, in function of the proper caracteri sties of the site, to come to an optimum inlet-outlet configuration leading to a minimum thermal impact and to a minimum recirculation of the heated waters at the intake. The paper particularly describes the original studies performed for the Flamanvilie power plant where a special computer controlled tank has been used.

Extensive field investigations were carried out over a two-year period on the process of sand deposition in the cooling water intake basin in the harbor of a power station. The trapped bed load weight and mean suspended sediment concentration were found to have good correlations with the Shields Parameter corresponding to conditions of simultaneous waves and currents. A simple numerical model was developed to estimate the total net transport rate of the bed load and suspended load. Calculated results from the model were found to give a reasonable estimate of the sand deposition in the intake basin.

Sea water intrusion into the diffuser system of a long sea outfall is likely to reduce the efficiency of the outfall. Recent surveys have revealed outfalls with a significant proportion of their diffuser blocked, and others discharging through as few as 50% of their diffuser ports. It is suspected that intrusion is the cause of these malfunctionings. Intrusion may be encouraged by the design requirement of low efflux velocities to obtain optimum dilution, or by over-design to cater for future increases in discharge. Although intrusion can only commence at low discharge velocities, when a port Densiometric Froude number falls below about 1, once it has started sea water will continue to enter the diffuser system until a state of balance is reached. Subsequently a considerably greater flow will be required to purge the outfall of all sea water. Intrusion may also be prematurely triggered by wave action, currents, or by reduced flow through a damaged port.

Hydraulic model tests at the University of Dundee are leading to an understanding of the intrusion mechanism as it affects the various diffuser systems in current use. A continuing research and site survey programme is aimed at design recommendations to eliminate intrusion or reduce its effects.

This study characterizes some wind stress effects on a coastal site which is a focus of small scale upwelling having a scale of the order of 10 km. Two time scales are considered. Firstly the seasonal character of wind stress with the associated sea temperature fluctuations is described. Secondly individual wind events of a few days duration are characterized by extent and rate of upwelling and offshore displacement of the thermocline front. Data on the thermocline displacement is fitted to Csanady's model of coastal upwelling, which leads to the prediction of upwelling parameters for given wind events.

Development of oil and gas fields offshore presents new problems in the design of the civil engineering works since in many cases little is known about prevailing conditions on the bed of the sea over the considerable lengths involved. Movement of pipelines after construction, is, therefore, not uncommon and may result in the failure of the line. Consequent losses can be considerable.

Pipelines are also vulnerable to mechanical damage by anchors and by fishing activities, the latter giving rise to political problems where the pipelines are laid across existing fishing grounds. To give protection to oil and gas pipelines as well as to marine outfalls and cables a method of protection has been developed using the well known concept of rock filled wire mesh or gabions.

To adapt this method to use in underwater works another well proven material, sand mastic asphalt, has been combined with the gabion mattresses to give the required qualities of strength, durability, flexibility together with sufficient weight to restrain movement in the pipeline.

The Southwest Ocean Outfall, with an overall length of 23,400 feet and a capacity of 450 million gallons per day, will be a major element of the Clean Water Program of the City and County of San Francisco, California, U.S.A. Offshore, the Outfall will cross one of the world's major active fault zones, the San Andreas.

Outfall construction started in 1981 and is scheduled to be completed by early 1985. The shoreward 3,000 feet is being built from a pile-supported trestle; offshore, the Outfall conduit sections will be placed using a barge designed and built specifically for the project. Oceanographic, coastal, geotechnical, and seismic conditions pertinent to design are presented along with discussions of specific Outfall design and construction features.

A study of sand and shell movement under the action of tidal flow was initiated in 1980 to ascertain the suitability of Whangarei Harbour at Marsden Point, New Zealand, for a proposed timber port. The aim was to assess the effects such a development may have on the sediment fluxes and the potential shoreline and channel instability that might be induced in the sandy inlet system.

Vertical water velocity profiles were analysed to determine bed friction coefficients which were subsequently broken into two component parts: one due to form drag and the other associated with the skin friction. It was shown that the Vanoni and Hwang (1967) equation for form drag can be extrapolated to include drag under tidal flow over megaripples. The skin friction component was obtained from the speed at 1m above the bed and the De5 grain size, by utilising the Karman-Prandtl equation. This was applied to analysis of bedload trap yields and bedform advance rates over megaripples and good agreement with the Yalin bedload equation for plane beds was obtained.

Suspended sediment transport was found to vary with U₁^7.75 and total load rates were in general agreement with the Engelund Hansen equation but deviations occurred due partly to expected form drag components being out of phase with the flow because of bedform hysteresis. Under tidal flow, the velocity at 1m was found to be a better predictor of sediment transport than the shear stress obtained from the velocity profile, thus methods presented in this paper use the lm speed for determination of total load transport.

During the flood year of 1981 the "Bot River Vlei", one of the largest lagoons in the south-western Cape of the Good Hope was opened to the sea by two artificial breachings of the sand bar which were aimed at lowering the excessively high water levels in the lagoon, and at the re-establishment of estuarine conditions.

The Coastal Engineering and Hydraulics Division of the National Research Institute for Oceanology of the CSIR surveyed and monitored the hydrological and sedimentological events in the Bot River lagoon and at the mouth channels in an attempt to formulate a future viable management policy, which had to take into account the conflicting interests of the defenders of continued artificial breachings and of those who advocate conditions of a closed estuary which will eventually change into a coastal freshwater lake.

The shear stresses and the associated friction coefficients at the interface and at the bed of an arrested saline wedge have been studied experimentally together with the detailed flow structure. Interfacial stresses, evaluated from hot film anemometer measurements and actual velocity profiles, agree well with those based on the integration of the equations of motion while the simplified onedimensional analysis gave considerably higher values. It was found that both the average interfacial friction coefficient, ${\overline{\text{f}}}_i$ and the average bed friction factor, ${\overline{\text{f}}}_{\text{o}}$ are best correlated with the dimensionless number, ReFr², where Re and Fr are the Reynolds number and the non-densimetric Froude number of the flow respectively, and with the relative density difference, $\text{Δρ/ρ}$. The results are presented in two families of curves with $\text{Δρ/ρ}$ as a parameter. The scattering of data points is minimal and the agreement with the results of some previous laboratory investigations and field data is good.

Predictions of time-varying salinity under arbitrary gate operations are presented of a tidal lake and its inlet. Particular regard is given to the modeling of salinity and gate operations and the applying the Markov process with Kalman filtering. In contrast to previous papers that predict only water levels the present paper also estimates salinity downstream the gate and of the tidal lake, simultaneously. It is postulated that present water level and salinity are expressed by the explicit linear functions of water levels, salinity, and gate operations at the other and the same positions in the past step. Using physically plausible estimate for these parameters in the functions the results of the predictions are in good agreement with observation.

The coastal region of the South Cape is presented as a typical semideveloped coastal zone with a limited environmental dataset. The author places himself in the position of a coastal engineer requiring wind data for design purposes. The various sources of wind measurement are discussed. Time series are presented depicting responses at different sites to the same large-scale synoptic situation. Spatial variations, both across the coastal boundary and in the offshore region are emphasized. These variations are brought to the attention of those engineers who may have to extrapolate wind conditions to their site of interest.

Limited information is available on currents in the semi-diurnal and diurnal frequency bands for the coastal ocean areas around Southern Africa. However, recently mooring data have become available from sites on the east, south and west coasts, and this paper analyses these results in an attempt to assess the importance of tidal and inertial currents.

It is clear that on the narrow shelf on the east coast the Agulhas Current dominates the energy spectrum, and tidal currents should be relatively unimportant at such sites. In the south on the Agulhas Bank the Current is still important, but comparable energy resides in inertial and tidal fluctuations. Modal analysis indicates the tides are primarily barotropic, with the inertial fluctuations mainly baroclinic.

In the absence of a major current on the west coast, most of the current variance occurs in the tidal and inertial bands; a complex vertical structure is also found. It is therefore clear that there are regions where such currents cannot be ignored.

This paper describes the method employed to refurbish an existing marine outfall which was deteriorating rapidly. It reviews the repair options that were available and how each was investigated and rejected until the best solution was obtained. Finally the execution of the project is described in detail through each stage to successful completion.

The indigenous water resources of the south west of Australia are under increasing stress. One solution to this problem may be water importation in the form of Antarctic icebergs. This possibility is discussed in this paper. To be cost effective, an iceberg water scheme needs to provide water at less than A$ 0.14 kl⁻¹ from the reservoir, and at yearly volumes of the order of 10⁸ m³. A number of analyses of the selection and towing of icebergs have been given in the past but the offshore processing of icebergs has been somewhat neglected. A conceptual solution to the processing problem is given in this paper and cost estimates for the acquisition and processing of icebergs are given. Our analysis indicates that iceberg water may be a feasible and economical urban water augmentation system for yearly volumes greater than 2 × 10⁸ m³. Further work on iceberg water augmentation is warranted.

The traditional concept of representing a "random" sea state by just a variance spectral density has been found to be insufficient for modelling the slow drift oscillations of large moored ships. This paper illustrates, through experimental investigations, the importance of including wave grouping as an additional design parameter. A special technique called SIWEH, developed by the Hydraulics Laboratory of the National Research Council Canada, for the generation of realistic wave climates which include wave grouping, is presented. However, when generating the grouped sea state, one also has to properly create the group-bound long wave components. The effect of proper compensation for the spurious free wave components is illustrated by the test results on the moored vessel response.

Problems arise during adverse sea conditions with ships moored to the ore loading jetty at Saldanha Bay. Under certain wave conditions moored ships undergo large motions which cause breakage of mooring lines. In order to obtain a quantitative insight into the problem an extensive measurement system was installed at Saldanha Bay. This system comprises instrumentation for the measurement of short and long waves, currents and wind, mooring line and fender forces and the motions of the moored ships. Preliminary results show that during storm conditions long waves with periods between 50 and 150 s occur at the jetty. These periods fall within the same range as the observed natural periods of horizontal oscillation of the moored ships.

This paper discusses the investigation that the U.S. Army Corps of Engineers, Portland District undertook for possible modifications to the Columbia River entrance channel. Because of the characteristically high sea conditions at this location, wave induced ship motions were considered an important criteria to be evaluated. Phototype ship motion measurements were obtained along with wave data. The data and preliminary results are summarized. Emphasis is given to the analysis of a relationship between environmental conditions and vertical excursions, and the subsequent use of the relationship to determine a new channel depth.

Extensive field monitoring programmes were implemented to study the behaviour of deep-draught ships in the entrance channels to two major South African ports. These programmes form part of a study to assess the future accessibility of these ports to ships of greater draught.

This paper describes briefly the monitoring techniques and methods of data analysis used, and reviews the results with regard to the environmental conditions during the monitoring operations.

The prototype results on ship response to waves will provide valuable data for the calibration and validation of both mathematical and physical models. The results are also used to establish allowance criteria based on safe underkeel allowances for ships presently using these harbours.

The new port of Richards Bay on the east coast of South Africa (Figure 1), was officially opened in April 1976 and was mainly built for the export of bituminous coal. Exports increased from an initial 2,5 million tons per annum to 26,5 million tons per annum in 1981. Extensions are now under way to increase this further to 44 million tons per annum by 1987.

A comprehensive study has been conducted for the Small Craft Harbours Branch, Department of Fisheries and Oceans Canada, to provide improved criteria for acceptable wave climate in small craft harbours (Ref. 3). An important part of the study was directed to model tests of vessel response to waves, and comparisons of these to field measurements and to simplified analytical predictions.

The objective of the present paper is to describe these specific comparisons and present the corresponding results in the context of improving harbour entrance designs. The model tests results and theoretical predictions are adequate to show quantitatively the dependence of a sailboat's response to different wave lengths with sufficient accuracy for wave periods, heights and directions to be selected as variables in formulating the required wave criteria.

In harbors affected by ocean swells,cargo handlings are often interrupted and irooring lines are broken as a result of severe ship motionsl). In order to decrease such accidents, the noored ship notions in a harbor basin must be studied. In this paper the ship notions in the harbor basin are investigated by using three dimensional Green's function and close agreement between theoretical and experimental results can be found. New methods to reduce noored ship motion are also proposed. The efficiency of these methods is verified theoretically and experimentally.

This paper describes a study carried out at Port Taranaki, New Plymouth, New Zealand to determine ship's track and motions at the port entrance. The results of the study being used to establish the extent (plan area) and optimum depth of proposed capital dredging works. The time lapse photographic technique, incorporating reference levels and bearings in each frame, used to record vessels entering and leaving the port is described, Maximum increases in ship's draft due to sea conditions are given.