Coastal Dynamics '01

Most computation-intensive numerical models of coastal morphology change have not yet incorporated behavior related to equilibrium states. This behavior may involve limiting depths for given waves and currents, limiting slopes, and limiting volumes of morphologic features such as ebb-tidal shoals. As a result, the simulation time scale in these models is in doubt. It is unclear whether a prediction has proceeded to a reasonable physical state or to a spurious one dictated by the nonlinear processes contained in the governing equations and imprecise initial and boundary conditions. Regularity in coastal morphology and equilibrium forms from micro-scale through regional scale can guide modeling efforts. This paper discusses the benefit of incorporating equilibrium properties in coastal morphology modeling. An example illustrates how micro-scale physical-based calculations might be combined effectively with a bounded macro-scale model of tidal inlet shoal evolution and sediment bypassing.

This paper reports on laboratory morphological studies of a shore-parallel porous breakwater system. A physical model study of the Elmer breakwater scheme, West Sussex, was conducted in the UK Coastal Research Facility (UKCRF) at HR Wallingford as part of an EPSRC-funded composite model evaluation (LUPY project). Mobile bed experiments, using both sand and anthracite as model sediments are described and discussed. Subsequent analysis showed that the evolution of equilibrium bays and their final width depend not only on the breakwater length and gap width, but also on the properties of the chosen model sediment and the permeability of the structure. It was also found that the 3D morphological changes influence the hydrodynamics, which in turn influences the evolution of the equilibrium morphological features.

This paper describes the erosion tests in front of submerged and emerged permeable breakwaters carried at the CIEM flume in Barcelona within the EU Scarcost Project. The maximum erosion for various wave trains is presented, and compared with some existent formulations. The influence of several parameters (wave height and period, breakwater freeboard) is also considered. The paper finally presents some numerical simulations of the wave induced current fields, which are potentially important for describing the barrier effect of the structure and where the influence of reflected wave trains is observed.

A pocket beach is known to be a relatively stable system in terms of longshore sediment transport and shows only a seasonal variation (swing) of a shoreline configuration depending on the dominant wave. Kagoshima Prefecture Government Office has conducted studies on an artificial (submerged) reef project in a pocket beach of Nagasakibana Coast since 1996. Coastal structures that affect a longshore sediment transport system are supposed to be constructed from the down-drift side. However, a field study and a theoretical analysis on optimal set-up of artificial reefs in the pocket beach revealed that artificial reef and also detached breakwater should be installed in the middle of the pocket beach first to minimize the maximum shoreline recession at a terminal beach.

This article presents how it is possible to numerically model the morphological evolution of a sandy beach in the presence of a detached shore-parallel breakwater. Seasonal variations (with weak and heavy swells) are reproduced and the annual evolution is compared with the one observed on a physical model. Thus six geometrical shapes of breakwater are studied. Three lengths from 100 to 200 m and two distances from the shoreline 150 and 200 m. The initial beach has an uniform slope of 1.4%. It represents a breakwater which can be found in the mediterranean sea, thus without tides. The wave comes at right angles to the breakwater. The wave climate is composed of 8 incident wave conditions. Three specific modules for wave, current and sediment transport are coupled and constitute our morphodynamic modeling system. For one calculation cycle, the first module solves the mild-slope equation (Berkhoff), the radiation stress are introduced in the second module to calculate the currents, and then the bed evolution is determined with the third module using the Bijker transport law. At the end of one cycle, the bathymetry is updated for the wave and current modules, then another cycle begins until the end of the physical model action duration. As a half breakwater is used, only multi-period waves are considered (no multi-directional). The Pierson-Moskowitz spectrum is introduced as the incident wave.

Field measurements from a cross-shore array of pressure sensors are used to examine the cross-shore variability of infragravity waves on a naturally barred beach. The data were obtained continuously over a three week period during the Delilah experiment, held at the Army corps of Engineers Field Research Facility on the Outer Banks of North Carolina in the fall of 1990. Infragravity pressure variances (based on approximately 1 hr records) are compared at 9 positions spanning the width of the surf zone from very near the beach face to 4.5 m depth. Ratios of pressure variance at different depths are used to make qualitative assessments of the cross-shore variability and seaward decay (or shoreward amplification) of infragravity waves. During low waves, the seaward variation of total infragravity energy is consistent with the theoretical decay of a mix of edge and leaky waves which decay seaward as h^–1 and h^–1/2, respectively, where h is the water depth. However, during larger waves the observed infragravity pressure variances are approximately uniform across the surf zone, suggesting that scattering effects are important in a wide turbulent surf zone. The observations also indicate that infragravity energy is locally maximum over the shallows of the sand bar, indicating an amplification of infragravity energy that suggests a possible link between energetic infragravity waves over the bar and morphological evolution.

Wave transformation on a transect crossing the surf zone of a fine grained, gently sloping barred beach has been further investigated with data from a 5-element coherent bottom-mounted pressure sensor line. In particular, field evidence is given for phenomena such as construction of merging shocks occurring on a short distance (several wave lengths) and deeply changing the wave field. The initially Rayleigh height distributions in 9-m depth were observed to be modified by shoaling and breaking into new distributions which present surplus at the medium waves and deficits at the lowest waves (compared with the Rayleigh wave distribution). Such phenomena are of great interest for random wave shoaling and breaking models.

Cross-shore variation of relevant wave characteristics at a barred beach are estimated through the separation of the wave field in incident and reflected components. Regular and irregular surface elevation data from a large-scale experiment are considered. The analysis shows that the reflection coefficient at a barred beach greatly depends on the wave period. Discussion of the spatial distribution of the net wave energy flux seaward and landward of the breaking point is carried out. Dissipation at surf zone per unit horizontal area is determined showing a strong influence of the wave height at breaking. This paper also highlights the limitations of using linear wave separation methods.

Nonlinear analyses are applied to surface elevation time series obtained from nearshore field measurements. Data sequences of wave sizes of all detectable scales are extracted and their fractal dimension is studied for evidence of deterministic behaviour. We observe that for some of our data, the fractal dimension appears to saturate for an embedding dimension of approx. 10. Surrogate data series would indicate that these wave size sequences may be deterministic, however the creation of surrogate time series shows this to be an artifact perhaps due to the narrow bandwidth of typical wave spectra. This re-confirms the conclusions of Elgar and Mayer-Kress.

The COAST3D project has made measurements of waves, currents, sediment transport and seabed/beach evolution at sites at Egmond in the Netherlands in 1998, and at Teignmouth, south Devon, UK, in 1999. The resulting data have been used to derive new understanding of coastal hydrodynamic and sediment dynamic processes, to test computational models of morphological evolution, and to formulate tools for coastal zone management. The project was undertaken within the framework of the European Commission's MAST-III Research Programme by a consortium of 11 organisations from five EU states.

Suspended sand transport measurements carried out on a meso-tidal beach at Egmond aan Zee, The Netherlands, during two Coast3D field campaigns in spring and fall of 1998 were used to examine the sand suspension mechanism and transport rates under waves. Two tests measured under similar wave conditions showed very different suspension patterns, due to different scales of suspension related to the bed form dimensions. Despite a wide range of bed form dimensions found in the field, the general trend of the data could reasonably well be predicted using a numerical model. Driven by the locally measured hydrodynamics and using a fixed bed roughness height the model predicted the current-related suspended transport rates near Egmond aan Zee shows reasonably good agreement with measured values (based on tuning of bed roughness).

In this work a complete analysis on the evaluation of longshore sediment transport rates along a Mediterranean coast has been presented. The objective is to analyze the different sources of uncertainty in the estimation of "measured" and "calculated" transport rates and how they influence the calibration of a transport formula. Results show the high variability introduced by the selection of different assumptions through the analysis, Due to this, the calibrated coefficient is only valid to forecast future transport rates and shoreline evolution by using the same wave data source and schematization and by using the same constraints imposed in the estimation of the inferred transport rates. All these points can be considered as one of the main sources in the different K-values reported in the literature.

The longshore sediment transport is studied both theoretically and experimentally. A way of modelling of the sediment-flow interaction is proposed, in which the flow is divided into three layers, namely bedload, contact load and outer layer. The luminescent tracer experiment was carried out at Ancao Peninsula (Portugal) under the MAST 3 research project INDIA in 1999. The model results were compared with the above data, as well as with the radio-tracer data collected previously at Lubiatowo (Poland).

Experiments have been performed for regular, bichromatic and random waves propagating with and against a turbulent current in a laboratory flume over a two-dimensional slatted bed roughness and over a smooth bed. Measurements of mean, wave-induced and turbulent velocities, instantaneous bed shear stresses, wave lengths and wave heights have been presented. The results provide an insight into the complex behaviour of combined wave-current flows and will be of use in the validation of wave-current models.

This paper considers the interaction between regular waves and a jet-like current in three dimensions. The waves are generated on still water and propagate to encounter the current at an arbitrary angle of incidence. A comparison is made between data obtained from an experimental programme and the predictions from a parallel modelling study. It is found that good agreement existed between theory and experiment, provided that the vertical structure of the current was taken into account.

A method for determination of the "apparent" bed roughness using "k-l" turbulence closure model is developed. An approximate expression for apparent bed roughness calculations as function of wave and current parameters is derived. The domain of variable parameters is selected accordingly to the Israeli near-shore conditions. Flow patterns on the Tel-Aviv coast have been computed by employing the 3-D CAMERI flow model and by implementing apparent roughness maps, calculated by means of the approximate expression. Comparison of flow patterns computed with and without accounting for wave generated turbulence in the boundary layer shows that disregarding of the increase in bottom resistance in the 3-D model leads to essentially different results. Implementation of the apparent bed roughness maps to the 3-D model results in decrease of flow velocities and divergence of the velocity field around marine structures.

This paper describes details of a hydrodynamic wave-current model and its applications to a tidal inlet (the Barra Nova) in the Algarve, southern Portugal. The model, as a core part of a Q3D model, is capable of computing wave characteristics and depth-averaged tidal and wave-driven currents, allowing for the effects of wind, wave diffraction and refraction, wave breaking and wave-current interaction. Application of the model was made to a highly dynamic tidal inlet — the Barra Nova to study the significance of wave-current interaction by running the model dynamically with both long waves (tides) and short waves (wind waves), as well as in a static mode with frozen tidal states. This newly-opened tidal inlet and its vicinity has been used during 1999 for a four-month intensive field campaign under Project INDIA (Inlet Dynamics Initiative: Algarve) funded by EU MAST III Research Programme. The model results obtained from dynamic runs were first compared with the field data measured by a number of remote sensing instruments, including X-Band Radar, and then with those obtained from static runs, which demonstrated the significance of the wave-current interaction on wave distributions for situations where strong tidal currents were presented.

The GENESIS model has been widely used for predictions of long-term shoreline evolution along wave-dominated open-coast beaches. However, to be applied in the vicinity of coastal inlets new capabilities needed to be implemented. This paper describes two new capabilities recently implemented in the model. The first is the extension of the model to calculate the longshore sediment transport rate by means of a generalized version of the Bagnold formula which makes it possible to describe longshore transport and shoreline change originating from multiple sources of a longshore current, such as wind and tide. The second is provision for inclusion of a pre-specified offshore contour to allow dominant local or regional bathymetric features to be represented in the wave transformation. Illustrative examples indicate promising results for these major enhancements of the model.

Malagueta beach has been a study case about the evolution of a urban beach, after its nourishment. A new one-line model is derived considering cross shore sediment transport in the swash zone. The model is applied to simulate the coast line evolution. Archive of field data have been carried along monitoring tasks, from 1991 surveys. The results fit very well to the obtained evolution data set from the prototype.

Shoreline change on Sendai Coast, Japan has been being measured every two weeks since 1993. The shoreline shows very complicated variation, though, there exists periods when two distinct patterns can be observed: one caused by cross-shore sediment transport and the other by longshore sediment movement. Thus, an empirical eigenfunction method is applied to the shoreline position data to detect these predominant modes. It is noted in the previous studies, this method is utilized for cross-shore beach profile, whereas in the present study, shoreline position is analyzed by means of this method. It is found that the measured shoreline data can be successfully separated into the modes caused by cross-shore and longshore sediment transports, accompanied with higher modes with negligibly smaller magnitude. The time-function of predominant modes shows close relationship with wave characteristics.

A numerical method able to simulate the flow generated after wave breaking is presented. The flow is calculated by a direct solving of the incompressible Navier-Stokes equations involving two phases (water and air) associated with an accurate interface tracking algorithm. The whole method is validated on the case of permanent non-linear wave propagation. A plunging breaking of a periodical wave is next simulated. The tracking of a "marked" fluid volume standing in front of the wave before the impact allows us to find the origin of the fluid involved in the splash up. This latter appears to be mainly composed by fluid particles that were initially in the wave through.

Storm waves responsible for severe beach erosion break in the mode of spilling when they approach the beach. Sediment transport processes associated with such spilling breakers have been still veiled due to difficulties in field observations under rough seas. A laboratory experiment was performed using a wave flume and a newly designed video-camera system with the purpose of investigating the flow characteristics and the influence of the turbulence on bottom sediments in the breaker zone of spilling breakers. It was revealed that (1) aside from the oblique vortex, previously observed, a new type of large-scale turbulence, named the "downburst", is present in the breaker zone and its structure is a descending water mass without marked rotational features; and (2) the downburst agitates sediment particles more vigorously than the oblique vortex, and the intensity of the downburst on bottom sediments is closely related to the energy of breaking waves.

Recent field observations of edge-waves on barred beaches show that bar-trapped modes can occur over the same frequency ranges as incident waves. Though `co-incident' edge waves must derive their energy from the incident waves, they are unlikely to be forced directly by modulations in the radiation stress gradients caused by wave groups since these cause forcing at infragravity frequencies. However, the radiation stress patterns around the breakpoint are not quite sinusoidal in space and time, so that the forcing pattern in that region has components that occur at multiples of the group frequency and wavenumber. The forcing due to these higher order harmonics is contained within the zone between the maximum and minimum breakpoint. Since the degree to which energy is transferred to edgewaves depends on how similar the cross-shore shape of the forcing is relative to that of the edgewave, transfer of energy to coincident edgewaves on barred beaches will be far more likely to occur over a wide range of frequencies than transfer to coincident edgewaves on plane beaches. Forcing conditions caused by incident wave group patterns are derived using a semi-numerical approach to demonstrate that co-incident edgewaves require a modulating breakpoint to occur. The degree to which the amount of forcing energy at higher order harmonics relative to that at the fundamental is shown to depend on the angle of incidence of the group at the breakpoint.

The formation and evolution of a new inlet at Chatham Harbor, Massachusetts, has been studied for thirteen years to document and understand the complex processes of inlet evolution and morphodynamic interactions. Morphologic evolution over the past thirteen years can be characterized by several changes. First, the inlet mouth has widened to over 1.8 km caused by the concurrent retreats of the adjacent barrier shorelines with recurved spits. Second, the southern part of the detached spit formed a barrier island that migrated southward and into the estuary, finally welding to the mainland shore. This resulted in closing off circulation to the southern part of the estuary, drastically changing the wave, tidal and littoral forces in what was a multiple inlet system. The entire tidal prism of the Bay now flows through the newest inlet. Third, with an abundant supply of sediments, the formation of substantial ebb and flood shoal complexes occurred. In order to maintain a navigational channel at this dynamic inlet system, a study was undertaken to identify morphologic patterns in inlet evolution and channel migration. With an understanding of this inlets dynamics, suggestions were made on navigation channel orientation that fit with the natural inlet evolution.

The present study focuses on the prediction of the long-term morphology of coastal inlets and tidal basins under the influence of accelerating relative sea level rise. The adopted model approach (ASMITA) considers the geophysical elements of a tidal inlet system at an aggregated scale, schematising the system into maximally three spatial units such as ebb-tidal delta, channel and flat. Analysis of model results shows that under the external forcing of a constant relative rate of sea level rise the system may tend towards a new dynamic equilibrium that depends on the sea level rise rate. However, if the rate of sea level rise exceeds a certain threshold value (stability limit), the system will no longer regain a state of dynamic equilibrium and the tidal basin will drown.

As a component of an assessment of coastal sand resources in Florida, crude estimates of planform area and volume of the flood-tidal delta are calculated for 61 inlets. At new inlets both area and volume tend to increase monotonically at a rapid rate initially, followed by slower growth in later years. A significantly non-monotonic growth may reflect major episodic effects. Inasmuch as most inlets have been extant.for several decades, the deltas have matured, although in some cases they have been significantly modified by dredging. Notwithstanding influences from this and other causes, area and volume of the mature delta have been plotted against spring tidal prism for all inlets. Weak correlations suggest that area and volume increase with prism, while the deposit thickness remains nominally invariant.

We solve the equations governing forced, dissipative, shallow water edge waves on a planar beach. Dissipation introduces waves that progress obliquely across-shore, away from regions of forcing, carrying energy to surrounding regions of dissipation. We derive matching conditions so that arbitrary profiles of water depth and dissipation strength can be simulated by patching together planar beach solutions. Forcing outside the surf zone, combined with dissipation inversely-proportional to water depth, generates a mixture of standing and shoreward-progressive waves. Shoreward decay of propagating waves in the surf zone weakens or prevents the shoreward increase in wave amplitudes predicted for simple topographies by non-dissipative models. The oblique shoreward propagation, net shoreward energy flux, and cross-shore homogenisation of surf beat amplitudes predicted by the dissipative edge wave model are consistent with field observations and are not predicted by non-dissipative models.

The effects of changes in the orientation of the coastline on edge waves in case of a landward concave coast are investigated, assuming circular depth contours. Both by analytical and numerical methods it is shown that fully trapped edge waves do not exist for realistic bathymetric contours. However, quasi-trapped waves with the properties of edge waves but with a small energy leakage do exist below some critical cut-off wavelength which is 1.15R for the 0 mode and that can be approximated by ΠR/2(2n + 1) for the higher modes where R is the curvature radius. The curvature produces a small increase of wavelength for a given period which can reach, for instance, some 12%, 18% and 24% for the 0th, first and second modes, respectively.

In this paper the nonlinear nearshore circulation model SHORECIRC is applied to simulate the infragravity wave conditions in the nearshore in response to windwave forcing during the 1990 Delilah campaign. For this purpose, the model has been extended to solve the energy equation for the short waves which acts as a wave driver for the long wave model. In order to simulate the wave conditions the directional offshore wavefield is synthesized from data including the bound, directional low-frequency components. The model-data comparison shows that the mean short wave transformation and mean longshore current on a real bathymetry are predicted correctly. Furthermore, the structure of the infragravity spectrum as measured in the nearshore array is correctly represented, albeit that the energy levels are somewhat underpredicted. Two-dimensional frequency longshore-wavenumber spectra show that the infragravity wave energy consists of edge waves and shear instabilities, and that the model accurately reproduces the main features in the measurements.

Sediment transport on natural beach profiles can be described in different ways, with respect to the spatial complexity and temporal variability of morphodynamic processes. The different transport phenomena display the results of the intensity of hydrodynamic processes acting along the beach profile. Field measurements indicate clearly that under storm conditions the suspended load dominates the total sediment transport, while under moderate wave conditons its contribution to the total transport decreases. Komar gives some examples on the relation of the two transport phenomena. The aim of this paper is to describe these transport phenomena qualitatively and quantitatively with the help of large wave flume tests which have been conducted within the MAST-III SAFE project in the `Grosser Wellenkanal' (GWK) in Hannover.

Suspended sand concentrations and bedforms under waves were measured in the controlled environment of a large wave flume. Three suspension conditions are discussed here; those occurring with regular (monochromatic) waves of height 0.55m over anorbital ripples, regular waves 1.0m high over orbital bedforms, and repeating wave groups (with a significant wave height of 0.6m) also over orbital-scale features. In all cases the wave-to-wave variability in suspended load was high (~30%). Patterns of suspension were dependent on the bedform type and on instrument location relative to the bedform. Regular waves suspended an order of magnitude more sediment than groupy waves with a similar significant wave height illustrating,the importance of sequences of high waves in pumping-up sediment concentration into the water column.

Detailed measurements of sediment concentrations in the sheet flow layer and in the suspension layer for flat bed conditions under prototype wave groups are presented. These measurements are part of experiments on near-bed sand transport processes, carried out in the Large Wave Flume in Hannover, Germany (SISTEX99). The results show that concentrations in the sheet flow layer are highly coherent with the instantaneous near-bed velocity. In the suspension layer, however, concentrations respond much slower to changes in near-bed velocity: at some distance from the bed concentrations increase and decrease on the time scale of the wave group, with a time delay relative to the peak wave within the wave group.

The mechanisms and time scales of sand suspension by irregular waves are considered on the basis of field experiments "Novomichailovka'93" (Black Sea), "Norderney'94" (North Sea) and "Ebrodelta'96" (Mediterranean Sea). The synchronous data for the suspended sediment concentration and the fluid velocity components were recorded by optical and electromagnetic sensors with a high frequency response. The sensors had been installed in a distance above the bottom of four to fifteen centimeters. The mechanism of the vortex ejection due to the shear instability of the bottom boundary layer is the most probable reason for sand suspension from the bed in area which are located seaward of the wave breaking point. In those areas the waves are greatly deformed and the bottom is nearly flat. The suspension events coincide in time with the first half part of decelerating phase or with the flow reversal after the passing of the wave crest. For such conditions the statistically significant coherence between sediment concentration and cross-shore velocity is confirmed for the peak frequency of the wave spectrum and its first harmonic. The concentration follows the velocity fluctuations with phase lag of –Π/4 at the frequency of wave spectrum peak. At the low frequency band (f<0.08 Hz) the sediment concentration poor correlate both with the velocity and its envelope.

The third-generation shallow water wave model SWAN is continually being updated to improve or extend its scientific and operational performance for applications in coastal engineering. The latest improvements and extensions include (nearly) diffusion-free propagation on any arbitrary scale (operational) and interaction between swell and windsea (experimental). Some effects of these improvements and extensions are shown in this study with a case of a major storm in the North Sea. The wave field is computed from the North Atlantic Ocean, through the North Sea to behind the barrier islands of The Netherlands with a resolution varying from 50 km to 40m. Some comparisons with observations are made.

The paper describes a short wave model which accounts for wave shoaling, refraction and breaking for the nearshore region. In a more general form this wave model will be suited as a wave driver for nearshore circulation models such as the SHORECIRC (SC) model. The main objective of the present paper is to examine which improvements can be obtained in the prediction of the short wave properties and forcing by using a wave theory more accurate than the sine wave theory used in models such as the REF/DIF1. The model is based on cnoidal wave theory for waves outside surfzone, and on the theory of nonlinear hydraulic bore and semi-theoretical analysis for surfzone waves. The time averaged short wave properties (short wave forcing) which drive nearshore circulation are also formulated. Comparisons to lab experiments show that this model estimates short wave characteristics such as wave height, phase speed and surface profile as well as short wave forcing very well for nearshore waves.

A numerical model based on the Boussinesq equations was developed to simulate wave transformation in the nearshore zone with emphasis on describing the effects of wave breaking. Two different formulations for estimating momentum losses due to depth-limited wave breaking, namely one proposed by Watanabe and Dibajnia and another one by Schäffer et al., were incorporated and tested in the numerical wave model. Detailed, high-quality data from Duck, North Carolina, collected during the DUCK85 and SUPERDUCK field experiments using photographic means were employed to evaluate the applicability of the Boussinesq model to surf-zone conditions. Model simulations were compared with measured time series of water surface elevation as well as with derived statistical quantities, such as the skewness, kurtosis, and root-mean-square wave height. Overall the model results were satisfactory, although the Boussinesq model employed failed to accurately reproduce the strong nonlinear shoaling that occurred prior to breaking.

A preliminary nonlinear breaking wave model is presented for a regular wave in an attempt to generalize it to field data by considering a Rayleigh wave height distribution to take into account the wave randomness. Surface elevation and velocities, represented as Fourier expansions, are calculated by solving Boussinesq type evolution equations. The dissipation of energy due to wave breaking is introduced in the set of evolutionary equations by including a damping term based on the Eldeberky and Battjes approach. The model is compared with the laboratory data of Ting and Kirby. In their experiment they considered regular waves propagating and breaking over a uniform sloping bottom. The comparisons show good agreement with the observed skewness of the free surface elevation. At the time of the redaction of this paper, the model is extended to a bar topography and compared with field data of Duck94, North Carolina.

Two profile models have been compared with field data measured at the Egmond sandy beach site (The Netherlands) within the framework of the European COAST3D project. Three hydrodynamic storm events and two morphodynamic events (storm and post-storm periods) have been selected for model comparison. The two profile models UNIBEST-TC and CROSMOR are process-based models; the UNIBEST-TC model is a deterministic model, whereas the CROSMOR model is probabilistic model based on a `wave by wave' approach. The models show reasonable results for wave height, longshore current and offshore bar migration.

A new method of predicting long-term morphological evolution based on a combination of depth-averaged, process-based modelling and a quick morphological updating scheme is presented. The hybrid method is first shown to represent free morphological behaviour with an accuracy depending on the frequency of updating the hydrodynamics and sediment transport. The method is then tested against large-scale developments in some outer deltas in the south-west of the Netherlands, over a period of 36 years, during which large-scale human interference in the form of the Delta works took place.

A new method to update wave and current values in a time evolving beach morphology model is developed. The method is based on the assumption that small bottom perturbations induce small variations in the hydrodynamic magnitudes resulting in a new set of linearized perturbed equations that can be solved easily. The method has been tested for 1-D as well as for 2-D bathymetries showing its capabilities.

A finite difference numerical model has been developed to represent the large-scale morphodynamics of complex coastal systems. The model is based on the Coastal System Tract (CST) concept. This model contains representations of coastal features such as tidal inlets, bays and rivers, beach systems (including littoral drift and profile migration), as well as the offshore zone comprised of the shoreface, shelf and upper continental slope. The purpose of the model is to diagnose the large-scale morphodynamical interactions between the components of the CST that have previously only been represented as separate or incompletely link elements. The results show that there are important forcing factors that result from different characteristic response times of the independent system components. This means that the system components are never properly tuned to respond in unison to perturbations in the large-scale forcing parameters.

An extensive field data collection effort was undertaken in Fall 1999 to examine wave propagation and currents through an inlet entrance. These data support a circulation and wave model for Grays Harbor, Washington, a jettied entrance with a large tidal prism. Both the field data and model results show wave attenuation in the inlet entrance, flood currents strongest on the north side of the inlet, and ebb currents more uniformly distributed. The influence of the tidal current and water level on wave transformation was also examined. Ebb current produces the greatest change at the inlet entrance, increasing wave heights by as much as 0.5–1.5 m. Flood current increases wave height at the seaward end of the entrance due to the ebb shoal redirecting flow offshore, but reduces wave height in the inlet throat. Water level has a minimal impact on wave height in the inlet entrance, but does control wave height in the back bay.

Channels that are maintained through an inlet entrance often tend to realign at an angle from the inlet centerline. Through hydrodynamic numerical modeling combined with field monitoring, the authors have found that eddies created by the ebb-tidal jet are a contributing factor to channel realignment at Shinnecock Inlet, Long Island, New York. Eddy formation is a non-linear process associated with the advective terms in the hydrodynamic equations. Such eddies can grow and migrate over the ebb-tidal cycle, according to the phasing of the ocean tidal wave and the ebb flow. Interaction of. the eddies with, the ebb-tidal jet forces the jet to migrate and change shape. Because the strength of the ebb jet is sufficient to move sand during most of the ebb cycle, the jet will scour the bottom over the path that it traverses. This scouring action moves the entrance channel from its dredged position to the location of the ebb jet.

Within the framework of the COAST3D project of the EC MAST-III programme, data from a measuring campaign in Teignmouth, United Kingdom, are used to evaluate the performance of two hydrodynamic area models (PISCES and Delft3D, developed by HR Wallingford and WL | Delft Hydraulics, respectively). The dataset contains spatially dense hydrodynamic measurements of a complex coastal area. In this study, predictions from both models are compared with measured velocities for various tidal and wave conditions. Both models show good overall agreement with the measured velocities, particularly if the complexity of the site is taken into account. The fact that both models were applied with standard values, were calibrated on the measured water level at one station only, and yet yielded satisfactory results affirms the utility of applying these models in consultancy projects.

The relative importance of remote and local forcing on the subtidal response in Galveston Bay was studied using water level and wind data observed during the winter and spring months from 1997 to 2000. The study confirmed the importance of remote forcing through Eckman transport for the water level response and local forcing for the surface slope response. These two forcing mechanisms act independently since the estuary axis was oriented roughly orthogonal to the coastline. A neural network model was introduced which used the meteorological data to predict the water level anomaly which; when added to the harmonic tides, provided good estimates of the total water level at the bay entrance (remote forcing).

Velocity and sand concentrations observed at two elevations in the throat of the Frisian Inlet are analyzed. The analysis is limited to the flood phase and period of little wave action. The results show a clear threshold velocity between 20 and 30 cm/s, which is in accordance with the Shields criterion. For all tidal cycles, a hysteresis between concentration and current speed is observed. Vertical concentration gradients are larger during decelerating than accelerating flow. The observations suggest that there is no unique relationship between local current velocity and concentration that covers the entire tidal cycle.

An artificial tidal inlet cut through the barrier island system of the Ria Formosa (South Portugal) has been studied. A conceptual model of functioning of the Barra Nova tidal inlet is proposed, mainly based on the morphology survey of the ebb-tidal delta during fair weather conditions following an important storm event, and on the long-term survey by video techniques.

Six tripods were deployed at shallow (~14-m) and deep (~24-m) sites on the northern, middle, and southern flanks of the Grays Harbor, Washington, U.S.A. ebb-tidal delta from early October through December, 1999 to measure waves, currents, temperature, and suspended-sediment concentrations as part of a wave-refraction and sediment-transport experiment. Directional wave spectra show that the general direction of wave approach shifted from WNW to WSW as the North Pacific weather pattern shifted from summer to winter, and we were fortunate enough to capture a large storm (offshore significant wave heights of ~8 m) in late October and a sequence of about 8 smaller events with ~4 to 6-m waves in November and December. As expected, wave directions indicated refraction around the ebb-tidal delta, and varied with incident wave period and direction. Direct estimates of sediment flux about 0.4–0.6 m above the bottom, and modeled estimates of depth-integrated suspended sediment flux indicate net offshore and northward transport. By comparison, estimated net bedload flux was onshore, but at much lower rates. These results indicate that sand on the ebb-tidal delta can be mobilized frequently under winter conditions, and can bypass the inlet at depths of at least 24 m. The data also suggest that significant offshore transport occurs at these depths and that offshore suspended-sediment transport during winter is so great that it might not be balanced by onshore bedload transport.

Field observations of hydraulic gradients and the corresponding friction factors (f) from rip systems are presented. From a range of field experiments we find f =0.032±0.014 in rip feeder channels. A model is formulated describing rip current systems as wave pumps. That is, the incoming wave power is related to the useful pump power of the rip system through the pump efficiency ε_rip. The presently available data indicate that ε_rip is almost constant around 0.035. This gives a predictive capability when the wave conditions and certain aspects of the topography are known. We present ε_rip-data from both laboratory and field. In our study mean-water-level-data with centimeter accuracy were effectively collected with simple stilling wells.

Measurement data of a rip-current system are used in a comparison with model results to verify a coastal area model operating on the time scale of wave groups. The measurements were performed as part of the RDEX field experiment at Palm Beach (Australia). Continuous coverage of rip current velocities over a large number of tidal cycles was obtained, with wave conditions varying from calm to storm conditions. Video time exposures of the surf zone are used to calibrate the model dissipation coefficients by comparing the measured intensity and the computed roller energy. Selecting a subset of the measurements covering 11 days at the beginning of the experiment, the agreement with computed results is generally good.

This paper presents the numerical hindcasting of the first eleven days of the main experiment of the Coast3D project, carried out near Egmond (the Netherlands) in the autumn of 1998. In this period a rip channel develops with an orientation correlated with the wave direction. Hindcasting the wave height and velocity measurements with a numerical model resulted in an increasing underestimation of wave breaking going closer to the shore. Consequently, there is good agreement between the computed and measured longshore velocities on the seaward slope of the inner breaker bar (Station 1A), but less so for these velocities on the shoreward slope of the inner breaker bar (Station 1D). We expect this to be due to the inability of the wave model to predict rapid decrease of the wave height across the inner bar. Cross-shore velocities in the main transect are dominated by rip currents and wave breaking-induced return flow. Qualitative agreement is fairly good but quantitative agreement between the measurements and the model results is only reasonable, probably caused by the application of a 2DH mode, inaccurate prediction of the location of rip currents and differences in the bathymetry.

A two-dimensional physical model study of oscillatory flow over sand ripples has been carried out in a wave flume. The main aim of the tests was to provide flow data in the bottom boundary layer over a rippled bed. This was carried out under real waves rather than pure oscillatory conditions such as those produced in oscillatory flow flumes or oscillating bed models. Simultaneous horizontal and vertical velocities were collected using a two-channel laser Doppler anemometer over vortex ripples. This allowed the direct calculation of the turbulent kinetic energy, Reynolds stress and eddy viscosity.

In the framework of the MASTIII COAST3D project, the nearshore multiple bar system of Egmond-aan-Zee (NL) was morphologically and sedimentologically investigated on the basis of very-high resolution digital side-scan sonar imagery and intensive samplings. An analysis of the small-scale morphology revealed a number of bedforms and could be interpreted in terms of the variations in current velocities. Although, the area is generally devoid of bedforms some regions occur where complex multi-scaled ripple patterns occur such as lunate megaripples along the seaward flank of the outer bar and in the throat of a rip channel, In the trough between the outer and inner bar, cross-shore oriented symmetrical ripples can be distinguished showing sharp delineations with zones of no bedforms. A clear relation exists between the spatial distribution of the surficial sediments and the morphology. The area is characterised by medium, well-sorted sands except for the trough between the inner and outer bar where mainly coarse moderately sorted sand is found. Fine sands are sporadically present, as well in the trough as along the seaward flank of the outer bar. A sediment trend analysis (STA) was performed to calculate residual transport directions on the basis of the areal distribution of some sediment parameters and the results were correlated with the results of the COAST3D modelling group. Confirmed by the modelling results, the STA approach shows a dominant residual tidal current in a northeastern direction; offshore directed flows in the rip channel and a wave-induced longshore current in a southwards direction.

A simple bar migration model was used to characterize observed differences in bar migration rates from three bars analyzed at Duck, USA and from one bar analyzed at Katwijk, NL. In these cases, bar migration was consistent with the model, indicating that they migrate toward a time varying break point at a rate that depended on the wave height and the instantaneous distance between a bar and the break point. Differences in the response rate characterization depended significantly on differences in the mean beach slopes and differences depended on the mean water depth occupied by the bars. These differences were accurately parameterized by a scaling argument based on bed load sediment transport.

Tidally-induced longshore currents are observed to occur outside the surf zone on a steep, macro-tidal, mixed (sand and shingle) beach. The tidal currents are of sufficient velocity to mobilize shingle-sized sediment of D₅₀ = 0.2m. Only westerly-flowing tidal currents were experienced upon the beach face. In moderate winds tidal stage and range had greater influence on resultant longshore velocity and direction than wind speed or direction. The intertidal shingle bank experienced longshore currents in the opposite direction to the prevailing wind direction.

The presence of a horizontal flow circulation on an intertidal beach with a cell topography is studied for a flood period with small waves (H_1/3 = 0.46m) and a flood period with shore-normal incoming waves (H_1/3 = 1.3m). A horizontal flow circulation developed in case of small water depths over the intertidal bar with local relative wave heights over 0.35 m. In this case, the flow circulation was driven by the onshore mass fluxes of water (Stokes drift and roller flux). The measured mean current velocities of the feeder channel and the computed mean velocities, based on the mass flux balance at the intertidal beach, were in the same order of magnitude.

The development and testing of a numerical hydrodynamic model, intended for use in the study of (steady) nearshore currents in the presence of complex bathymetry, is described. Governing equations for conservation of mass and momentum are quasi-3D, in that analytical depth-integration is split at the wave trough level so that fluxes associated with the organized wave motion and the breaking wave roller are separated from the underlying current. The currents/set-up model is driven by a simple, period-averaged wave transformation model that describes shoaling, refraction, and depth-limited breaking, which is also used to drive a planform model for the breaking wave roller. Initial results for a 9,000 ft long stretch of beach that is fronted by large expanses of submerged rock reef demonstrate that even for highly complex bathymetry the model is both robust and efficient, indicating that it may be suitable for large-domain, long-term analyses.

A numerical model is presented for the simulation of wave-current driven sediment transport and it is developed as research tool. In line with previous papers the present paper validates its applicability as a research tool improving sand-transport formulations. This paper compares the Stokes drift derived from laboratory experiments with the numerically simulated Stokes drift. Further, two analytical test cases are included, one for sand transport in steady flow and a second case for oscillatory bed concentration. Based on a sensitivity analysis, we recommend other researchers to also validate their sand-transport models against these analytical test cases.

A sediment transport model based on the two-phase mass and momentum equations is introduced with appropriate closures on the near bed boundary conditions and fluid turbulence. The pick-up function proposed by van Rijn is implemented as the boundary condition for the vertical sediment mass flux. The performance of the proposed model is tested against seven laboratory U-tube measurements. It is found that the two-phase model has an adequate ability to predict the time-averaged concentration under various combinations of maximum free-stream velocity and oscillatory period. However, some discrepancies are observed in the comparison of the time history of the concentration profiles. We anticipate that this is due to the approximations in the near bed boundary conditions in the sediment-phase. Some future work towards a better understanding of the near bed boundary conditions is suggested.

In the coastal zone, rather large variations of the sediment composition can be observed, as a result of spatial differences in hydrodynamic conditions. In order to be able to calculate the selective sediment transport, the full grain size distribution of the sediment has to be taken into account. Measurements have been carried out in a wave flume with a well-graded, rippled sediment bed to determine the net sediment transport per fraction. The data show that the finer sediment fractions are more easily transported, resulting in a coarsening of the remaining bed.

In this paper, a series of wave flume tests on wave-induced seabed responses of a fine sandy soil (d₅₀= 0.069 mm) were conducted to investigate near bed sediment suspension upon fluidization. One sediment suspension-concentration probe and several pore pressure transducers were deployed few centimeters above and at different depths inside the test soils for simultaneous measurements under monochromatic wave actions. The measurements display that the mean values of suspended sediment concentration start to increase at about several wave cycles after the set-off of soil fluidization with significant pore pressure build-ups in the test sandy soil. Under continuous wave actions upon the fluidized soils, suspended sediment concentrations used to increase to a maximum and then decrease slowly. In the following runs of wave actions with similar heights and periods, thickness of fluidized seabed generally reduces while the maximum suspended sediment concentration drops off as well. These results were always confirmed with clear observations of sediment pluming events over the fluidized beds.

The influence of prominent harbour moles (2.5 km cross-shore, 1 km longshore) upon a wave-dominated shoreface has been identified. It is shown that the effects extend beyond the shoreline and localised shoreface (i.e. up to 2 km from the moles) such that they can be identified over the entire shoreface up to 18 km either side of the moles. These effects are identified as shoreface activity, which is shown to increase with temporal period up to the data limits (32 years). Hence, anthropogenic intervention can have a much greater influence than often appreciated. These results have coastal engineering and managerial significance, especially over the larger temporal and spatial scales.

A processes-based shoreline change model is applied to simulate historical shoreline behaviour and predict future change along a 40-km stretch of coast in the US Pacific Northwest. The sensitivity of the modeled shoreline change to different schematizations of the wave climatology is examined. The modeling results are most sensitive to directional changes in the incident waves as well as to the anomalous wave climatology associated with the 1997–1998 El Niño. The sensitivity of shoreline change to a future climatology based on a trend of increasing significant wave height and peak period, was minimal. The impact of the 1997–1998 El Niño was compared to 1999–2000, a more typical wave year, suggesting that on annual time-scales cross-shore transport processes dominate over alongshore processes.

Nearshore bathymetry data collected along the US Pacific Northwest is analyzed to determine relationships between nearshore morphology and shoreline change at a variety of spatial scales. Sandbar properties are compared at three sites, locations separated by tens of kilometers but subject to similar hydrodynamic forcing and having like sediment characteristics. These three sites, an accreting coast, an eroding coast, and a dynamically stable coast, are shown to exhibit different large-scale coastal behaviour. The nearshore profile slopes at these locations are correlated with recent shoreline change history. Measurements have revealed offshore trending sandbars that are continuous over alongshore distances of tens of kilometers, behaviour that spatially mimics the net offshore bar migration documented on several other coasts.

Modern coastal processes at the southeastern coast section of the Sea of Azov are estimated in this work. A mathematical model is proposed. It affords to predict a rate of the coastal abrasion in conditions of the sea level rise. Recommendations are given for the choice of the coast protection methods in this region.

At Duck (N.C.) and four Dutch sites, subtidal nearshore sandbars have a multi-annual lifetime during which they are generated near the shoreline, move offshore through the surfzone, and decay offshore. We use complex empirical orthogonal function analysis to quantify bar length and bar amplitude across the surf zone, and to compute the average time interval (or, `cycle period') at which bars are formed near the shore or decay offshore. Viewed as a function of mean depth, the bar length curves are statistically identical for all sites and the shapes of the bar amplitude curves are remarkably self-similar. Cycle periods range from approx. 3.2 years at Duck to 16 years at one of the Dutch sites. The inter-site differences regarding bar amplitudes curves and cycle period could not be linked to differences in simple bulk statistics of the nearshore wave and waterlevel climate, or to differences in the nearshore bed slope.

Alongshore uniform (2-D) and nonuniform (3-D) nearshore bar crest morphology was quantified for a 3.4 year data set of videoed alongshore bar crest lines collected at the double barred beach of Noordwijk, Netherlands. Typical spatial and temporal scales of bar crest change were derived from this description. Cross-shore bar migration (2-D change) was characterised by a dominant yearly offshore trend with superimposed monthly (seasonal) and weekly variations. In contrast, 3-D changes mainly varied on weekly to monthly scales. Observed 3-D features, like crescentic bars and rips, had typical life-times of several months, and migrated alongshore or changed in cross-shore amplitude over weeks to months. Total bar change was thus dominated by 3-D change on weekly to monthly scales and by 2-D change on yearly scales, meaning that the common assumption of 2-D bar change dominating over 3-D bar change is only valid on restricted time scales.

Unusual streaks morphologically similar to transverse bars of the nearshore zone of oceans and seas have been recently observed on the barred beaches of a large man-made lake. These features ran perpendicular or oblique to the shoreline and showed the bands to be bathymetric ridges with a relief of 0.2–0.5 m. Sometimes the bars were found between the shoreline and inner longshore bar but more frequently offshore the inner bar. It has been found experimentally that wave-breaking and associated littoral processes were responsible for the transverse bars formation on the tideless beach of the man-made lake.

This paper considers the nature of Large Scale Coastal Behaviour (LSCB) data sets, the importance of which has been highlighted by the nearshore research community. It is suggested that the application of a data reduction approach to the analysis of these morphological data sets will greatly facilitate their interpretation. A data technique, consisting of the parameterisation of sand bar features is outlined and applied to two examples from LSCB data sets; a bathymetry and an image of wave breaking patterns in the nearshore region. The data reduction technique succeeds in reducing the magnitude of the data whilst still encapsulating a significant proportion of the information content of each of these data sets. The paper then proceeds to demonstrate the utility of the data reduction technique in the subsequent interpretation of the evolution of behaviour evident in these LSCB data sets. It is shown that it is possible to emulate the results obtained from analysis of the original data sets from the parameterised data set.

The East Anglian coastline (UK) is used as a case study to illustrate an innovative quantitative approach to large-scale beach classification. This encompasses a systematic application of a statistical analysis technique to beach volume change, representing the spatial variability of the inter-tidal beach. This technique is used to classify the East Anglia coastline into regions of similar inter-tidal profile behaviour. Classification results can be used to inform ongoing debates about the location and boundaries of coastal management units. The technique proposed in this paper is generic and can be applied elsewhere.

The US Army Corps of Engineers (USACE) initiated the Regional Sediment Management (RSM) Demonstration Program in recognition of the interaction of engineered navigation and beach restoration projects with adjacent coastal projects. A major component of this initiative is the creation of a regional sediment budget for a 360-kilometer stretch of Gulf of Mexico shoreline that includes nine federal navigation projects and one federal beach restoration project. This paper outlines two elements of creating a sediment budget for a single inlet system within the RSM region: identifying sand transport pathways and computing changes in sand volume. Both elements rely heavily on comparison of successive SHOALS data sets collected at East Pass, Florida, USA. The paper then describes how the information inferred from the comparisons will be incorporated into the macrobudget for the entire region.

A four-county beach segment in northwest Florida, USA, suffered severe to moderate erosion from Hurricanes Eloise, Opal, and Georges in 1975, 1995, and 1998, respectively. This paper analyzes 13 sets of high quality beach profile data — measured at 1000-ft longshore spacing along approximately 50 miles of coastline in the four counties over a period of 25 years — to document hurricane impacts on the subaerial and subaqueous portions of the beach and to assist in making beach and dune management decisions following future storms. The historical beach profile database is used to develop a beach profile averaging method to generate characteristic beach profiles and to characterize spatial and temporal changes in beach condition. These characteristic profiles are used in conjunction with high-frequency dune erosion modeling and the observed performance of assisted recovery activities (emergency sand berms, beach scraping, dune revegetation, and sand fences) to develop a tool for predicting — before the occurrence of a severe storm — the probability of success or failure of assisted recovery activities meant to accelerate natural beach recovery.

This paper discusses the effects of a change in discharge regime of the River Rhine on the cohesive sediment transport in the Dutch coastal zone and towards the Wadden Sea. Two 3D numerical models are used to quantify these effects in a comparative study. It is shown that a different discharge regime may influence the sediment concentration in the Wadden Sea.

Morphodynamic models can play a key role in assessing the likely impact of coastal developments. A major goal is the prediction of predicting coastal evolution over a regional scale. The deterministic approach to morphological prediction was pioneered by Pelnard-Considere who presented a one-line equation to forecast changes in coastline position. Subsequent models have extended this to multiple depth contours. Such models, however, do not forecast accurately the variability likely to be experienced in real situations. One alternative to the deterministic modelling approach is to employ stochastic forecasting methods, such as Monte Carlo methods. A key input to this type of model is the statistical description of the wave climate or other driving forces. However, to obtain physically meaningful results very many realisation are needed. In this paper we formulate and solve moment equations for the shoreline position. These equations describe the averaged or long-term solution, and its dependence on wave-climate; and eliminates the need for computationally intensive Monte-Carlo simulations.

A theoretical study is made of the shape, dimensions and growth rate of a spit. The spit is growing without changing its shape, being fed by the littoral drift along a long straight coast up drift of it. A one-line model for coastline development is used as basis for a simple analytical model. This is based on the assumption that the local longshore sediment transport is determined by the local orientation of the coastline relative to the incoming waves. The analytical model predicts an exponential shape of the spit. The analytical model gives a growth rate inversely proportional to the length scale and this model predicts an infinitely narrow spit to emerge. The sediment transport around the spit is investigated by two-dimensional models. The shape of the spit is still taken to be exponential and the growth rate is estimated on basis of the longshore sediment transport along the base of the spit. The maximum growth rate is found for a finite length scale of the spit and a spit of a finite width is therefore expected to emerge.

Monthly records of shoreline positions, taken between Sep. 1983 and Sep. 1999 at IBW PAN Coastal Research Station at Lubiatowo, Poland, were examined with advanced statistical methods — Singular Spectrum Analysis (SSA) and Principal Oscillation Patterns (POP). The SSA allowed the extraction of shoreline change patterns. The most significant one reflects long-term trends of shoreline change, having a clearly visible periodical structure with cycles ranging between 8–10, through 16, up to 25–30yrs. Detailed periodic structure of these trends was examined with an FFT routine. The POP was applied to them and found that trend variations can be understood as sum of two cyclic patterns with periods very similar to those found with the SSA (10 and 18yrs) and a pattern reflecting simple exponential decay of a given shoreline configuration. These patterns may serve as a data-driven model for prediction of future trends of shoreline evolution.

The goal of this research is to evaluate a two-dimensional bottom boundary layer and sediment transport model with field observations of velocity and suspended sediment concentration. The model-data comparisons show several interesting phenomena. First the model over-predicts bed shear and underrepresents mid-bottom boundary layer shear. The Engelund and Fredsøe bedload model favorably compares with nearbed concentration observations. The model predicted suspended sediment plumes are of similar order magnitude to the observed plumes. These model-data comparisons will help field observation interpretations over complex topography and quantify the strengths and weaknesses in the model's physics.

The Boussinesq-type modeling concept provides an adequate and efficient means to describe wave dynamics in coastal regions. They are an extension of the depth-averaged shallow-water model and include the propagation of short waves as well as nonlinear effects. The Boussinesq-type model applied in this paper has been developed to obtain an accuracy sufficient for practical purposes within limited computing times. We present the extension of this model with a 2-D wave breaking model based on a combination of the eddy viscosity concept and the surface roller concept. The combination has a number of features that makes it suitable for nearshore applications. Mass and momentum are strictly conserved while the wave breaking model only dissipates energy, which is in agreement with physical laws. The results and the comparison with experiments under very different wave conditions demonstrate the good performance of the model.

Shallow foreshores considerably affect wave propagation and hence wave run-up on coastal structures. This concerns for instance the evolution of wave height distributions and wave energy spectra between deep water and the toe of coastal structures. Based on numerical model investigations and physical model tests it was concluded that wave run-up and wave overtopping can best be described using wave conditions at the toe of the coastal structures, in particular using the wave period T_m-1,0. With this wave period the effects of arbitrarily-shaped wave energy spectra on wave run-up and wave overtopping can be taken into account. In this paper estimates of wave conditions, including estimates of this wave period T_m-1,0, are obtained using a spectral wave model and a time-domain Boussinesq-type wave model. Especially the applied Boussinesq-type wave model appeared to be appropriate for conditions with severe wave breaking on a shallow foreshore where also a considerable amount of energy is transferred to lower frequencies. Also wave interaction with the dike itself was modelled numerically.

Advances in computational power have permitted the use of Lagrangian particle methods to model fluids, particularly separated flows and free surface flows with splashing. Here we present a smoothed particle hydrodynamics (SPH) approach to a numerical wave tank, capable of studying several types of waves impinging on structures and walls.

Both the effects of alongshore non-uniformities in the bathymetry and the importance of a roller dissipation mechanism in the modeling of longshore currents are examined and compared with measurements from a barred beach during the Delilah experiment. Comparisons are made with predictions from Delft3D, a depth-averaged two-dimensional hydrodynamic model. Measured cross-shore and alongshore variations of the mean longshore current are compared with the model for both uniform and measured bathymetry with and without a roller dissipation mechanism. It is demonstrated that the longshore current maximum in the trough measured during Delilah is the result of both the roller advecting momentum into the trough and alongshore pressure gradients associated with non-uniform bathymetry.

The calculation of long term coastal sediment inputs often ignores key components and uncertainties. These issues are analysed using the calculation of fine-grained sediment input from the English East Coast (Margate to Berwick on Tweed) to the North Sea between 1850 and 1994. Error analysis suggests a potential error of up to 21%. A key constituent of this error is sediment production from shoreface erosion below Mean Sea Level (MSL), which is responsible for approximately half of the total inputs. This source has been ignored in many earlier budgets. Therefore, improved estimates of both historic and future sediment inputs require better understanding of shoreface evolution on eroding coasts, as well as the composition of the eroded material.

Investigations of sediment dynamics and geomorphological changes on seasonal and annual scales have been carried out in two different sandy nearshore areas of the southern Baltic Sea. A new method called `tracer stick method' was applied to collect data, about erosion and accumulation to an accuracy of centimeters. On a seasonal basis there is a geographical variation of maximum erosion and deposition values, with the zone of maximum mobility shifting towards offshore during the stormy winter period. At both investigated areas the high sediment mobility does not automatically result in conspicuous morphological changes.

The coastal morphology of the Mouth of the Columbia River (MCR) has changed dramatically over the past century. Since the construction of jetties in 1914 and 1917, the inlet deepened and stabilized, the outer ebb delta migrated northward and offshore several kilometers, and the adjacent shorelines to the north and south prograded several hundreds of meters. Recently, high rates of erosion along these adjacent shorelines affecting parklands have raised questions about the cause of the erosion. The purpose of the sediment budget analysis presented here is to address these questions by quantitatively examining the links between the morphology change of the shoreline and the ebb delta and historical changes in local and regional sediment supply.

Detailed measurements of individual swash events, the shape of the swash lens, pore-water pressures in the beach and hydraulic gradients were obtained for a gravel beach on the south coast of England. Swash events were classified using the surf/swash scaling parameter and showed that surging breakers occurred in 80% of the runs with the remaining 20% resulting from plunging breakers. Two types of swash behaviour were identified. Swash events at the start of a run showed conditions when there was no interaction between backwash and the uprush of the next swash event; whereas the swash time histories illustrated conditions towards the end of the run, when swash/backwash interaction was prevalent. Downward-acting hydraulic gradients (infiltration) were measured for most of the uprush and backwash cycle. Values were an order of magnitude greater than infiltration events measured on sand beaches.

Hydrodynamic parameters of wave-induced water motion in the swash-zone, predicted by a numerical solution of the Non-Linear Shallow Water (NLSW) equations, are compared with full-scale data on two beaches, one mild and the other steep. The model can predict many aspects of the swash hydrodynamics, particularly horizontal velocities, which is important with regard to sediment transport. The results are sufficiently reliable for the prediction of sediment transport in the swash-zone. However, agreement is better for the mild beach than for the steep one, probably due to higher permeability which exists on the steep beach with a larger grain size.

This paper describes studies of the bed friction factor in non-stationary free surface flow, for the specific case of bores moving in shallow water and breaking waves on beaches. Data from a set of experiments carried out in a laboratory flume, including measurements of water level displacement and fluid velocity through LDV, are used to evaluate the bottom stress and the mean flow velocity. The results of other measurement techniques in similar conditions are used to critically assess our results. It is found that the friction factor is higher and the transition from laminar to turbulent boundary layer takes place at higher Re numbers than values obtained in separate studies of turbulent bottom boundary layer including an additional source of turbulence.

A new formulation is presented for estimating the relative compatibility of borrow material used in beach nourishment projects, that is the determination of the amount of required borrow material needed to yield one unit that is consistent with the native material. The formulation is based on Galofre (2000) model of equilibrium grain size distribution. According to this model, the borrow material placed at the beach face will be rework and its distribution will change until an equilibrium is reach. This equilibrium condition depends on the wave climate and the borrow material characteristics. The formulation defines two different overfill ratios, the Stable sand overfill ratio, S, and the Remaining sand overfill ratio, R. The first ratio accounts for the borrow material that fits the equilibrium distribution and consequently is stable in the active beach profile. The second ratio adds the coarse sand that, although is not "stable" in the active beach profile, is transported shoreward and, from the nourishment point of view, is not lost. A critical diameter, φ^*, is define in order to evaluate the R-ratio. Use of these formulations requires knowledge of the mean of the borrow sand and the standard deviations of the borrow and native material.

A statewide monitoring program has been documenting changes within the beach and nearshore system along the South Carolina coast since 1993. Additional beach and geophysical surveys completed at recent beach nourishment projects have provided a consistent means to assess project behavior over long periods. Response of three recently nourished beaches has generally, followed the pre-nourishment patterns of erosion. Historical hotspots and highly vulnerable areas rapidly lost the constructed subaerial beach sand while adjacent areas maintained volume near or in some cases above expected levels. Localized exchange of sediment across the shoreface, observed in profiles and time series of side scan sonar mosaics, may explain some of the deviation between theoretical and observed fill behavior.

Results from a detailed 2-dimensional numerical modelling of the coastal processes in the vicinity of 1,136,000m³ shoreface nourishment, north of Torsminde, Denmark (west coast of Jutland facing the North Sea) are presented and discussed. The morphological development of the shoreface nourishment is estimated using the ISE approach and the results compared with measured morphological changes during the winter of November 1997 to April 1998. Results from a full morphological model in an idealised test case is presented and used to illustrate some main features of the morphological development of a shoreface nourishment.

Beach nourishment introduces an abrupt perturbation to nearshore morphology. With time a readjustment of the profile to a quasi-equilibrium occurs. Characteristics of the placed sand, however, are often significantly different than the native material causing a change in morphology over greater time scales. Based on field data from various sources, this study investigates morphodynamic changes experienced by nourishment projects. In particular, theoretical changes in the equilibrium profile due to grain size are compared to field observations of profile change over time.

Using a vertical array of four streamer-type traps, the on-offshore sand transport rate was measured in the swash zone on a natural beach. The vertical distribution of the onshore transport rate showed that the maximum occurs in the bottom layer and that the rate gradually decreases with distance from the bed. The corresponding distribution for the offshore transport rate also indicated that the transport was concentrated in the bottom layer with an even sharper decrease in the rate going upwards. The total onshore and offshore transport rate integrated in the vertical direction were proportional to the second power of the onshore and offshore current velocity, respectively.

A formula to compute the net transport in the swash zone was developed based on a bedload sediment transport relationship proposed by Madsen. Predictions by the formula reproduced well field measurements of the sediment transport in the swash carried out at a sandy beach on the Pacific coast of Japan. The formula was also incorporated in a numerical model of the profile evolution in the swash under erosive conditions, and comparisons between model simulations and field data showed encouraging agreement. However, improved results are expected when the swash zone model is coupled to a surf zone model in order to properly simulate the interaction between these zones. Furthermore, the scarp which developed in some of the experimental runs was not satisfactorily modeled using the present formula. A wave-impact based sediment transport model is needed to describe the response of the scarp.

Measurements of concentrations and suspended sediment sizes above well-sorted and mixed sand beds in sinusoidal flows in a large oscillatory flow tunnel are presented. The results show that sediment grading is fundamentally important to the entrainment and suspension of sand from the bed. The depth of bed mobilised by the flow depends on the proportion of coarse sand in the bed, with a greater proportion of coarse sand resulting in a smaller mobilised depth. The distribution of the mobilised sand above the bed strongly depends on the grading. Sand is suspended to higher elevations when the proportion of fine sand is greater. Conversely, a lower proportion of fine sand leads to lower suspended sand concentrations high above the bed and higher concentrations within the sheet flow layer. These observations are intuitively sensible: the experimental results provide quantitative measures of the observations.

New experiments with graded sand were conducted in the large oscillating tunnel of Delft Hydraulics. The measured transport rates show that size gradation of sand particles has a small effect on the total net transport rates. However, the transport rates of individual grain-size fractions are strongly affected due to the interactions between the sizes. Finer sizes in the mixture are less transported as part of a mixture than as `uniform' material. Moreover, the transported sand mixture is coarser than the original sand bed. Comparing the new experimental results with three different transport models show that using the hiding/exposure correction factor of Day with the model of Ribberink leads to better predictions for the composition of the transported sand. The model of Dibajnia & Watanabe overestimates the transport rates of the two fractions under investigation. The model of Bailard overestimates the transport rates and the contribution of the fine fraction in the transported sand.

The study of morphological processes along a coastal unit of 40 km in length in the nontidal Baltic Sea revealed a substantial variation of the shore-parallel component of wave energy flux, which is assumed to be proportional to the mass of alongshore sediment transport. A reduction of flux in transport direction e.g. due to changes in the shoreline alignment or in the shape of the coast implies a jam in the longshore transport. It means that part of the sediment arriving must be deposited or diverted in offshore direction, i.e. in shore-normal transport. In order to resolve this contradiction sea surveys from different sources were used and analyzed as difference maps. It could be verified that instead of deposition at this erosive coast the reduction of longshore energy flux along a coastal stretch led to an offshore diversion of sediment out of the surf zone and accretion in 8 m to 12 m depth region, seaward of the active zone (5 m depth at the site), more than 2,000 m distant from the shore.

Beach profile modeling is still a task undergoing development. Because of the simplifications required to the present day conventional models, the applicability to real world problems remains limited. In addition, a lot of physical parameters are involved in the application process, introducing unknown percentages of uncertainty and even randomness. The main aim of this paper is, therefore, to quantify the overall uncertainty in profile dynamics predictions. More specifically, the paper will quantify the contributions of wave, current, turbulence, sediment transport, etc. predictions to this uncertainty. This will allow a scientific and engineering reading of the problem building "blocks" in a profile model. For this purpose and in order to make the exercise more useful, a widely used and tested profile model (SBEACH) has been selected. This paper is the first of a set checking all the involved parameters.

Complex principal component analysis (CPCA) was applied to beach profile data at Hasaki, Japan, to investigate morphological change on a barred beach. The first eigenfunction mainly represents a seaward bar crest migration with a period of a year, while the second one provides the combination of the shoreward migration of an accumulated area with a period of several years and that with a period of several months. The third eigenfunction represents the beach profile change predominant in the foreshore. The beach processes based on the first and the third modes have strong correlation with the incident wave energy flux although the process based on the second mode does not. The results of CPCA were compared with those of conventional principal component analysis.

The present paper treats the problem of modelling the recession of a coastal cliff in order to predict future positions of the top of the cliff, and the occurrence of major failure events. The behaviour of the cliff is predicted based on available data, by applying various probabilistic and statistical tools. The use of some of these methods is believed to be original while others draw on previous work. A first stage comprises building an adequate model for the cliff taking account of the characteristics of the recession process and the available data. In a second stage this model is used to predict the recession in future years. Predicting the probability that a particular cliff-top asset will be lost in a given time enables an assessment of risk, and can provide the basis for economical appraisal of management measures at the site under study. The methodology is applied to a wide variety of cliffs in the UK located at The Naze, Isle of Wight, Herne Bay and Scarborough. The results suggest that uncertainty in cliff predictions can be large due to the typically sparse data and uncertainty in recession processes. The models presented here are able to provide useful predictions of the recession of the top of a cliff or of the number of major landslides in a given period of time, indicating that simple statistical methods have predictive capability. Results are expressed in probabilistic rather than deterministic terms to reflect the uncertainties.

This paper discusses the quantitative interpretation of video observations of breaking waves, in support of a model to map subtidal surf zone bathymetry from time-averaged video images. Careful investigation of intra-wave time series of pixel intensities has provided a new concept to interpret pixel intensities. The new breaker intensity model isolates the roller contribution by removing the effect of persistent foam, and relates image intensities to the. roller energy E_r. Application of the technique against data obtained from experiments at Duck (NC, USA) and Egmond (The Netherlands) shows good model capabilities at both single and multiple barred beaches. Results further suggest to lower the roller dissipation coefficient β to 0.05 as compared with the commonly used value of 0.10. Application of the model to quantify subtidal beach bathymetry is subject to future research.

This paper discusses the analysis of full colour video images obtained at hourly intervals in order to quantify shoreline and nearshore hydro-sedimentary processes at three locations along the South African coast. The analysis of video images is conducted in tandem with measured wave heights and tidal levels in order to monitor beach profiles and shoreline changes associated with beach nourishment. The analysis procedures make use of pixel H-S values and neural network classification algorithms for locating the position of the waterline in video images. These techniques are shown to be more accurate and efficient than existing methods of analysis such as grey scale thresholding. Keogram images are used to visualize many of the time dependant properties of individual wave events, including run-up characteristics. Cross-correlation of the image with a directional gaussian identifies the elongated structure of the waterline. These techniques have been used to monitor the short and long-term evolution of a coastal waterline in response to storm and beach nourishment events over a 9 month period.

In December 1999, a non-intrusive directional lidar wave gage (LWG) was field-tested. The LWG uses proven lidar technology to directly measure water surface elevation. So unlike bottom-mounted gages, the surface elevation measurement from the LWG is independent of other hydrodynamic processes. The prototype wave gage consists of 4 rangefinders, where each rangefinder collected water surface elevation time series at a rate of 10 Hz. A compass and inclinometer allowed for accurate positioning of the footprints on the water surface. Throughout the field test, ground truth was collected concurrently with the LWG data using a PUV (pressure gage and current meter) mounted directly beneath the LWG. This paper describes the prototype LWG and compares the directional-spectral analysis results from the LWG data with those from the ground truth data.

USGS and NASA are investigating the impacts of hurricanes on the United States East and Gulf of Mexico coasts with the ultimate objective of improving predictive capabilities. The cornerstone of our effort is to use topographic lidar to acquire pre- and post-storm topography to quantify changes to beaches and dunes. With its rapidity of acquisition and very high density, lidar is revolutionizing the. quantification of storm-induced coastal change. Lidar surveys have been acquired for the East and Gulf coasts to serve as pre-storm baselines. Within a few days of a hurricane landfall anywhere within the study area, the impacted area will be resurveyed to detect changes. For example, during 1999, Hurricane Dennis impacted the northern North Carolina coast. Along a 70-km length of coast between Cape Hatteras and Oregon Inlet, there was large variability in the types of impacts including overwash, dune erosion, dune stability, and even accretion at the base of dunes. These types of impacts were arranged in coherent patterns that repeated along the coast over scales of tens of kilometers. Preliminary results suggest the variability is related to the influence of offshore shoals that induce longshore gradients in wave energy by wave refraction.

In May 1999 a field campaign has been carried out at Omaha beach, a macrotidal beach situated along the coastline of Normandy in North France. This beach is characterised as a typical ridge and runnel beach and four intertidal bars can be distinguished. During a period of two weeks extensive morphological and hydrodynamic data have been collected. Key objective during the field period was to understand the three-dimensional development of the ridges and runnels and to recognise the processes responsible for origin and mobility of the ridges. The paper presented here attempts to combine integrated sand transport rates, measured by fluorescent tracers injected at the seaward flanks of the two highest ridges, with the residual topographic evolution of the intertidal beach. Fair weather conditions prevailed and as a result a slight onshore migration of the ridges has been monitored (max. 12 m). The dispersion of the tracer clouds reveals a dominance of cross-shore over longshore transport. However, this can not be comfirmed by the hydrodynamic data measured seaward of the breaker line. It illustrates the limited influence shoaling wave processes have on sand transport at the injection points during conditions of onshore bar migration. A predominance of swash action is assumed to be the main acting cross-shore sediment transportation process.

Bedload transport on Spratt Sand is a result of migrating bedforms and a lack of sediment clearly limits the growth and development of these bedforms. Maximum bedload transport based on dune-tracking is observed for a combination of waves and currents, probably due to an increase in migration rates rather than an increase in ripple dimensions.

A short-term field observation was conducted to investigate hydrodynamics, sediment suspension and morphological processes on an intertidal flat. The topography on the tidal flat was found to fluctuate by about 8 cm during two weeks although long-term accumulation rate was estimated to be only about 4 cm/y. Suspended sediment concentrations were highly correlated with bed shear stress due to tidal current, wind-induced current and wave. Episodic erosions were frequently observed with high turbidities, which were caused by relatively high wave and strong current between the high water and the mean water during ebb tide.

Sand waves form a prominent regular pattern in the offshore seabed of sandy shallow seas. A 2DV morphological numerical model describing the behaviour of these sand waves is under development. The first goal is to make the new numerical model reproduce the results obtained with a linear stability analysis. We start with a constant current after which the behaviour of sand waves is investigated in a tidal environment. Furthermore, calculations using Delft3D have been made to test the general setup of the new numerical model.

Sand bars which are normal to the coastline have been described on sandy beaches from low to moderate wave energy. A generating mechanism for such bars is here proposed that is based upon the coupling between water motions and the evolving topography in case of normal wave incidence. Basically, the momentum carried by the incident waves is focused by the bars because of topographic refraction. This causes a strong onshore current on the bars and return flow at the troughs in between. Since wave stirring is typically decreasing shoreward, the onshore current over the bars produces accretion and the offshore current at the troughs produces erosion. A positive feedback is thus created between circulation and morphology and this explains the growth of the bars. A nonlinear numerical model of surf zone morphodynamics has been set up to show the growth of such bars from a number of different small amplitude initial perturbations of the alongshore uniform equilibrium.

The simple cross-shore sediment transport model put forward by Plant et al. is used to predict equilibrium beach profiles. The stability of such profiles with respect to alongshore uniform perturbations is also analyzed. Preliminary results indicate the growth of an unstable mode with a nice resemblance to observed break point bars. Although the mechanism was already suggested by previous empirical and qualitative studies, the present contribution provides a promissing quantitative model to explain the formation and migration of such bars.