Study of the Evolution of a Beach Nourishment Project Based on Computer Models J. Galofre,* FJ Montoya

Study of the evolution of a beach nourishment project based on computer models

J. Galofre,* F.J. Montoya," R. Medina* o Tarragona Coastal Service, Coastal Department, Ministry

of Public Works, Transportation and Environment,

PL Imperial Tarraco, 43005 Tarragona, Spain * Ocean and Coastal Research Group, Universidad de

Cantabria, Dpto de Ciencias y Tecnicas del Agua u del Medio Ambiente, Avda. de los Castros s/n, Santander

39005, Spain

Abstract

This paper describes the role of beach change numerical modeling in the study of the evolution of a beach nourishment project. A brief review of the processes involved in the evolution of a nourished beach is made. Available models for simulating the different processes are classified by means of the spatial and temporal scale they solve. This classification is used to determine present needs of new models and to select the appropriate existing model. Capabilities of models and application of models to the prediction of a project's performance is discussed. A set of models for the study of the evolution of a beach nourishment project based on numerical and empirical models is proposed. These models are applied to a beach nourishment project in Spain.

1 Introduction

Beach nourishment is a major area of concern in the field of coastal engineering. Prediction of the performance of a beach nourishment project has received con- siderable attention during the last decades (see Work and Dean^, as a general reference). Although the sediment processes involved in the changes of a beach are nonlinear and have great variability both in space and in time, prediction of beach evolution with numerical models has proven to be a powerful technique that can be used to assist in the determination of project design and/or eval-

250 Computer Modelling of Seas and Coastal Regions

Table 1: Beach Forcings

Short Term Middle Term Long Term Waves Platform-Currents Winter-Summer Waves

Tides Fortnight-tides Platform Currents Waves-cur rents Storms M.S.L. variations

uation. Furthermore, numerical models describing the response of beaches to different coastal forcing have become increasingly numerous and sophisticated in the recent years. At present, however, there is no model that can be used to solve all the spatial and temporal scales of variability involved in beach nourishment evolution and, consequently, different models must be used.

This paper describes the use of numerical models (either physics-based or empirical) in the evaluation of a beach nourishment project. Various types of numerical models are used. The models are classified by their spatial and temporal domains of applicability. The objective of the work is to propose a set of models for the study of a beach nourishment evolution that takes into account the different scales of variability.

2 Processes, Scales and Tools

Erosion, accretion and beach change in offshore bottom topography are controlled by wind, waves, currents, water level, nature of the sediments and its supply and coastal structures. The time scale of variability of these forcings varies from short-term (less than 15 days), middle term (15 days to 6 months) and long term (years), see Table 1.

The response of beaches to these perturbations and variable forcing can be found in a wide range of spatial scales, see Table 2.

Besides the wide range of temporal and spatial scales of variability, coastal evolution processes are often three-dimensional., Still, important aspects of the coastal behaviour can be understood and predicted on the bases of lower-dimensional models that take advantage of the circumstance that the response of a beach often exhibits a different behaviour with essentially different length scales in three mutually orthogonal space directions (vertical, cross-shore and longshore) (De Vriend*).

This has led to a range of numerical models that focus on a particular time

Computer Modelling of Seas and Coastal Regions 251

Table 2: Beach Responses

Small Scale Meso Scale Large Scale

Bed forms Profile changes Coast-line erosion Beach cusps Planform changes Coast-line accretion Bars Crescentic bars Eustatic response

Table 3: Models

Short Term Middle Term Long Term Small Scale Meso Scale Large Scale Wave propagation Wave propagation Wave propagation Forcing Tide propagation Tide propagation Tide propagation

Waves- cur rents Waves- currents Waves- currents

Profile models Profile models Parametric models Response Local sed. transp. N lines models N lines models

or space scale and a particular forcing (e.g. wave propagation models) or beach response (e.g. one line models), see Table 3.

3 Beach Evolution Models

To improve predictive models for beach response, an accurate description of the forcing is necessary. From Table 3, it can be seen that existing hydrodynamic models (wave-, tide-propagation and wave induced currents) can be used for solv- ing the forcings at almost all spatial scales of interest. The choice of a particular model should be done in relation to the response model to be used.

When selecting a response model, several physical facts must be taken into account. Cross-shore transport is very important just after thefill . A nourished beach reaches its equilibrium profile within the first year after thefil l (Kamphuis and Moir*. Several existing models can describe the post-fill evolution of the profile (usually neglecting longshore transport). However, cross-shore transport at greater time scales (month to years) remains a challenging problem that has not received a great deal of attention (Work and Dean ^).

252 Computer Modelling of Seas and Coastal Regions

Longshore sediment transport is found to be important near the shoulders of the fill in the beginning. Effects of the longshore gradients propagate after- wards into the nourished region. N-line models can represent these changes of the coastline in the mid-long term. The one-line approach imposes limitations by neglecting the influence of cross-shore transport. However, this can be over- come if the model is calibrated adequately (Hanson and Kraus*). For the long term evolution prediction, parametric models (e.g. equilibrium profile-coastline models) and statistical models (e.g. P.C.A. models) can help N-line models. Still better models are desirable.

In this study we focus on the mid-long term evolution of a beachfill .Con - sequently, profile models are not considered. One-line model is selected for the prediction of the mid-long term evolution of the coastline. A propagation model based on the mild slope equation is chosen in order to solve the wave propagation. This kind of model is assumed to be adequate for the needs of the one-line model.

A wave-induced current model is also used to better understand the results of the one-line model. Following Medina et al * a 3-way P.C.A. model is used for the long-term scale.

4 Altafulla Case Study

4.1 Field Site and Data Collection

The site of the field study is Altafulla (Fig. 1), a sandy beach located 10 km north of Tarragona and 80 km south of Barcelona, on the Mediterranean Coast of Spain. Altafulla is a half-opened beach 2.3 km long located between two capes,

"Els Munts" to the east and "Tamarit" to the west. A small river flows during storms in the middle of the beach.

There are two predominant directions of wave approach: SW and E. More than three-quarters of the deep-water waves approach Altafulla from those sec- tors. The annual average significant wave height is about 0.5 m with typical winter storm waves of Hs of about 3 meters. Tides at Altafulla are negligible.

The native beach sand has a mean diameter between D$Q — 0.12 to 0.2 mm and the beach profile slope changes from 1.2% to 2% from shoreline to bathymetric -5 meters. This bathymetric contour is considered the profile closure depth at Altafulla. This value was determined from the three-year monitoring program of profile changes. The bottom of the sea is sandy up to the 10 m bathymetric contour line.

Computer Modelling of Seas and Coastal Regions 253

27CO 2900 3100 3333 =530 3700

Figure 1: Site Location Map

4.2 Beach Nourishment

Several erosion problems occurred in the northern part of the beach. A seawall was built to prevent backshore building damage. Southward littoral drift has been theorized as a major factor in the erosion that has been witnessed on Altafulla. A beach nourishement project was undertaken in 1989. Beach nourishment started in late 1990 and was completed in 1991. The Beach nourishment works consisted of 160,000 m^ of borrowed sand volume. Borrowed sand had a median diameter averaging D$Q = 0.6 mm. A detached breakwater was also built in the middle of the beach (see Fig. 1). The breakwater was 100 m long and was placed at the 5 m bathymetric contour line.

4.3 Monitoring Program

A monitoring project is being carried out to evaluate the evolution of the fill. The monitoring program started in 1991 at the conclusion of the fill. The field program includes bathymetric beach profile and sediment samples. Profile surveys are taken every two months. Each profile is surveyed from permanent monuments landward to a depth of approximately 10 meters. Sediment samples are collected along three profiles simultaneously with the profile survey.

254 Computer Modelling of Seas and Coastal Regions

\N\\\\S\\N\N\N\N.\S\N\\\\\\\\NS\\\\\\\N\\\\\\\\\S\\\\S\\\\\\\\\\S\\N\\\N\\\\^S&K^SS^^ \ . \ \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\x\\\\\\\\\\\\ \

\\x\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\v\\\\\\\\\\\\\x\\\\\\\\\\\\\\\\\\\\\\\\\ \\V\\\\N\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

\\N\\\V\\N\\\ \\ \\N\\\\\\\

Figure 2: Wave Propagation Field

- 10 tg. - + 45

Figure 3: Wave-Induced Currents

4.4 Beach Dynamics

Beach dynamics were computed by means of the numerical models discussed pre- viously. Computations were carried out for conditions before and after the fill and the construction of the detached breakwater. Different wave height, wave period and wave approach direction were used. Typical results from the computations are shown in Figures 2 and 3. Figure 2 shows a vector plot of the wave propagation field computed by the parabolic wave propagation model. Figure 3 shows the wave induced current determined from the wave field shown in Fig. 2. The main conclusion drawn from the results of these simulations was that the beach is mainly controlled by wave refraction. Diffraction is, however, the most important feature behind the breakwater and close to the capes. The residual wave-induced current computed by adding all the wave-induced current fields with their associate probability is southward.

4.5 Beach Evolution

The basic procedure followed in this study to analyze the beach nourishment evolution was a one-line model and a statistical model. The one line model used was GENESIS (Hanson and Kraus^). The model was calibrated with the profile evolution data following the procedure described by Hanson and Kraus*. Figure

4 shows the results of the beach changes behind the breakwater obtained from the model. -i »- Initial Shoreline —* Calculated Shoreline —• Breakwater * Diffracting Groin

-100 10 20 30 40 50 60 70 90 ALONGSHORE COORDINATE (cell spacing = 25 m) Figure 4: Shoreline Change from One-line Model

The statistical model used was a Principal Component Analysis Model (three- way PCA). The method was used to objectively separate the spatial and temporal variability of the beach profile data and the sediment grain size data, as described by Medina et al [3]. Results show a seasonal variability in the beach profile data

256 Computer Modelling of Seas and Coastal Regions

and a long-term trend. This trend is inferred to be associated with the littoral drift.

5 Concluding Discussion

Numerical models of beach change are becoming more accurate and prolific and they will be increasingly used in the study of the evolution of beach nourishment.

Still, there is no model which can be used to analyze all the processes involved in the evolution of afill . Several models must be used depending on the temporal or spatial scale of interest. Furthermore, some methods which are lacking can be found in the short-, middle- and long-term scales.

Users have to select the appropriate models for a particular objective. This paper has attempted to demonstrate the utility and benefits of some numerical modeling to the study of beach nourishment evolution. Although emphasis was on numerical modeling of beach changes, it is recognized that a shore protection project will involve a wide range of techniques and tools.

References

1. Hanson, H. and N.C. Kraus, 1991. "Comparison of Shoreline Change obtained with Physical and Numerical Models". Proc. Coastal Sediments'91, ASCE, pp. 1785-1799.

2. Hanson, H. and N.C. Kraus, 1989. "GENESIS - Generalized Model for Simulating Shoreline Change", Vol. 1. Reference Manual and Users Guide, Tech. Report CERC-89-19, U.S. Army of Engineers, Waterway Experiment

Station, 247 pp. 3. Kamphuis, J.W. and J.R. Moir, 1977. "Mean Diameter Distribution of Sed-

iment Sizes before and after Artificial Beach Nourishment". Proc. Coastal Sediments77. ASCE, New York, pp 115-125.

4. Medina, R., M.A. Losada, I.J. Losada and C. Vidal, 1994. "Temporal and Spatial Relationship between Sediment Grain Size and Beach Profile". Marine Geology, 118, pp. 195-206.

5. de Vriend, H.J., 1992. "Mathematic Modelling of 3D Coastal Morphology". Proc. Short Course on Design and Reliability of Coastal Structure. 23rd I.C.C.E. Venecia, Chapter 10, 25 pp.

6. Work, P.A. and R.G. Dean, 1995. "Assessment and Prediction of Beach- Nourishment Evolution". Journal of Waterway, Port, Coastal and Ocean Engineering, Vol. 121, No. 3, May/June, pp. 182-189.