2008 Taiwan-Polish Joint Seminar on Coastal Protection

Simulation of Shoreline Change behind a Submerged Permeable Breakwater

Chin-Wen Hung1 Hong-Bin Chen2 Ching-Piao Tsai1

ABSTRACT
This paper reports the numerical simulation of the wave height variation and the shoreline change in lee of a permeable submerged breakwater. Firstly, the time-dependent mild slope equation involving the porous parameters of the submerged structure is adopted for the computation of the wave field. Secondly, the longshore sediment transportation is calculated by the breaking wave energy flux behind the structure. The long-term shoreline change model is then established based on the calculation of the continuity equation of the sediment. For the verification of the numerical results, the experiments in a two-dimensional movable bed wave basin were conducted. The numerical results of the wave height variations and the shoreline change show well in agreement with the experimental results. Keywords: Submerged permeable breakwater; Shoreline change; Time- dependent mild-slope equation; Continuity equation of sediment.

1. INTRODUCTION
When waves transmit over a submerged permeable breakwater the wave heights may decay due to wave energy dissipation; and the circulation current systems behind the structure is likely to make the sand deposit. The submerged breakwater consists of rubble mounds in which the porous medium has the function of ecological restoration. This submerged and permeable structure has less impact on the coastal environment than a traditional offshore breakwater, thus it is regarded increasingly by costal engineers for the coastal defense. About the issue of the submerged permeable breakwater, most of literature has been only paid attention to the wave transformation, rather than the prediction of the shoreline change in lee of the submerged offshore breakwater. In the past, many numerical models were proposed to simulate the shoreline changes. These models include two-dimensional line models or three-dimensional models for the topography variation. For example, Pelnard-Considere (1956) proposed the one-line contour model, Bakker (1968) developed the multiple-line contour model, and Willis (1980) considered the so-called 1.5 lines contour model. Perlin and Dean (1978) developed the N-line model, which is the extension of one- and two- line contour models to an arbitrary number of contour lines. These line models do not need compute the wave and current fields so that they have the computational efficiency. Later,
1 National Chung Hsing University, Taichung, Taiwan, R.O.C. 2 Kainan University, Taoyuan, Taiwan, R.O.C. Email: cptsai@dragon.nchu.edu.tw (Ching-Piao Tsai) B-49

2008 Taiwan-Polish Joint Seminar on Coastal Protection

Watanabe et al. (1986) proposed a three-dimensional model to calculate short-term shoreline changes and coastal topography variations by the continuity equation of the sediment based on the quantities of wave and current fields. Besides, GENESIS proposed by Hanson et al. (1989) was often adopted to the prediction of the shoreline changes due to coastal structures. This model considered wave transformations including shoaling, refraction and diffraction. Most of the late applications of the GENESIS model were combined with RCPWAVE (Ebersole et al., 1986) model. However, up until now, it seems that no effective models were developed to simulate the shoreline changes behind the submerged permeable breakwater. This study attempts to establish the numerical model to simulate the wave height variations and the shoreline change due to the submerged porous structures. For this purpose, the calculations of wave field around the structure are based on the time-dependent mild slope equation which the porous properties of the structure are considered. Then the long-term shoreline changes are calculated from the continuity equation of the sediment which the modified longshore sediment transportation due to the coastal structures is considered.

2. COMPUTATION OF WAVE FIELD

Considering the waves progress over a layer of porous structure on a seabed, referring to Tsai et al. (2006), the time-dependent mild slope equations involving the parameters of the submerged porous mediums are expressed as follows: Q + C 2 h = 0 (1) t
1 (2) + h [nQ ] = 0 t n in which h =(/x, /y) is the horizontal gradient operator, t is the time, is the water surface elevation, Q is the flow rate per width and C is the wave celerity. The wave celerity C and n are given as:
C2 = g [SH] k [CH]

(3) (4)

n=

1 A 1 + 2 [SH] [CH]

and
2 2 (sinhkhb ) + b (coshkhb ) A = khb ( b b ) + kh b 2

(5) (6) (7)

[SH] = bsinh(khb )cosh(kh) + bcosh(khb )sinh(kh) [CH] = bsinh(khb )sinh(kh) + bcosh(khb )cosh(kh)

where h is the water depth above the porous medium, hb is the thickness of the porous

B-50

2008 Taiwan-Polish Joint Seminar on Coastal Protection

medium, k is the complex wave number, b is the porosity of the medium and

b =Cr-if, Cr is an inertia coefficient of the flow, and f is the linearized friction factor.
For the computation of the wave field in the surf zone, referring to Watanabe et al. (1986), the energy dissipation due to wave breaking are added in eq. (1) and expressed as:

Q 1 2 + C h (n) + f DQ = 0 t n where fD is the energy dissipation factor and is expressed as (Mizuguchi, 1980):
f D = 2.5tan g Q ( - 1) h Qr

(8)

(9)

is the amplitude of the in which tan is the average beach slope in the surf zone, Q
flow rate, and Qr = gh3 is the maximum flow rate amplitude in the wave recovery zone. It needs to estimate the location of wave breaking in the numerical calculation for eq. (8). The breaking criterion of Goda (1975) is used in the present model, given as:
Hb h 1 + 15tan 4 / 3 = 0.17 1 exp 1.5 L0 L0

(10)

in which Hb is the height of the breaking wave, L0 is the wavelength in deep water. The computed wave height at a location is compared with the value calculated from the right-hand side of eq. (10) to obtain the location of wave breaking. A finite-difference scheme is adopted in this study. The provision of numerical stability is satisfied with Courant stability criterion, x/C t , where x is the spacing grid in the x-direction and t is the time interval. According to Watanabe et al. (1986), the boundary conditions can be classified into three types. They are internal boundaries with arbitrary reflectivity, nonreflective virtual boundaries and offshore open boundaries. The first one is assumed that partial standing waves exist in front of a reflective boundary. The second one is the side boundaries established for performing the numerical analysis, which are treated as having no reflection that waves pass freely through them. And the third one is initiated the calculation by the given incident wave conditions from the offshore open boundary. This boundary exists not only incident waves but outgoing wave due to the reflection from structures and the shoreline. It is noted that the submerged permeable breakwater is regarded as a boundary of a part of the porous medium on the seabed. By contrast, the emerged offshore permeable

B-51

2008 Taiwan-Polish Joint Seminar on Coastal Protection

breakwater is treated as an internal boundary with reflectivity.

3. COMPUTATION OF SHORELINE CHANGES

3.1 Longshore Sediment Transport Rate
The longshore sediment transport rate is the major factor influencing on the evolution of the shoreline change. Komar and Inman (1970) related the volumetric longshore sediment transport rate, Ql, with the incident breaking wave energy flux, and expressed as:

Ql =

K ( ECg )b sinbcosb ( s - ) g

(11)

where ( EC g ) b is the incident wave energy flux evaluated at the breaker line, b is the breaking wave angle to the shoreline, the coefficient K is an empirical coefficient, is the porosity of sediment, s is the density of the sediment, and is the water density. Only the quantification of the breaking height was considered to estimate the longshore transport sediment rate in eq. (11). However, it has to consider the contribution of the wave diffraction effect in case of the existence of the coastal structures. Ozasa and Brampton (1980) modeled the longshore sediment rate in the vicinity of the coastal structure and verified the prediction expression. The empirical formula was given as

sin 2 K cos (H/x) ] Ql = H b2C gb [ K 1 bs 2 bs b

(12)

in which Hb is breaking wave height, Cgb is the group velocity, bs is the breaking wave angle to the shoreline line. The term (H/x) b was introduced to describe the sediment transport arising from the breaking wave height alongshore caused by and K in equation are diffraction at structures. The dimensionless parameters K 1 2 expressed as:
= K / [16( / 1)(1 )(1.416)5 / 2 ] K 1 1 s
= K / [8( / 1)(1 )tan(1.416)5 / 2 ] K 2 2 s

(13a) (13b)

in which K1 and K2 are the empirical coefficients so called the transport parameters. In the present computations, the parameter K1 is fixed in a value of 0.77 as suggested by Komar and Inman (1970), and K2 is adjusted to achieve the best agreement between calculated and observed shoreline features near the major structures. The governing equations of the long-term shoreline changes are based on the continuity equation of sediment, expressed as y 1 Q ( ) + ( )( l q) = 0 (14) Ds x t
The quantity q denotes the net rate of sediment from the landward and seaward boundaries, x-axis is taken to be parallel the alongshore direction, and y-axis points

B-52

2008 Taiwan-Polish Joint Seminar on Coastal Protection

offshore. The quantity Ds means the critical depth, beyond which the bottom profile does not change. Hellermeier (1983) expressed it as Ds = [2.3 10.9( H 0 / L0 )]H 0 (15) where H0 and L0 are the deep water wave height and wavelength. In finite-difference form, the equation is given as t Q y = ( )( (16) - q) Ds x In the present computations, the fine meshes are performed for the wave field to reach the required accuracy. However, it will cost much computational time but have less efficiency in the calculation of the shoreline change if it also adopted the same built meshes of the wave field. Hence, the enlarged grids were formed to calculate the shore change. The technique of moving average was also adopted in the calculation.

4. MODEL TESTS
To verify the numerical results, a model test was conducted in this study. A 1:50 scaled model was used in a wave basin for the investigation the wave heights and shoreline change. The test conditions are listed in Table 1. Fig. 1 shows the arrangement of the model test. The tests were conducted for two cases. One was the case of the submerged breakwater that the top of the breakwater is very close the water surface, i.e. ds/d 0. The other was the situation of the emerged offshore breakwater which its breakwater height was high enough that the wave overtopping did not occur. To measure the wave heights variation around the breakwaters, 56 wave gauges were installed, as shown in Fig. 2. It notes that the X- and Y- axis used in the experimental are different with the numerical analysis. The wave heights were measured for every 80cm interval distance, i.e., in Y= -240 cm, -160 cm, -80 cm, 0, 80 cm, 160 cm and 240 cm, in which Y= 0m is the middle of the breakwaters. In X direction, it was measured for every 50cm interval points. As shown in Fig. 2, the position of permeable breakwater is located at 320cm to 360cm in the X direction and 80cm to -80cm in the Y direction.

B-53

2008 Taiwan-Polish Joint Seminar on Coastal Protection

Table 1 Test conditions Prototype Wave Period (sec) 8 3 80 20 70 1/20 4 Model 1.13 0.06 1.6 0.4 1.4 1/20 0.08

Related Wave height (m) Breakwater length (m) Breakwater width (m) Offshore distance (m) Bed slope Breakwater water depth

(a)

(b) Fig1. The arrangement of the model test, (a) side view (b) plan view.

B-54

2008 Taiwan-Polish Joint Seminar on Coastal Protection

Fig2. Measured points for wave heights

5. RESULTS AND COMPARISONS 5.1 Wave height variations

Fig. 3 shows that the numerical solutions of wave height are well in agreement with the experimental results. The solid and circle points are represented the measured wave heights of Y section in the positive (0 Y 240cm) and negative direction (-240 cm Y 0) respectively. Figs. 3(a) and 4(b) are the emerged and submerged cases of permeable offshore breakwaters, respectively. It shows that the measured wave heights are slightly larger than those of the computed results. The errors are perhaps induced from the quantities of the permeable parameters of the porous structure. It notes that b = 0.35, f = 1, and Cr = 1 are set in the computations, which the value of b is obtained by the experiment, but the quantities of f and Cr are assumed. Fig. 4 indicates the comparisons of the computed wave height variations of the emerged permeable offshore breakwater (ds/d >0) and submerged permeable breakwater (ds/d 0). The results show that the wave height variation at the section Y= 240cm are almost not influenced by the breakwaters. At the sections Y= 0 and 80cm, it can be seen that the wave heights behind the submerged breakwater are larger then those of the offshore breakwater due to the waves transmission over the submerged breakwater. The fluctuation due to wave reflection from the breakwater can also be found obviously at the section Y= 0 cm.

B-55

2008 Taiwan-Polish Joint Seminar on Coastal Protection

Fig3. Comparisons of the computed and experimental wave height (a) permeable offshore breakwater (b) permeable submerged breakwater

5.2 Shoreline changes

The experimental and computational results of the shoreline changes in lee of the permeable breakwaters are shown in Fig. 5, in which (a) is for the emerged permeable breakwater and (b) is for the submerged permeable breakwater. The results show that the shape of salient shoreline was formed behind both breakwaters under the wave conditions used in the experiment. The simulated shoreline changes show good tendency with the experimental results. The computed wave heights behind the breakwater were slightly underestimated (as seen in Fig. 5) that possibly induce the simulated salients are slightly overestimated. The errors maybe due to the uncertainties of the permeable parameters of the porous medium; it has to be suggested from the calibration of the experiments in the study. It was found from experimental and computational results that the smaller salient beach was formed in lee of the submerged case of the permeable breakwater.

B-56

2008 Taiwan-Polish Joint Seminar on Coastal Protection

Fig4. Comparisons of the numerical solutions of submerged and offshore breakwaters.

Fig5. Comparisons of experimental and computed shoreline changes for (a) permeable emerged breakwater (b) permeable submerged breakwater

X(cm )

X(cm)

6. CONCLUSION
A simulation model for the shoreline changes in lee of the submerged permeable breakwater was presented in this paper. The model includes the calculation of the wave height and the long-term shoreline changes. The porous properties were involved in the computation of the wave height. The effect of the wave diffraction to the breaking wave energy flux behind the submerged breakwater was considered in the calculation of the continuity equation of the sediment. The experiments were also conducted in the study to verify the numerical solutions. The larger wave height and the smaller salient shape of beach in lee of the submerged permeable breakwater was found, when compared to the case of the emerged offshore permeable breakwater. The comparisons showed that the numerical solutions are well in agreement with the experimental results.

ACKNOWLEDGMENT
The authors are grateful to National Science Council, R.O.C., for supporting this study under grant No. NSC95-2625-Z-005-008.

REFERCNCE

B-57

2008 Taiwan-Polish Joint Seminar on Coastal Protection

[1] Bakker, W. T., 1978. The Dynamic of a Coast with a Groin System. Proc. 11th Conf. On Coastal Eng., ASCE, pp. 492-517. [2] Ebersole, B. A., Cialone, M. A., and Prater, M. D., 1986. Regional Coastal Processes Numerical Modeling System. Report 1, RCPWAVE-A linearwave propogation model for engineering use, Technical Report CERC-86-4. Coastal Engineering Res. Center, U.S. Army Corps of Engineers, Waterways Experiment Station, MS. [3] Goda, Y., 1975. Irregular Wave Deformation in the Surf zone. Coastal Engineering in Japan, ASCE, 18, pp. 13-26. [4] Hanson, H., and Kraus. N. C., 1989. GENESIS- Generalized Model for Simulate Shoreline Change. Report 1, Technical Reference. Technical Report CERC-89-19. Coastal Engineering Res. Center, U.S. Army Corps of Engineers, Waterways Experiment Station, MS. [5] Hellermeier, R. J., 1983. Sand Transport Limits in Coastal Structure Design. Proc. Coastal structure 83, ASCE, pp. 703-716. [6] Komar, P. H., and Inman, D. L., 1970. Longshore Sand Transport on Beaches. Journal of Geophysical Research 75,pp. 5914-5927. [7] Mizuguchi, M., 1980. A Heuristic Model of Wave Height Distribution in Surf Zone. Proc. 17th Coastal Eng. Conf., ASCE, pp. 278-289. [8] Ozasa, H., and Brampton, A. H., 1980. Mathematical Modeling of Beaches Backed by Seawells. Coastal Engineering 14, pp. 47-64. [9] Pelnard-Considere, R. ,1956. Essai de theorie de l'evolution des forms de vivage en plages de sable et de dalets. IVeme Journees de l'Hydraulique, Les Energies de la Mer, Question III, Report, No. 1, pp. 289-298. [10] Perlin, M., and Dean, R.G., 1978. A Numerical Model to Simulate Sediment Transport in Vicinity of Coastal Structures. Miscellaneous Report 83-10, Coastal Engineering Res. Center, U.S. Army Corps of Engineers, VA. [11] Tsai, C.P., Chen, H.B., and Lee, F.C., 2006. Wave Transformation over Submerged Permeable Breakwater on Porous Bottom. Ocean Engineering 33, pp. 1623-1643. [12] Watanabe, A., and Maruyama, K., 1986. Numerical Modeling of Nearshore Wave Field under Combined Refraction, Diffraction and Breaking. Coastal Engineering in Japan 29, pp. 19-39. [13] Willis, D.H., 1980. Sediment Load under Waves and Current. Proc. 16th Conf. On Coastal Eng., ASCE, pp. 1626-1637.

B-58