An Artificial Surfing Reef in São Pedro Do Estoril Beach, Portugal: Numerical and Physical Modeling Studies

AN ARTIFICIAL SURFING REEF IN SÃO PEDRO DO ESTORIL BEACH, PORTUGAL. NUMERICAL AND PHYSICAL MODELING STUDIES Conceição Fortes1, Maria da Graça Neves1, Ana Mendonça1, Liliana Pinheiro1, Luís Leite1, Lourenço Mendes1, Pedro Monteiro1, Artur Palha1, Pedro Bicudo2, Nuno Cardoso2 This paper describes the Numerical and Physical Modelling components of the main project entitled “Viability study of the artificial surfing reef of S. Pedro do Estoril”. These studies aim to establish the main characteristics (dimensions, orientation and location) of the artificial surfing reef for S. Pedro do Estoril and to analyze its hydrodynamic performance for different incident wave conditions. The main conclusions of both studies were that the reef solution permits the improvement of the surf conditions in the area of S. Pedro do Estoril especially for the most frequent wave in the region. However, in both studies some problems related with strong reflection and vortices in the reef’s corners where detected and should be improved in the final solution.

INTRODUCTION

Due to the importance of surf in the Cascais maritime area, the Municipality of Cascais (CMC) proposed the construction of an artificial surfing reef in São Pedro do Estoril beach, Fig. 1, to make this beach an even more relevant surf spot in Portugal. The reef aims to enhance the surf conditions in this area, fitting it also to expert surfers.

Reef

Beach

Fig. 1. São Pedro do Estoril and site of the artificial reef.

São Pedro do Estoril is a sandy beach located in Estoril, Cascais, Portugal, Fig. 1. It is 400 m long and its width varies between 25 m and 35 m with rock formations at both ends. This coast was selected due to its consistent waves, producing good surf mostly in the fall, winter and spring. In fact, it provides good surf conditions for beginners and intermediate surfers during the whole year. The site for the artificial surf reef is limited in the West end by the NeoGothic Castle and in the East side by the Avencas beach and biophysical reserve, Fig. 1. The latitude and longitude centre of the area are respectively N 30º41’35” and W 9º22’31”. The coast is composed of 20 m to 30 m high cliffs. There, the 1

2 waves break in a natural reef area, rarely used by surfers since the existing surf has poor quality. The artificial surf reef is designed to create a new surfing area, of international quality, for experienced surfers. For this purpose, the CMC, the Instituto Superior Técnico (IST), the Laboratório Nacional de Engenharia Civil (LNEC) and the Faculdade de Ciências da Universidade de Lisboa (FCUL) agreed in a protocol to produce the necessary studies for the development of strategic projects in the coast of Cascais, including the present project of an artificial reef for the Beach of São Pedro do Estoril. In particular, IST and LNEC were contracted to provide the Viability Study for the reef. This viability study includes two main components: the numerical and physical modelling studies and the environment impact evaluation. The numerical and physical modelling studies aim to establish the main characteristics (dimensions, orientation and location), of the artificial surfing reef in S. Pedro do Estoril beach, Fortes et al. (2007) and Bicudo et al. (2007a, b), and to analyze its hydrodynamic performance for different incident wave conditions. The aim of the Environmental Impact Assessment is to identify the main impacts of this project locally, nationally and even internationally, including environmental and social impacts, Custódio et al. (2008). In this paper, only the numerical and physical model studies are described. The paper starts with the establishment of the local wave regime at São Pedro do Estoril beach in order to define the design waves for which the artificial reef will be functioning. After, an analysis of the artificial surfing reef performance, based upon the numerical wave propagation model results is performed. That led to the reef main characteristics after tested on the physical model. A description of the physical model set-up, of the tests and its results are presented and discussed. Finally, a preliminary comparison between physical and numerical results will contribute to establish a final solution for the artificial surfing reef. NUMERICAL MODEL STUDY

The methodology used, Neves et al. (2007a), was: 1. Establishment of the local wave regime; 2. Based upon (1), definition of the design wave parameters for which the artificial reef will be functioning; 3. Analysis of the artificial surfing reef performance a. Definition of geometry, orientation and location; b. Propagation of the design wave parameters, using the numerical model FUNWAVE, (Kirby et al., 1998), for each reef solution; c. Determination of surfability parameters. Step 3 is repeated several times for each solution of the artificial reef. Several solutions were made by varying the slope, orientation and the location of the artificial reef. The comparison of the numerical results for different artificial reef designs permits to establish the reef characteristics to study on physical model.

3 Local wave regime

Offshore wave regime

The local wave regime is already a result of a numerical modelling study. For the characterization of the local wave regime at the São Pedro do Estoril Beach, Fortes et al. (2007a), the following steps were performed: • Step 1 – Characterization of the offshore wave regime of the São Pedro do Estoril beach, which was based upon the wave data (~13000 data sets) transferred to offshore from the Figueira da Foz buoy; • Step 2 – Transference of the offshore wave regime to the maritime area near the São Pedro do Estoril beach, using an irregular wave propagation model of the TRANSFER methodology, see Fig. 2; • Step 3 – Characterization of the local wave regime in front of the São Pedro do Estoril beach, i.e., Point 4, see Fig. 3.

F.Foz

P4

a)

b)

Fig. 2. a) TRANSFER methodology b) Establishment of the local wave regime (Point 4).

a)

b)

c)

Fig. 3. Local wave regime at S. Pedro do Estoril (point P4).

At point P4 (Fig. 2), the range of directions are between 200 and 270º being the more frequent wave directions between 220° and 255°. The significant wave heights are less than 5.5 m and the more frequent ones are in the range between 0.0 and 1.0 m. Zero crossing wave period varies between 3 s and 17 s, being the more frequent between 6 s and 7 s.

4 Design wave parameters

Based upon the local wave regime presented at Fig. 3, a definition of the design waves to be considered for the numerical wave propagation is made with and without the reef. Waves with zero crossing wave periods (between 11 s and 15 s), significant wave heights from 1.5 m to 4 m and wave directions between 220º to 255º, were considered since they are interesting conditions for surfing. Three tide levels were considered 0.3 m CD, 2.0 m CD and 2.7 m CD, which corresponds to the lower, medium and higher tide level. Analysis of the artificial surfing reef performance

This analysis begins with the definition of its main characteristics, namely geometry and location. For each reef solution, the wave characteristics around the reef are determined in order to calculate the surfability parameters. The analysis of those parameters permits to enhance the reef characteristics. Several reef solutions and locations were tested, Bicudo et al. (2007a). In this paper, only the results concerning the final solution are considered, Fig. 4. The chosen geometry corresponds to a 200 m long reef with an almost triangular cross section.

Fig. 4. Artificial reef solution. Location in the field. Geometric characteristics.

Propagation of the incident wave conditions with and without the artificial reef In order to perform the characterization of the wave conditions around the artificial surfing reef, the weakly nonlinear Boussinesq FUNWAVE model (Kirby et al., 1998) was used for the propagation of the design wave parameters. FUNWAVE (Kirby et al., 1998) is a wave propagation model based upon the extended Boussinesq equations derived by Wei et al. (1995). These equations describe the nonlinear evolution of waves over a slopping impermeable bottom without considering wave breaking. Their range of validity extends from shallow up to intermediate water depths. The model was used for regular waves, so wave height, wave period and wave direction are used. Numerical tests were performed without the reef and with different reef configurations and locations, for the different design wave

5 parameters. In this simulation, a 676 m long finite difference grid was used with a node spacing of ∆x=2.0 m, leading to 338 spatial nodes. The time step selected was ∆t=0.1s. Two sponge layers were put at the beginning and at the end of the domain, with lengths of 40 m and 20 m. The total simulation time was 600s. As an example, in the following sections, only the results for an incident wave with a period T= 11 s, direction θ= 220º, wave height, H= 2.0 m and the tide level of 2.0 m are analysed. Fig. 5 to Fig. 6 presents the wave heights, wave directions and velocity fields without and with the reef.

Fig. 5. FUNWAVE model. Wave heights: a) Without reef; b) With reef. Incident wave height of T= 11 s, θ= 220º and H= 2.0 m. Tidal level= 2.0 m (C.D.).

Fig. 6. FUNWAVE model. Velocity components: a) Without reef; b) With reef. Incident wave height of T= 11 s, θ = 220º and H= 2.0 m. Tidal level= 2.0 m (C.D.).

From Fig. 5 to Fig. 6, it is clear that the presence of the reef alters significantly the wave heights and wave directions, due to refraction and diffraction effects. There is an increase of the wave height along the reef as a consequence of the decrease of the depth. Following the increase of the wave height, the wave breaks farther from the coast. The wave directions are also modified due to the refraction effect of the reef. Moreover, there are significant modifications in the velocity components in the area surrounding reef. In fact, two vortices appear very close to the reef, which can be problematic to surf.

6 Surfability parameters The hydrodynamic analysis is based upon the surfability parameters, namely the surf lane, the Iribarren number, the peel angle, the wave height at breaking, the amplification of the wave height and the length of the wave wall. The Iribarren number ( ξ b ), for analysis of the conditions required for breaking, provides an indication about the breaker shape (terminology by Galvin (1968)) that varies between spilling ( ξ b < 0.4 ), plunging ( 0.4 < ξ b < 2 ) and surging/collapsing ( ξ b > 2 ) breaker. The peel angle, related to the break angle and the wave obliquity at the broken depth, determines the speed that the surfer must generate to stay ahead of the breaking section of the wave. Peel angles vary between 0°-90°, with zero peel angle corresponding to what is referred as a ‘close out’ where the waves break simultaneously along the entire crest. As peel angles increase the speed of breaking along the crest, which approximates the surfer velocity Vs , decreases to a speed suitable for experienced surfers. This occurs around 30°-45° with the optimal peel angle for most recreational surfers considered to be in the range 45°-65°. In this paper, only the results concerning the peel angle and Iribarren number for different incident wave conditions are considered, since those parameters are the most important surfability parameters to design an artificial reef. The peel angle (Walker, 1974) and Iribarren number (defined as Battjes, 1974) are calculated by using the following formulas, respectively: →

senα =

ξb =

c Vs s

(1) (2)

Hb L0

where α is the peel angle, Vs is the surfer downline velocity and c is the wave celerity, s is the bottom slope, H b the wave height at breakpoint and L 0 the wavelength.

7

a)

b)

c)

Fig. 7. Wave breaking area and wave heights along the wave breaking line for an incident regular wave of T= 11 s, θ = 220º: a) H= 1 m; b) H= 2 m; c) H= 3 m.

b)

a)

c)

Fig. 8. Wave breaking area and wave heights along the wave breaking line for an incident regular wave of T= 15 s, θ = 220º: a) H= 1 m; b) H= 2 m; c) H= 3 m.

Fig. 7 and Fig. 8 present the wave breaking area on the domain and the wave heights obtained from FUNWAVE results for an incident wave conditions considering T= 11 and 15 s, respectively, and H= 1 m to 3 m and θ= 220º. Fig. 9 and Fig. 10 present the Iribarren numbers and the peel angle along the wave breaking line, for the same conditions. 4

5 4. 4 3. 3 2. 2 1. 1 0. 0

T= 11 s

3.5 3 2.5 2 1.5

H=1 m

1

H=2 m

0.5

H= 3 m

0 0

a)

50

100 Breaking line [m]

150

200

T= 15 s

H=1 m H=2 m H= 3 m

0

b)

5

10 Breaking line [m]

15

20

Fig. 9. Iribarren number along the wave breaking line for an incident regular wave of θ = 220º: a) T= 11 s; b) T= 15 s.

8 35

T=11s

30

30

25

25 Peel angle [º]

Peel an gle [º]

35

20 15 H=1 m 10

15 H=1 m H=2 m

5

H=3 m

0

H=3 m

0

0

a)

20

10

H=2 m

5

T=15s

50

100

Breaking line [m]

150

200

0

b)

50

100 Breaking line [m]

150

200

Fig. 10. Peel angle number along the wave breaking line for an incident regular wave of θ = 220º: a) T= 11 s; b) T= 15 s.

From the above figures, one can notice clearly that there is only one continuous breaking line with the same orientation of the reef, especially for wave heights below 2 m, for the wave periods tested. In contrary, for H = 3 m and especially for T=15 s, there are two lines, which indicates that the breaking occurs from both side of the reef. Waves should break in a plunging manner. This means Iribarren numbers higher than 0.4 and lowers than 2, and a value of around 0.6 at the start of the wave ride for take-off. The take-off value is present in all the above cases, for both wave periods. The surfable part of the reef, for T=11s, H=2m and H=3m, is around 140m length, that corresponds to a plunging break. For lower wave heights the possible length of ride decreases to around 50m. For the wave period of T=15s, the Iribarren number increases about 25% for the three wave heights tested. The surfable part of the reef is around 140m length for H=3m, decreasing the length of ride to about 50m length for lower wave heights. The peel angle, related to the surfer skill and the wave height, expected to start at around 27° for professional surfers, varies in the tested cases from 0°30°. For the wave period of T=11s the surfable part of the reef would be around 100m for both H=1m and H=2m. For H=3m, this value decreases for less than 50m. For the wave period of T=15s, for H=1m the possible length of ride would not exceed 50m, and for both H=2m and H=3m would increase for around 150m. However, analysing the breaking figures for T=15s and H=3m, is observed two wave breaking fronts, meaning that the wave is breaking at around the same time in both extremes of the reef converging in the middle. The surfer could have the opportunity to choose between the two extremes to ride the wave, but since the plunging breaker occurs mainly in the first 150m of the reef, this would be the best option. For the wave periods tested, there is an increase of the values with the incident wave height. Moreover, the Iribarren values increases with the wave period. For the wave period of 11 s and wave heights below 3 m, one can notice that in the first 150 m of the reef the wave breaking is of a plunging type

9 ( 0.4 < ξ b < 2.0 ), which is specially adequate to surf. The same happens for H= 3 m in the first 50 m of the reef. In contrary, in the other situations, the value of the Iribarren mumber is higher than 2.0 and so the wave breaking is surging. The peel angle, always below 30º, represents an adequate velocity for very experimented surfers. Moreover, when the peel angle is below 25º, the velocity for surfers is too high and it is impossible to surf. So, in general, the good surf conditions occur for H= 1 and 2 m, in the first 100 m and 150 m, for T= 11 s and 15 s, respectively. PHYSICAL EXPERIMENTS

The aim of the physical model study was to analyze the hydrodynamic performance of the artificial reef solution for different incident wave conditions. For that, tests without and with the reef solution were performed in order to evaluate the wave breaking conditions in those situations. This study included: • The model construction (bathymetry and artificial reef) and the establishment of the test conditions; • Physical model tests without and with the reef for different incident wave conditions, to measure the wave heights in the area of the reef as and to identify the position of the wave breaking line and the type of breaking; • The analysis of the results. • Physical experiments were performed at one of LNEC’s wave tank, Fig. 11. The model scale considered was 1:30. The setup includes a 30m x30m wave tank, a wave generator, curved wave guides to propagate the wave without energy dispersion, a reduced scale bathymetry and topography of the site, and numerous instruments to observe and measure the wave breaking parameters.

Fig. 11. Overview of the physical model without and with the artificial reef

Test conditions

The tide levels and the incident wave conditions adopted on the physical model tests were defined by Bicudo et al. (2007a,b) based upon the local wave regime, Fig. 3, and the results of the wave propagation from offshore to the water

10 depth -10 m CD (which corresponds to the depth for which the wave makers are located). Some additional wave conditions, which are not frequent, such as high wave periods for different wave heights, were considered since they are interesting conditions for surfing. So, the tests conditions adopted are, for regular waves: • wave heights ranging from 1.0 m to 6.0 m, • wave directions of 220º and 235º, Fig. 12a, • wave period of 11 s, 15 s and 19 s, • tide levels of 0.3 m,+2.0 m and +3.7 m CD. Additionally, irregular wave tests were made for incident wave with peak period of 11 s and 15 s and significant wave heights of 3 m and 4 m. At each regular wave test, 7 wave gauges, Fig. 12b, measure the free surface elevation. The first one is located near the wave maker to measure the incident wave. The other six wave gauges were positioned along a line, as shown in Fig. 12b. This position of this line varied according the positions P1 to P6, Fig. 12c. All tests were photographed and filmed to identify the location of the breaking line and the type of breaking. 1 2 3 4 56

235

220

Total of 36 positions

Fig. 12. a) Model bathymetry and wave directions; b) Gauges; c) Gauges positions.

Results

Fig. 13 a, b present the wave heights obtained in several points around the area of the reef, for an incident wave of T= 11 s, H= 2 m and wave direction of 220º, with and without the reef. Fig. 13 c presents the wave heights along the line marked in Fig. 13 a, b as L4 line. The tidal level was +2.0 m (CD). Fig. 14 presents a photo and a wave breaking line for the same incident wave conditions. There are presented as well the wave breaking lines corresponding to H= 1 and 3 m. The following analysis is done only for regular waves.

11

a)

b) H=2.0m, T=11s, 220º 3.0 2.5 H (m)

2.0 1.5 1.0 0.5

without reef

with reef

0.0 0

2

4

6

8

10

12

Gauges position (m)

c) Fig. 13. Wave heights. T= 11 s, H= 2 m and θ = 220º. Tidal level=+2.0 m CD.a) without the reef; b) with the reef; c) along the L4 line.

H= 1 m H= 2 m H= 3 m H= 4 m H= 5 m H= 6 m

a)

b)

Fig. 14. T=11 s, θ = 220º. Tidal level=+2.0 m CD. a) Wave breaking lines. H=1 to 3 m; b) Photo: H= 2 m.

From the physical model tests without the reef it was clear that the reef will not have any influence on the cases where wave breaking occurs before the reef. In the other cases, it may have some influence. From the physical model tests with the reef, it was concluded that the reef is good for surfing for some incident wave conditions. In fact, the reef presents good surfing condition, for progressively higher wave heights as the tidal level increases and for the low and medium tidal level. Moreover, for the tested periods and directions, the reef enhances the surfing conditions for the frequent wave height (1 to 3 m).

12 In general, the reef has better results for wave directions from 235º than from 220º, with an increase of the range of wave heights which are better for surf. The reef presents the best results for wave periods of 11 s and 19 s for the lower wave heights. In the other cases, or the wave breaks before the reef (higher wave heights), after the reef (small wave heights), or has no good conditions for surf because it breaks at the same time. Comparison Between Physical And Numerical Results

The physical model results are compared with the FUNWAVE numerical results, for similar incident wave conditions tested (wave periods of 11 s and 15 s, wave heights below 4 m, the wave direction of 220º and tidal level of 2.0 m CD). First of all, during the tests it was observed a strong reflexion on the vertical part reef and the occurrences of vortices in the corners of the reef, which was already expected with the numerical model, Fig. 6. These are one of the aspects that should be improved in the final solution of the reef. Secondly, comparing the Fig. 6 with the breaking area from the numerical model and the photo of physical model was observed in both the occurrence of two different breaking lines that are jointed in one single point. Finally, when comparing the physical model series and the corresponding one form numerical model, Fig. 15, the shape and the magnitude of the free surface elevation is quite similar as well as the wave heights. 2 FUNWAVE S7 EXPERIMENTAL S7

Water Surface ELevation [m]

1.5

1

0.5

0

-0.5 350

370

390

410

430

450

Time [s]

Fig. 15. FUNWAVE and experimental results. T=11 s, H=2 to 6 m and θ = 235º. Tidal level=+2.0 m CD.

However, there are some differences: FUNWAVE shows always higher values which is partially due to the empirical wave breaking technique that is implemented. A sensitivity analysis on the FUNWAVE parameters that control the wave breaking can improve this results and make FUNWAVE a good tool to achieve the final solution of the artificial reef. CONCLUSIONS

This paper describes the Numerical and Physical Modelling components of the main project entitled “Viability study of the artificial surf reef of S. Pedro do Estoril”. The numerical modelling studies established the main characteristics (dimensions, orientation and location) of the artificial surfing reef and confirmed

13 a good performance for the chosen design for the reef. The physical modelling confirmed the good performance of the reef for some of the incident wave conditions. However, the physical modelling showed some problems on the shape of the reef, namely the strong reflection and the existence of vortices in its corners. Those aspects should be improved in the final solution. ACKNOWLEDGEMENTS

Financial support of the office of the Cascais Municipality and of FCT projects PTDC/ECM/67411/2006 and PTDC/ECM/66516/2006 are acknowledged. Special thanks to João Santos, Rui Capitão, Branca Branco, Carlos Galvão, Odair Maurício, Ana Leandro, Ana Passarinho, Metcheld van voorde for their help on the tests. REFERENCES

Battjes, J.A. 1974. Surf similarity. Proc 14th International Conference on Coastal Engineering, 466-479. Bicudo, P.; Cardoso, N. 2007a. Numerical Modelling for the Wave Generators of the Physical Model, CMC/IST/FCUL/LNEC-MOD_IST/06. Bicudo, P.; Cardoso, N. 2007b. Numerical Modelling for the Orientation and Slope of the Reef’s top. CMC/IST/FCUL/LNEC-MOD_IST/07. Custódio, A. M.; Bicudo, P. ; Nogueira, M. J.; Figueiredo, P. M. 2008. Environmental Impact Evaluation of Artificial Surf Reef in S. Pedro do Estoril. CMC/IST/FCUL/LNEC-MOD_IST/08. Fortes C.J., Capitão, R., Neves, M.G., Monteiro, P.P., Mendes, L.S. 2007a. Viability Study of an artificial Surf Reef in the S. Pedro Beach. Numerical and physical modelling. Wave regimes. LNEC Rep. 172/07. April. Fortes, C.J.E.M.; Neves, M.G.; Mendes, L.; Monteiro, P.; Palha, A. 2007b. Viability Study of an artificial Surf Reef in the S. Pedro Beach. IV CPGC, Madeira, October. Galvin, C.J. 1968. Breaker type classification on three laboratory beaches. Journal of Geophysical Research, 73(12), 3651-3659. Kirby, J.T.; Wei, G.; Chen, Q.; Kennedy, A. B.; Dalrymple, R. A. 1998. FUNWAVE 1.0 – Fully Nonlinear Boussinesq Wave Model Documentation and User’s Manual. Report No.CACR-98-06, University of Delaware. Neves, M.G.; Fortes C.J.; Mendes, L.S.; Monteiro, P.P 2007a. Work methodologies for the artificial reef of S. Pedro do Estoril, Portugal. LNEC Report 182/07-NPE. Walker, J.R. 1974. Recreational Surf Parameters. Tech. rept. 30. University of Hawaii, James K.K. Look Laboratory of Oceanographic Engineering. Wei et al. 1995. Wei, G.; Kirby, J.T.; Grilli, S.T.; Subramanya, R. A fully nonlinear Boussinesq model for surface waves. Part 1. Highly nonlinear unsteady Waves, J. Fluid Mech. v. 294, p. 71-92, 1995.

14 KEYWORDS – ICCE 2008 PAPER TITLE: AN ARTIFICIAL SURFING REEF IN SÃO PEDRO DO ESTORIL BEACH, PORTUGAL. NUMERICAL AND PHYSICAL MODELING STUDIES Authors: Conceição Juana Fortes, Maria da Graça Neves, Ana Mendonça, Liliana Pinheiro, Lourenço Mendes, Pedro Monteiro, Artur Palha, Pedro Bicudo, Nuno Cardoso Abstract number: 1112 Artificial surfing reef S. Pedro do Estoril Numerical modeling Physical modeling Surfability parameters FUNWAVE model