MODEL TESTS OF THE PROPOSED P.E.P. REEF
INSTALLATION AT VERO BEACH, FLORIDA
Robert G. Dean
Albert E. Browder
Matthew S. Goodrich
Don G. Donaldson
August 17, 1994
Board of County Commissioners
Indian River County, Florida
MODEL TESTS OF THE PROPOSED P.E.P. REEF INSTALLATION
AT VERO BEACH, FLORIDA
August 17, 1994
Board of County Commissioners
Indian River County, Florida
Report Prepared by:
Robert G. Dean1
Albert E. Browder1
Matthew S. Goodrich1
Don G. Donaldson2
2 Indian River County, Vero Beach, Florida
1Department of Coastal and Oceanographic Engineering, University of Florida,
Gainesville, Florida 32611
TABLE OF CONTENTS
LIST OF FIGURES ........................................
FOREW ORD ............................................
OBJECTIVES OF A SUBMERGED BREAKWATER INSTALLATION .
THE PLANNED INSTALLATION SITE .....................
THE PLANNED DESIGN ...............................
OBJECTIVES OF A SUBMERGED BREAKWATER
............. .... ....... .. .5
EFFECTS OF A SUBMERGED BREAKWATER .........................
(1) Reduction of Wave Height .................................
(2) Modification of the Nearshore Current System ................... .
(3) Modification of the Cross-Shore Sediment Transport System ...........
(4) Modification of the Longshore Sediment Transport System ............
ROLE OF SECONDARY FLOWS ...................................
LABORATORY TESTING AND RESULTS ..................................
MOVABLE BED STUDIES .......................................
FIXED BED STUDIES ...........................................
A) Initial Fixed Bed Tests ...................................
(1) Wave Height Measurements ..........................
(2) Current Measurements ..............................
B) Additional Fixed Bed Tests .................................
DISCUSSION OF MODEL RESULTS ................................
SUMMARY AND RECOMMENDATIONS ...................................
SUM M ARY ..................................................
I BOTTOM DROGUE TRAJECTORY/VELOCITY RAW DATA
I SAMPLE CALCULATIONS FOR P.E.P. REEF MODEL WAVE
LIST OF FIGURES
Net Transport, Vero Beach, FL., Positive Transport is Southward, Walton (1976) ...... 2
Significant Wave Heights, HI, for Vero Beach, CDN Data 1986-1989 ............... 3
Typical Profile at the Center of the Project Site, Monument R-81 ................. 3
Long Term Shoreline Changes for Vero Beach, FL. Based on DEP Monuments R-77
through R-84 ................................................... 4
Schematic of Wave Diffraction Around a Breakwater .............
Definition Sketch ....................................
Wave Basin Schematic .................................
Reef Unit Arrangements Tested in Model .....................
Cross-Shore Wave Height Profile Lines ......................
Cross-Shore Wave Height Profiles for Freeboard Ratio f/h = 0.0 .....
Wave Height Transmission Coefficients ......................
Wave Height Transmission Coefficients versus Freeboard, f, divided by
Incident Wave Height, H ...............................
Bottom Drogue Trajectories, f/h = 0.0 ......................
Reef Arrangements for Additional Model Testing ................
Control and Type A Case, f/h = 0.0. Wavemaker is to the right. Units in ft/s ........ I
B and C Cases, f/h = 0.0. Wavemaker is to the right. Units in ft/s .............. II
Type D Case, f/h = 0.0. Wavemaker is to the right. Units in ft/s ............... III
Control and Type A Case, f/h = -0.2. Wavemaker is to the right. Units in ft/s ...... IV
B and C Cases, f/h = -0.2. Wavemaker is to the right. Units in ft/s ............. V
Type D Case, f/h = -0.2. Wavemaker is to the right. Units in ft/s ............... VI
Control and Type A Case, f/h = -0.4. Wavemaker is to the right. Units in ft/s ..... VII
B and C Cases, f/h = -0.4. Wavemaker is to the right. Units in ft/s ............ VIII
Type D Case, f/h = -0.4. Wavemaker is to the right. Units in ft/s ............... IX
Type E Case, f/h = 0.0. Offset Distances = 2w and 4w. Units in ft/s ............. X
Type E Case, f/h = 0.0. Offset Distance = 6w. Units in ft/s .................. XI
Type F Case, f/h = 0.0. Offset Distances = 2w and 4w. Units in ft/s ............ XI
Type F Case, f/h = 0.0. Offset Distance = 6w. Units in ft/s .................. XII
Type G Case, f/h = 0.0. Offset Distance = 4w. Units in ft/s ................. XIV
Don G. Donaldson
Indian River County
Indian River County and the City of Vero Beach have for many years considered a shore protection project at Vero
Beach. Sand nourishment has been proposed twice, however, the local citizens have rejected funding sand
nourishment because of environmental concerns. The county has now shifted its focus away from a project that
provides protection from high energy erosional events. The revised project is to provide some measure of storm
protection without adversely impacting the local environment.
The county desires a project that provides sediment retention capabilities and that will reduce the incident wave
energy. Various types of sediment retention projects were considered, including beach dewatering, emergent
breakwaters, and submerged breakwaters. The submerged breakwater unit manufactured by American Coastal
Engineering, Inc., the Prefabricated Erosion Prevention (P.E.P.) Reef, was selected for its apparent simple
construction and installation requirements. Also, given that the P.E.P. Reef is the only prefabricated submerged
breakwater installed in Florida, at the time of its selection, the County felt it would have a better chance of success
by using a local product.
OBJECTIVES OF A SUBMERGED BREAKWATER INSTALLATION
The submerged breakwater is not a primary storm protection device. This is because it is subject to continuous
wave overtopping and its effect is diminished by the increased water depth associated with a storm condition. The
submerged breakwater can only enhance the landward shore protection features by providing limited wave energy
reduction. Therefore, the objective of the Vero Beach submerged breakwater is to reduce the incident storm wave
energy sufficiently to retain the existing beach profile long enough to lessen the damage to upland structures. In
addition, during less energetic storms the breakwater is supposed to trap sand so that dune protection is enhanced.
The County has had difficulty in quantifying its objective. This is due in part to a lack of information regarding
equilibrium conditions for submerged breakwaters and a lack of numerical model routines that can account for the
hydrodynamic impacts associated with a submerged breakwater. The county contracted the University of Florida
Coastal and Oceanographic Engineering Department to run various physical model tests to help establish the best
configuration of the Reef for the Vero Beach project. The purpose of the study is to determine what the nearshore
hydrodynamic regime and wave energy reduction are for various configurations of the P.E.P. Reef breakwater.
A submerged breakwater must balance wave energy reduction with the hydrodynamic impacts. The hydrodynamic
regime associated with a particular submerged breakwater configuration is dependent upon several factors, including
the dimensions of the breakwater, the depth of the water, the distance offshore, incident wave characteristics, and
the planview configuration. This study has attempted to resolve just two of these factors, the depth of unit
placement and the planview configuration given constant wave conditions that are normally incident and the average
beach profile for the project location.
MODEL TESTS OF THE PROPOSED PEP REEF INSTALLATION
AT VERO BEACH, FLORIDA
The model test results presented herein were conducted to provide design guidance for the proposed Vero Beach
installation of approximately 4,000 feet of P.E.P. Reef. The model studies were funded by Indian River County
and were carried out in the model basin of the Department of Coastal and Oceanographic Engineering (COE) of
the University of Florida. The general purpose of these tests was to evaluate the effects caused by the Reef
structure on the wave and current system in the vicinity of the Reef. The tests were conducted at an undistorted
length scale of 1:16, which for Froude scaling, yields a time scale of 1:4. The tests commenced with and limited
results were obtained from a movable (sand) bed; however, the main body of tests was conducted on a fixed
(concrete) bed. Standard wave gages were used to quantify the wave field and several approaches were employed
to define the current patterns. The remainder of this report is organized as follows. The main body focuses on
the relevant general site conditions and the model test arrangements and results. The summary and
recommendations follow these tests. Appendix I presents the raw data from 22 of the 27 tests conducted.
Appendix H provides example calculations for the comparison of the model results with the equations developed
by Ahrens (1987).
THE PLANNED INSTALLATION SITE
The planned site for the P.E.P. Reef installation is offshore Vero Beach, Florida in water depths ranging from 8
to 10 feet relative to the National Geodetic Vertical Datum (NGVD). The Reef is to be located from 200 to 300
feet from the shoreline. Various hydrographic factors of relevance to the performance of this installation are
The spring and mean tidal ranges for Vero Beach are approximately 3.58 and 3.30 feet, respectively. The longshore
sediment transport characteristics have been developed by Walton (1976) based on analysis of wave data from ship
observations. The average longshore sediment transport varies seasonally with southerly and northerly directed
transport occurring during the winter months and summer months, respectively. The results from Walton (1976)
are shown in Figure 1 where it is seen that on an average annual basis, the seasonal changes in direction occur
between February and March and between August and September. The net and gross sediment transport can be
determined from Figure 1 as 36,000 yds3 per year (southward) and 645,000 yds3 per year, respectively. Other
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Figure 1 Net Transport, Vero Beach, FL., Positive Transport is Southward, Walton (1976).
estimates of net southerly longshore transport in the general vicinity are much higher, for example the U.S. Army
Corps of Engineers (1971) has estimated the net annual southerly longshore sediment transport at Port Canaveral
to be 350,000 yds3 per year and 200,000 to 250,000 yds3 per year at Fort Pierce Inlet. Based on other studies,
Coastal Tech (1991) has adopted the net annual southerly directed longshore sediment transport as 157,000 yds3 at
Sebastian Inlet, some 16 miles to the north.
The wave characteristics at Vero Beach can be determined from a number of sources. The Department of Coastal
and Oceanographic Engineering has maintained a wave gage off Vero Beach for several years as part of the Coastal
Data Network (CDN) program. The seasonal variation in wave height is shown in Figure 2 from the CDN for the
period 1986 to 1989.
There are two natural reefs off Vero Beach. The inshore reef is located approximately 300 feet from the shoreline
and the seaward reef, while poorly defined, is some 1,000 feet from the shoreline. The proposed P.E.P. Reef
4 -) Hs(ft)
Jan Feb Mar Apr MayJun Jul Aug Sep Oct Nov Dec
Figure 2 Significant Wave Heights, H,, for Vero Beach, CDN Data 1986-1989.
500 1000 1500 2000
Offshore Distance from Monument R-81 (ft)
Figure 3 Typical Profile at the Center of the Project Site, Monument R-81.
location is between the inner reef and the shoreline. Figure 3 presents a profile from approximately the mid-point
of the proposed P.E.P. Reef installation (Department of Environmental Protection (DEP) Monument R-81).
The long-term shoreline change characteristics in the planned installation area are relevant to the performance of
the submerged breakwater. Fortunately, long-term shoreline position data have been organized for all 24 of
Florida's sandy shoreline Counties as part of the Coastal Construction Control Line (CCCL) program. The
referenced shoreline positions are the Mean High Water shorelines which correspond to the + 1.97 feet NGVD
elevations. The available data span a time period of more than a century (1882 to 1986 for the area of interest here)
and are available at approximately 1,000 feet spacings along 650 miles of Florida's sandy beach shoreline. Figure
4 presents shoreline position changes averaged over DEP Monuments R-77 through R-84, a length of approximately
7,000 ft., which encompasses the proposed location for the P.E.P. Reef. The CCCL data in Figure 4 have been
augmented by survey data taken in July, 1993, by Morgan and Eklund, Inc. The changes span more than a century
and are relative to the 1882 position data. It is seen that from 1882 to 1947 the shoreline advanced, while over the
period from 1947 to 1966, the shoreline receded, possibly in response to the Great Ash Wednesday Storm of March
1962. By 1971, the shoreline had advanced by approximately 60 feet relative to the 1966 position and from 1971
to 1993 (the last date for which data are available in this data set) the shoreline has receded by approximately 6 feet.
Over this period, this would amount to an average annual recession of approximately 0.3 feet per year which would
translate volumetrically to less than 0.5 cubic yard per year per foot of beach frontage. It is recognized that
6 0 --- I -
.. .. ...... .... ... ... ... .. ... .. ... .. ... ... ..... I:... ... .. ... ... .
Figure 4 Long Term Shoreline Changes for Vero Beach, FL. Based on DEP Monuments R-77 Through R-84.
_20I i I------ i I i I i i
Figure 4 Long Term Shoreline Changes for Yero Beach, FL. Based on DEP Monuments R-77 Through R-84.
seawalls and sand added to the system may have reduced shoreline recession over approximately the last ten years.
In summary, the long-term shoreline changes in Vero Beach have been advancement; however, from 1971 to 1993,
the shoreline examined here has receded by approximately 6 feet.
THE PLANNED DESIGN
Although a major purpose of the model study was to develop a basis for improved design of the P.E.P. Reef
installation, prior to a decision to conduct the model studies, a design had been developed for the Vero Beach
The design called for a Reef approximately 4000 feet long constructed in two continuous segments along the Vero
Beach shoreline. The northern segment would be about 1000 feet long and extend from approximately Live Oak
Road to Indian Lilac Road. The southern segment would extend for a distance of 3000 feet from just north of the
Village Spires south to the south end of Humiston Park. The structure alignment ranges from 200 feet to 300 feet
from the shoreline in water depths ranging from 8 to 10 feet NGVD, resulting in a freeboard ratio, f/h (crest
elevation relative to still water level, f, to total depth of water, h) ranging from -0.25 to -0.40 prior to any
settlement of the units. Two of the purposes of the model studies were to evaluate whether or not there was merit
in arranging the units in a planform different than the planned alignment, such as a particular segmentation pattern,
and whether the planned depth was appropriate. It is important to understand the meaning of the freeboard ratio.
Freeboard ratio is defined as a negative quantity for a submerged structure in this report. For example, in a given
depth of water, a barrier with a lower freeboard ratio, such as -0.6, lies further below the water surface than a
barrier with a higher freeboard ratio, such as -0.2. This notation is adopted herein to remain consistent with
previously published works, Ahrens (1987), for example.
OBJECTIVES OF A SUBMERGED BREAKWATER
Breakwaters extending more or less parallel to the shoreline are called "detached" breakwaters. There are two types
of detached breakwaters: submerged and emergent. Emergent breakwaters are those with their crest elevations
generally greater than the highest anticipated astronomical tide elevation. The proposed installation at Vero Beach
is of the submerged type as described earlier.
The three main objectives of any breakwater system are the following: (1) To reduce the wave height, (2) To
increase the retention time of sand behind the breakwater, and (3) To not cause any adverse effects to the beach
behind the installation nor to the adjacent beaches.
EFFECTS OF A SUBMERGED BREAKWATER
There are several possible effects of a submerged breakwater, including the following: (1) Reduction of wave
height, (2) Modification of the nearshore current system, (3) Modification of the cross-shore sediment transport
system, and (4) Modification of the nearshore longshore sediment transport system. Submerged breakwaters may
function differently in sediment "rich" systems as opposed to nearshore systems that generally have a deficit of
sediment. Each of these possible effects are reviewed below.
(1) Reduction of Wave Height
Any object occupying a portion of the water depth will tend to cause a reduction of wave height in its lee. Both
the height and width of the structure are relevant to the magnitude of the reduction. In addition, the wave
characteristics are also a factor, with a greater relative reduction occurring for the higher waves, particularly for
structure-induced wave breaking.
Measurements at the Midtown Palm Beach installation result in a wave height at the inshore wave gage that is
(approximately) from 15% to 35% smaller than the wave height predicted at the inshore gage by the shoaling of
the incident waves. It is believed that a portion of this reduction is due to the fact that the inshore gage is located
in a shallower water depth than the outside gage (approximately 8 feet versus 14 feet) and that some reduction would
therefore occur due to breaking, irrespective of the presence of the breakwater system (Dean, (1994)).
(2) Modification of the Nearshore Current System
The primary effect on the nearshore current system is to allow the onshore transport of water over the breakwater
while restricting the offshore transport, thereby increasing the inshore setup. These effects can result in a net
increase in longshore currents. The magnitude of the increase is dependent upon the proximity of the structure to
the beach, the structure length, and the relative freeboard. If the breakwater is continuous, the longshore currents
will develop and discharge out both ends of the breakwater system. A sufficiently long breakwater could cause a
condition where return flows over the breakwater exist, however, flows out the ends would still occur. If the
breakwater is segmented, the currents may be relieved at the breakwater gaps. The effects of the induced currents
are obviously to transport sand in the direction of the currents and to deposit the sand in those areas where the
current is reduced, namely where the currents exit the breakwater system. There are at least two possible
approaches to reducing the induced longshore currents: 1) Segment and/or offset the breakwater units, thereby
relieving the ponding by seaward flows through the gaps, and 2) Install groins at each end of the breakwater system.
The latter is not considered appropriate for the Vero Beach installation due to its interference with the ambient
longshore transport system. While segmentation of the Reef can reduce the induced longshore currents, this must
be carefully considered due to the increased wave transmission at the gaps. There is the possibility that the
combination of the currents tending to flush the sand from behind the breakwater system and the waves tending to
cause sand deposition behind the system (discussed below) could approximately offset, however, this is considered
(3) Modification of the Cross-Shore Sediment Transport System
On a seasonal basis, sand is transferred from the dry beach and deposited in an offshore bar, to be returned to the
dry beach later. This seaward movement of sand usually occurs in response to winter storms and results in
narrower winter beaches. The milder summer waves cause the sand to be carried shoreward where it is deposited,
restoring the summer beach width. Some studies have shown that the average seasonal shoreline changes in Florida
are on the order of 20 to 30 feet. There is the possibility that the breakwater could cause a net landward transport
of sand. While this is an understandable objective, there is no documentation of this occurring either for continuous
or segmented submerged breakwater systems. For the case of a continuous breakwater, the likelihood of a net
landward cross-shore sand transport is considered unlikely. For the case of segmented breakwaters, the normal
landward transport of sand that occurs during the summer periods could deposit sand landward of the breakwater
and could be retained there through wave sheltering. However, it will be shown later that some designs of a
submerged breakwater can cause a significant net seaward flow of water either through the gaps or at the ends for
segmented breakwaters and at the ends of continuous breakwaters. This seaward flow of water would tend to result
in a net seaward flow of sand, thus tending to counterbalance any net landward sand transport.
(4) Modification of the Longshore Sediment Transport System
The primary interaction of breakwater systems with the longshore sediment transport is through modification of the
waves and nearshore currents, both discussed previously. Modification of the longshore sediment transport by these
two agents will be discussed below.
Consider waves propagating normal to the shoreline as shown in Figure 5. The breakwater causes some reduction
in wave heights and for emergent breakwaters it is known that the breakwater causes a diffraction of the wave crests
as shown. The obliquity of the nearshore wave crests results in sand being transported behind the breakwater where
it is deposited. Many model studies and field installations have demonstrated this characteristic. Next, consider
the case of waves which are more similar to those at Vero Beach, changing directions with the season. During the
times that sand is being transported southward, deposition occurs behind the breakwater on the north end. With
reversals in the longshore sediment transport such that sand is being transported to the north, the waves of reduced
height behind the breakwater are not able to access the sand behind the north end of the breakwater, yet the sand
transporting capacity of the waves immediately north of the breakwater is unaffected, thus sand will be eroded from
the beaches to the north. Although this discussion has been presented from the point of view of the
north end of the breakwater, consideration of the south end will yield the same results: a net amount is deposited
in the lee of the breakwater and the same amount must be eroded adjacent to and south of the south end of the
structure. The degree of erosion via this mechanism is related to the height of the barrier, since the diffraction
patterns are reduced as the barrier becomes more submerged. This overall behavior of causing erosional stress on
the beaches near the two ends of the structure argues strongly for: (1) Pre-construction placement of some amount
of sand in anticipation of such effects (the locations and amounts of sand could be based on numerical modelling
results), (2) A monitoring program to identify such effects, and (3) Plans to respond to erosionally stressed areas.
Oblique Incidence Oblique Incidence
Normally Incident Waves
S Diffracted Wave Crests
----- Adjusted Shoreline
Figure 5 Schematic of Wave Diffraction Around a Breakwater.
ROLE OF SECONDARY FLOWS
The flows induced in the model were fairly weak especially for freeboard ratios similar to those at the Midtown
Palm Beach installation (<-0.4) and planned at Vero Beach (probably less than -0.4, considering settlement). In
some cases in which flows are significantly smaller than other flows in the system, they are called "secondary"
flows, and it might seem that their effect on the sediment transport system would also be "secondary"; however,
this may not be the case. An example is the secondary helical flows that are found in meandering rivers: these
helical velocities are much smaller than the primary velocities along the channel axis. Yet these flows are
responsible for the development of the meander system and in some cases the short circuiting of the meander loops
and the resultant formation of "ox bow" lakes. In the case of interest here, the beach profile can be considered to
be in near equilibrium with the extant forces prior to the installation of the submerged breakwater. The submerged
breakwater induces weak currents that place the beach profile out of equilibrium and thus cause sediment transport
until a new equilibrium is reached. With the divergence of longshore currents associated with net onshore flows
over the breakwater, the effect would tend to be a scouring of the area landward of the breakwater and deposition
near the two ends of the breakwater where the currents weaken.
LABORATORY TESTING AND RESULTS
The P.E.P. Reef was evaluated for its hydrodynamic performance in the three-dimensional wave basin at the Coastal
& Oceanographic Engineering Laboratory in Gainesville, FL. Forty-eight individual 1:16 scale concrete units,
depicted in the definition sketch Figure 6, were fabricated from design drawings provided by American Coastal
Engineering, Inc. Original testing was conducted on a movable sand bed, and subsequent testing was performed
on a fixed bed. The fixed bed tests consisted of a 47 foot longshoree direction) beach with a 1:8 gravel slope
fronted by a 16 foot (cross-shore width) horizontal fixed bed, followed by a downward sloping section to a paddle
type wavemaker. A schematic of the laboratory setup is shown in Figure 7. Normally incident waves averaging
2.1 feet in height and eight seconds in period (prototype values) were used in all tests. These values were chosen
to be reasonably representative of the conditions at Vero Beach. The tests conducted and reported on herein are
not intended to encompass the full range of conditions that will occur at the Vero Beach site, rather they illustrate
mechanisms and trends allowing for qualitative extrapolation to conditions not included in the test program.
Evaluation of the Reef consisted of wave attenuation measurements as well as current measurements via dye and
drogues. Capacitance type wave gages were used to measure wave heights in both the cross-shore and longshore
Figure 65 Definition Sketch.
7/////////////I//11//11/11//11//////// I/1/1/ 1/1/1////1///////11///
P.E.P. Reef Model
1:8 Gravel Gridded
Beach Test Area
Figure 7 Wave Basin Schematic.
directions. These data were used to determine transmission coefficients, Kt, for each Reef arrangement tested.
Current measurements were taken from video-tape of drogues moving on the floor of the fixed concrete bed. A
one foot square grid painted on the bed provided distance measurements from which velocities were calculated.
A submerged breakwater, as discussed previously, acts in several ways in the surfzone. First and foremost, it
causes wave attenuation through energy reflection and dissipation of the incident waves. Secondly, it creates an
elevation in the water level in the lee of the structure. This elevation, called "ponding", is due to the interruption
of the natural return flow which occurs in the absence of the breakwater and the transfer of momentum from the
waves striking the reef. This elevation of water surface creates an elevation gradient which will drive currents to
relieve the water volume buildup in some fashion in the area of the reef. The Midtown Palm Beach Installation
has experienced a net volumetric erosion landward of the breakwater which could be due to augmentation of
longshore currents via this mechanism. Seelig and Walton (1980) discuss the desire to limit the induced average
velocity around breakwater gaps to 0.5 ft/s to prevent significant sediment transport out of the breakwater system.
Consideration of both wave attenuation and current patterns provides the focus of the analysis presented here.
MOVABLE BED STUDIES
The tests commenced with a movable sand bed installed in the basin. A 25 foot (cross-shore) beach consisting of
a 1:8 slope beachface and a 1:30 slope offshore area was used. The Reef was placed 15.6 feet offshore (250 feet
prototype) based on the design plans for the Vero Beach installation in 9 feet of water (NGVD). Eight hour tests
were conducted on several different arrangements of the Reef, with profile surveys taken every four hours. In
addition, wave height profiles were taken along with video-tape of dye and drogue motions.
Survey data of the tops of the units indicated the settling phenomenon also witnessed in the Midtown Palm Beach
Installation. Settlement of 1.3 feet (prototype) on average was observed over a 32 hour prototype timeframe. While
this figure cannot be directly compared to Midtown Palm Beach field data since the sediment used in the laboratory
was not scaled down according to the 1:16 ratio, the amount of settlement is of the same order as the field data from
Midtown Palm Beach and the calculations done in preparation for the Vero Beach Installation.
Volume change and wave height analysis, along with video tape analysis, provided little conclusive information.
Velocity measurements of the system were difficult to obtain due to the unsteady flows over the changing bar
formations in the basin. Based on these observations it was decided to concentrate on a fixed bed model and focus
on currents and wave height measurements.
FIXED BED STUDIES
Two sets of fixed bed studies were carried out: A) An initial set of four arrangements and B) an additional set of
three arrangements based on analysis of the initial tests. These two sets are presented separately to preserve the
sequence of testing, analysis, and consideration leading to the final arrangements.
A) Initial Fixed Bed Tests
Using the gravel and concrete fixed bed arrangement shown in Figure 7, four initial arrangements of the P.E.P.
Reef were evaluated at four depths, along with a no-reef control test at each depth. These were chosen to give
representative information from the myriad of possibilities of unit placement. The depths were chosen to provide
freeboard ratios, f/h (crest elevation relative to still water level, f, to total depth of water, h), of 0.0, -0.2, -0.4,
and -0.6. Again, these freeboard ratios are expressed as negative quantities to distinguish submerged barriers
(negative freeboard ratios) from emergent barriers. To provide reference, the Midtown Palm Beach P.E.P. Reef
installation has an average freeboard of -0.58, including settlement effects to date (Dean et. al, 1994). The four
plan arrangements chosen are depicted in Figure 8, and subsequent planform references are based on these four
A) 45 Continuous Units
B) Three 11 Unit Segments
C) 45 Staggered Units
J -1 1
!i I 77ii i i 1177 7I i i I
D) Five 5 Unit Segments
i11 i i iH'H m 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 L
Figure 8 Reef Unit Arrangements Tested in Model.
i i i
- I L
(A, B, C, and D) types. Twenty tests were conducted. The type A case includes 45 model units arranged in a line
parallel to the beach, equivalent to 540 feet of structure (prototype). This simulates the proposed arrangement and
the Midtown Palm Beach configuration. The type B case uses 33 units in 11 unit segments separated by gaps equal
in length to 10 units. This is equivalent to 132 foot segments separated by gaps of 120 feet in prototype. Case C
also contained 45 units, with every other five units offset seaward by two unit cross-shore widths, w. The fourth
case, case D, is similar to case C except with the absence of the four offshore segments. This case represents five
60 foot long segments separated by 60 foot gaps.
(1) Wave Height Measurements
A capacitance type wave gage was used to measure wave heights for each of the four P.E.P. Reef arrangements
shown in Figure 8 as well as a control case. The measurements were taken along cross-shore gridlines (0 ft.,
6 ft., and 12 ft. relative to the basin centerline) as displayed in Figure 9. The estimated error in the recorded
model wave heights is 0.01 ft. (about 10% of the recorded wave heights).
I I !
-12 ft. -6 ft. O ft. 6 ft. 12 ft.
Figure 9 Cross-Shore Wave Height Profile Lines.
An average of the five cross-shore wave height profiles was taken for each Reef arrangement and transmission
coefficients calculated from each. The f/h = 0.0 case is depicted in Figure 10. The graph shows the reduction
in wave height over the Reef in the initial four cases compared to the control profile. It is noted that the type B
and type D cases contain profiles in their averages that fall in gaps between the breakwater, as seen from examining
Figures 8 and 9. These gap profiles were included in the averages to represent the effect that areas in front of large
breakwater gaps will not experience as much attenuation of wave height and will result in a higher overall
transmission coefficient for that particular arrangement. In comparing these different Reef arrangements the wave
heights are examined close to the beach toe and compared to the control wave height at the same location. In the
0.0 freeboard ratio tests the average wave height profiles show wave attenuation to be the greatest in the type A and
O .0 8 ; .. .. .. ... .... ..
....... 45 Continuous (A)
0.04 3 Sement (B) ...
0.04 ------ 45 Stagered (C) ireef
---- 5 Unit Gaps (D)
0 1 2 3 4 5 6 7 8 9 10 11 12 13
Offshore Distance (ft./grid)
Figure 10 Cross-Shore Wave Height Profiles for Freeboard Ratiof/h = 0.0.
type C cases, those that have 45 units across the basin. The type B case and type C case showed less reduction in
wave height as a result of the gaps in their respective planforms. The peaks at the 9 and 12 foot gridlines indicate
the reflection of energy by the Reef. As the freeboard ratio decreases (from 0.0 to -0.6), the wave attenuation
decreases. At a freeboard ratio of -0.6, none of the four arrangements indicate significant wave height reduction.
The cross-shore wave data for each test were then longshore averaged over the 0 to 3 foot cross-shore gridlines and
divided by the average longshore wave height from the 10 to 13 foot cross-shore gridlines to calculate the
transmission coefficient, Kt. The ranges over which the wave heights are averaged and the distance from the Reef
provide some compensation for reflection effects and allow for wave reformation after passing over the Reef. The
transmission coefficients are plotted versus freeboard ratio in Figure 11. The plot clearly demonstrates wave
attenuation decreasing as freeboard ratio decreases. While a freeboard ratio of 0.0 results in a significant wave
height reduction, the -0.4 and -0.6 freeboard ratios produce little to no wave height attenuation. The type A
planform generates the greatest wave attenuation, followed by the type C case. The type B and type D cases do
not reduce the wave heights as much, since they contain gaps through which waves pass nearly unaffected.
Considering that the wave transmission coefficients at the Midtown Palm Beach installation are 0.65 to 0.85 and
that the average freeboard ratio, f/h, is approximately -0.58, the transmission coefficients obtained here are
considerably greater (less wave height reduction).
For the highest freeboard ratio, 0.0, the type A planform has a transmission coefficient of 0.75. The type B, C,
and D case values are 12.7%, 5.3%, and 20.2% higher than the continuous 45 unit case, respectively. At lower
freeboard ratios these cases vary less than 10 percent from the type A planform. Ahrens (1987) performed
experiments on submerged stone breakwaters. Figure 11 shows the experimental curve developed from that work
for the corresponding freeboard ratios of this study and demonstrates that it compares well to the type A continuous
length Reef case. At a freeboard ratio of 0.0, Ahrens (1987) predicts a Kt value of 0.68. The percent difference
between Ahrens' work and the present study in this case is less than 10 percent at a freeboard ratio of 0.0. and
decreases with decreasing freeboard ratio to less than 1.0% at the -0.6 freeboard ratio (for the continuous 45 unit
Figure 12 presents transmission coefficient data for the type A case plotted in terms of freeboard non-
dimensionalized by incident wave height. These are the same data shown in Figure 11 presented here to show the
1.1 II I I II I 1
Figure 11 Wave Height Transmission Coefficients.
- -- **. ^ .^
----------- "^ O
45 Continuous (A)
-- O 3 Segments(B ... .......
S 45 Staggered C)
5 Unit Gaps (D)
1I I I l*
0 .9 -- -- -- -- -- -- ---- --- --- --- -- --------- ---- -- -- --- .. .. ... .. ... .. .. ...
S0 .8 ........................................... ............. ...................
0 7 ---- -
A 45 Continuous (A)
0 .6 I i-I I
-5 -4 -3 -2 -1 0 1
Figure 12 Wave Height Transmission Coefficients versus Freeboard, f, divided by Incident Wave Height, Hi.
relationship of wave height and barrier height on wave attenuation. The figure indicates that as the magnitude of
the freeboard increases relative to the incident wave height, the effectiveness of the barrier is greatly diminished.
At a value of f/Hi of -4.5 the transmission coefficient reported by Ahrens (1987) for the P.E.P. model conditions
is 0.99. Below a value of -1.3 forf/Hi, 90% or more of the incident wave height is transmitted over the structure.
Using video-tape and the gridded bed, velocity measurements were taken in the vicinity of the Reef. One and one
half inch diameter balls made slightly negatively buoyant were used as bottom drogues to determine the current
patterns on the bed, presumably where most sediment transport would occur. Approximately 30 drogues were
tracked per test, and where possible their speeds measured. Raw data from the drogue tracking are presented in
APPENDIX I for the highest three freeboard ratios; the lowest freeboard ratio yielded no useful measurements as
the bottom currents were too small. Velocities were compared to the controls at each freeboard ratio to attempt
to determine the effects of the Reef on the current patterns.
Dean et. al (1994) presented a hypothesis for the flow over a submerged longshore barrier. Wave motion transports
water over the structure, and the presence of the structure impedes the return flow normally seen in waves
propagating onto a beach. This interruption forces at least some of the return flow to be directed alongshore until
it finds a relief area to flow offshore. Dean et. al (1994) proposed that the erosion of sediment observed landward
of the Midtown Palm Beach installation and the accumulation near both ends after the first four months survey with
the full Reef installed could be due to the currents generated in the longshore direction by this mechanism. Figure
13 shows a representation of the current patterns produced by a Reef arrangement of 45 continuous units at a
freeboard ratio of 0.0 (Still Water Level, SWL, at the crest of the structure). This represents a prototype Reef
length of 540 feet. The figure indicates a strong current pattern generated by flow over the structure directed to
either end of the Reef. The flow is then relieved around each end. As compared to the control velocities at that
depth, for this case the presence of the Reef increased the current magnitude on the bottom at the ends by five
times, and reversed its direction (from onshore to offshore). Similar effects are seen at greater depths and for other
arrangements, with the current magnitudes decreasing for greater depths. The magnitudes of the outgoing flow on
the bottom were measured up to 0.6 ft/s prototype on average for the above case, which is greater than the design
recommendation of Seelig and Walton (1980). While this water elevation is an extreme case, it demonstrates the
presence of the 'pumping mechanism' and validates its importance as a secondary flow, which, as previously
discussed, can be very effective in sediment transport.
For the 0.0 freeboard ratio cases, all four of the initial arrangements demonstrated a strong offshore bottom flow
around the ends of the Reef. Increases of two to five times over the control case currents were documented. In
only one case was an offshore directed flow seen anywhere other than the ends, that of the type B case. There the
flow was located only in one small area in one gap and represented an increase in velocity of two times that of the
Figure 13 Bottom Drogue Trajectories, f/h = 0.0.
The -0.2 freeboard ratio cases exhibited much the same behavior as the 0.0 freeboard case, with no offshore bottom
flow noted except around the ends of the entire installation. Increases of up to 4 times the control velocities were
seen at the ends of the structures, and again the direction of current was reversed in the presence of the Reef.
Where ends of the Reef were encountered, either at gaps or at the terminus of the Reef, drogues exhibited eddy
patterns characteristic of the pumping mechanism described previously.
At a freeboard ratio of -0.4, the current patterns become much less discernable. Some offshore flows are seen, but
the magnitudes of these flows are quite small, equal to the control magnitudes in most cases. Motions in the -0.6
freeboard ratio cases were too small to be distinguished above the controls. It is noted that the longest segment
length simulated in the model was 540 feet while the Vero Beach planned installation has one 1,000 foot segment
and one 3,000 foot long segment. The increased length of the field installations would magnify the pumping effect
over that observed in the model.
The drogue tracking represents bottom currents for all cases. In only a few cases was an offshore directed bottom
current detected in the lee of the Reef, which may seem counter intuitive based on wave theory. This would
indicate that the offshore relief is manifested in the upper parts of the water column. Dye studies performed in
conjunction with the drogue tests support this contention.
In cases where the pumping mechanism is strong and offshore flow is seen around the ends, dye behaves similarly
to the drogues throughout the water column. In cases where little offshore drogue motion was seen, the dye
indicated locations where the return flow occurred. These instances correspond primarily to thef/h = -0.4 and -0.6
cases where the Reef lies in deeper water. Under these conditions, dye was observed moving offshore near the
centerline of the Reef, similar to the control case at each depth. This behavior attests to the three-dimensionality
of the problem and the difficulty in controlling the location of the offshore flows. It is desirable to restrict or
prevent the pumping effect around the ends; however, simply putting in gaps along the length of the Reef may not
relieve this effect completely, as indicated by the dye/drogue tests.
While difficult to quantify, the pumping mechanism relates to wave transmission. In Figure 11 as the wave
transmission coefficient increases with decreasing freeboard ratio, the pumping effect decreases as well. At f/h
ratios of -0.4 and -0.6, the pumping effect is nearly indistinguishable, as is any wave attenuation. This would
indicate that, in reference to the Midtown Palm Beach installation, the times when the effects of the Reef are most
noticeable occur during situations of higher waves at lower water elevations, such as low spring tides. It is at these
times when the waves are the most attenuated (smaller K, values) and the Reef blocks a higher percentage of the
B) Additional Fixed Bed Tests
Based on analysis of the initial tests, a compromise arrangement was sought between the need for wave attenuation
and the reduction of the pumping mechanism generated by wave attenuation, as discussed above. The staggered
Reef arrangement appeared to hold the most promise, therefore, additional tests concentrated on variations on the
Type C case presented previously. Seven additional tests were conducted, and each test was performed at the 0.0
freeboard ratio case. This depth of water was chosen to clearly demonstrate the effects each arrangement causes,
since, as shown in the previous twenty tests, deeper water depths merely diminish the effects of the Reef.
Figure 14 shows the additional arrangements tested. The Type E case used 45 units in nine unit segments, offset
at three different distances, x = 2w, 4w, and 6w. These distances were based upon the cross-shore width of the
units, w, (fifteen feet prototype width). Each segment equals 108 feet in prototype, totalling 540 feet of tested
length. The Type F case was similar to the Type E case, only the offshore segments were reduced in length by
two units each, representing 84 feet in prototype length. The two units on each end of the offshore segments were
removed in an attempt to promote offshore relief flow between segments. The Type F case was also run at the
same three offset distances. The Type G case was conducted in response to the results of the Type E and F cases.
The Type G case consists of an 18 unit center section flanked by two eleven unit offshore segments. The segments
are separated by two units in the longshore direction and four units in the cross-shore direction. This is twice the
scale of the Type F case in the longshore direction with the offshore segments as long as the basin would permit.
Wave height and current measurements were performed on the final seven tests. In the cases where the offshore
segments prevented wave height measurements at the 10, 11, and 12 foot gridlines, an extra measurement was taken
at the 14 foot mark to quantify the offshore wave height at that point and was included in the average to compute
transmission coefficient. Transmission coefficients for each of the seven additional tests are presented in Table 1.
Table 1 Wave Transmission Coefficients for Additional Tests.
Arrangement f/h Offset Distance Kt
E 0.0 2w 0.79
E 0.0 4w 0.90
E 0.0 6w 0.79
F 0.0 2w 0.75
F 0.0 4w 0.96
F 0.0 6w 0.83
G 0.0 4w 0.78
E) 9 Unit Segments
F) 9 & 7 Unit Segments
) 18 l n -l U Ssllll i II IIIII
G) 18 & 11 Unit Segments
vmT I IF
Figure 14 Reef Arrangements for Additional Model Testing.
The results of the transmission coefficient analysis do not present a clear picture of the performance of the Reef.
Some variation in Kt results from reflection effects caused by both the Reef and the wavemaker. Measurements
taken offshore of the Reef may be affected by the reflection envelope created by the structure and the wavemaker.
If the offshore gridlines fall on an amplified portion of the envelope, the K, reported will be lower than it actually
is, and vice versa if the measurements fall on a reduced portion of the reflection envelope. The results are shown
here in the same fashion as the initial tests for completeness. A sufficient number of tests was conducted in the
initial phase to minimize the impacts of the reflection effects and assure that the reported Kt values were reliable.
I I I I 1 I I l l I l
To resolve the question of wave attenuation in the additional tests, the actual values of wave height in the lee of the
structures were examined. The average offshore wave height was measured to be 4.0 centimeters (2.1 feet
prototype). The average onshore wave height for the seven tests was measured to be 3.3 centimeters, resulting in
an average transmission coefficient of 0.81, which compares well to the results shown in Figure 11. In the cases
of the 4w offset distances where high Kt values were observed, the onshore wave heights measured were actually
lower than the average for all seven tests, measuring from 2.9 to 3.0 centimeters. This indicates that the offshore
wave heights measured were also lower, falling on reduced portions of the reflection envelope. The conclusion here
is that regardless of the offshore reflection patterns generated by the Reef and the wavemaker, the onshore heights
(those that would impact the beach) are being reduced by the Reef, and in all seven tests the amount of reduction
is approximately equal. This is expected in a horizontal bed test where each wave height profile runs over a Reef
segment. The gaps are sufficiently small such that the entire model length receives the same wave attenuation.
The other portion of a compromise solution involves the currents generated by the pumping mechanism. The
rationale for the arrangements chosen in the additional tests was to provide gaps for offshore relief of water along
the project length while maintaining an acceptable level of wave attenuation. Obviously an arrangement with any
sizable gaps would leave a portion of the shoreline exposed to unattenuated waves. By staggering the Reef
segments, a gap can still be included in the planform.
The staggering of the Reef system did show signs of reducing the pumping effect around the ends in some cases.
In the Type E case, offshore relief through these gaps was observed for the 4w and 6w offset distance tests. These
tests provided a large enough area to overcome the "bridging" effect between segments and allow fluid to flow
offshore. The relief flow, however, was not steady and not located at one particular gap. The relief, like the
nearfield flow, appeared unsteady and moved from gap to gap during the test. This was observed via dye releases
in the gaps during the testing. The magnitudes of the flows observed around the ends were not significantly
reduced, however, showing increases in velocity over the control case of up to five times. The 2w offset distance
test provided no benefit over the continuous Reef case, causing high velocity flows around the ends of the entire
structure. Again, the control case indicates onshore flows in the areas near the ends, while the Reef causes offshore
flows in these areas.
The Type F case tests showed some improvements in reducing the pumping effects. By shortening the length of
the offshore segments, small longshore gaps were created and offshore relief was promoted. Offshore flow around
the ends was still present in each of these tests, but the magnitude of the flow was noticeably reduced in some tests.
In the 4w offset distance case, the offshore relief was quite noticeable, and the velocity of the flow around the ends
was reduced to 3 to 4 times that of the control case. The 6w test showed no improvement over the 4w test.
Pursuant to the Type F tests, which showed the most promise of all the arrangements studied, an additional test was
conducted to determine if the segment lengths could be extended without creating a strong longshore flow. The
Type G test extended the lengths of the segments used in the Type F case to twice their length (208 feet in
prototype). This arrangement is of interest for the sake of construction cost, both in time and material. Longer
segments require fewer stabilizing tie downs at segment ends and can be installed more rapidly. The results of the
test were encouraging, as flow velocities measured around the ends of the entire structure were 2 to 3 times that
of the control case (and in the opposite direction). This velocity represents a prototype velocity of up to 0.4 ft/s,
below the critical velocity suggested by Seelig and Walton, 1980.
DISCUSSION OF MODEL RESULTS
The goal of any laboratory study of submerged breakwaters is to understand how the structure interacts with the
surrounding fluid and sediment. Currently this understanding is limited primarily to the behavior of wave height
attenuation, and the effects of such a structure in the surf zone in fact present a complex three-dimensional problem.
This work attempts to address this problem, both in the hydrodynamic and the sedimentary regimes.
It is apparent from the tests conducted that secondary flows (those of lesser magnitude than the ambient longshore
current) play a key role in the effects of a submerged breakwater. Ponding of water in the lee of the Reef is one
effect that would directly impact the sediment transport via the longshore currents generated. This ponding effect
is related to the relative height of the structure. As the relative height increases, more water is retained in its lee,
contributing to the elevation gradient in the longshore direction. At the same time, the amount of mass transported
over the barrier remains relatively fixed, as it is concentrated in the trough to crest level of the water column,
therefore the pumping effect increases. However, to create sufficient wave height attenuation, the structure must
block a significant portion of the water column. Thus a balance must be reached between the degree of wave
attenuation sought and the amount of pumping generated. From the model studies, a structure whose crest is located
at the SWL generated the greatest amount of wave attenuation but also generated the strongest pumping currents
around its ends. This relationship does not hold for emergent structures, obviously, where overtopping can be
limited and/or completely prevented while achieving significant wave attenuation.
The current magnitudes generated by a submerged obstacle are dictated ultimately by the length and relative height
of the structure (which controls the transmission of wave energy). When the pumping mechanism is strong, currents
around the Reef are dominated by this longshore flow. Referring again to Figure 13, any drogue or dye released
in the test area was ultimately drawn into the circulation pattern seen in the figure.
At lower freeboard ratios (-0.4 to -0.6), the pumping mechanism is reduced as the Reef retains less fluid behind
it. In these cases the flow field becomes more complex. The system generates both a nearfield and a farfield flow
pattern. The nearfield flow, as indicated by bottom drogues, is extremely complex and unpredictable. Reef
segments appear to be connected by a bridging mechanism that acts to both decrease wave energy in reefline gaps
and set up flow channels within segmented Reef systems. This bridging effect is not well understood but may prove
beneficial in providing some degree of wave energy reduction in a gap between two breakwater segments. The
reader is referred to APPENDIX I for examples of this complex flow. Drogues were observed in several tests to
move from one segment to the next and back again, never leaving the nearfield and never remaining in one location.
In some instances the drogues were observed to 'bounce' away from the nearfield and enter the farfield flow. This
farfield flow is much better defined and predictable, again driven by the longshore gradient in surface elevation.
In the two higher freeboard ratio cases, the farfield flow demonstrates this pattern. At lower freeboard ratios the
presence of the barrier becomes less and less significant, and the flow patterns closely resemble the control case.
One of the more interesting findings of the current study was the lack of a near-bottom based return flow generated
by wave action. This phenomenon is well documented in two-dimensional wave tank tests. In the basin study
conducted here, however, this is not the case. Studying the control tests at each depth, the return flow is highly
three dimensional, appearing to be located in the upper portions of the water column or on the surface, and
occurring at specific locations along the beach, almost like a rip current. Videotape of the control cases indicated
return flows located along or near the centerline of the basin, describing an overall circulation pattern for the basin.
In the instances of low freeboard ratio, the return flows were also located along the centerline region of the basin.
At higher freeboard ratios, the Reef dominated the flow patterns, reversing the control patterns in all cases.
The relative absence of bottom return flows raises the question of upwelling of water on the leeward side of the
Reef. This phenomenon, seen in the Midtown Palm Beach Installation, is visible as the trough of a wave passes
over the barrier. As waves pass over the Reef some of their momentum is transferred down into the water column.
This creates an eddy in the lee of the Reef, much like the eddy formed in the simple flow of water over a weir.
This mechanism draws water toward the Reef and directs it upward, creating upwelling patterns visible on the
surface. This behavior has been considered beneficial for preventing offshore loss of sediment over the Reef. The
results of this study do not necessarily support or negate this mechanism, however, the flow of water in this manner
is seen to be limited to the very nearfield of the Reef itself. Away from the nearfield in the lee of the barrier, most
bottom flows appear to be alongshore or onshore, which would not carry any sediment into the upwelling region.
The sediment transport in the area of a breakwater is directly linked to the hydrodynamic activity created by the
barrier. Any area where wave energy is decreased will experience a decrease in the usual wave-induced longshore
current. This will increase the residence time of sand passing through the area, causing some sediment to be
deposited in this region. This is the mechanism by which downdrift areas are adversely impacted by shore parallel
structures. This effect is in part mitigated by the pumping mechanism, although sediment must be transported well
out of the shadow zone of the structure to reenter the longshore flow. To summarize, the overall farfield transport
of sediment is dictated by the natural longshore flow and the pumping mechanism described previously. At lower
water levels, the pumping mechanism plays a strong role in sediment transport. At higher water levels (i.e. lower
freeboard ratios), wave attenuation is slight and the pumping effect is diminished, allowing the natural longshore
current to dominate sediment transport.
This concept also describes what would happen in the case of oblique wave incidence. The ponding phenomenon
would still occur; however, the divergence of the longshore flow resulting from it would no longer be symmetric.
Less flow would be seen opposing the ambient longshore current, transporting more sand in the lee of the Reef in
that direction. When the longshore transport reverses its direction the same mechanisms apply only in the opposite
direction. In each case, sediment close to the ends of the structure cannot be transported by the waves passing over
the Reef, so beaches to either side of the installation would experience erosion. In the case of continuous normally
approaching waves (a situation that does not often occur in nature), the only area that would erode due to a shore
parallel submerged breakwater would be in the lee of the structure.
The pumping mechanism can be partly mitigated by the use of offset segments in the planform and small gaps
between the offset segments. These gaps and offsets (tested at approximately 24 and 60 feet, respectively, in the
most successful tests) serve to provide offshore relief of the pumping current along the length of the entire project
rather than entirely at the ends. This reduces the magnitude of the induced longshore current, thus reducing the
sediment transported from the lee of the structure. At the same time, the small gaps do not leave lengths of
shoreline exposed to unattenuated waves. The pumping mechanism cannot be completely eliminated, as it is a result
of the presence of the structure itself. Thus the use of offsets and small gaps appear to provide a reasonable
compromise. In addition, placing the end segments of the installation in the offshore position may provide a more
gradual transition for the longshore current moving out of the project area. This may aid in retaining sand in the
lee of the structure, albeit at the ends. This sand might then be pushed onto the beach during mild wave conditions.
It is noted that this work does not present an 'optimal solution' in the ideal sense. The nature of this research is
extremely opportunistic; there are many ways to arrange units along a large stretch of coastline. An optimal
solution would require field experimentation at the specific site, something that is obviously not feasible. The
information presented herein provides a basis of performance of such structures and should be treated as such.
SUMMARY AND RECOMMENDATIONS
The following summary statements follow based on the model tests.
1. For the limited movable bed tests, the settlement of the units was surprisingly similar to that found at the
Midtown Palm Beach Installation.
2. The "basin effects" are small in the fixed bed basin in which most of the studies reported herein were
3. Based on wave height measurements, the Reef effects decrease with decreasing freeboard ratio and are
almost nil nearf/h = -0.4 to -0.6.
4. For all the tests in which the Reef caused a significant wave height reduction, the effects of the Reef were
to cause net onshore flows over the Reef. This flow must be compensated by offshore flows elsewhere.
5. The effect of the Reef on the current system was especially noticeable for the higher freeboard ratios and
consisted of onshore flows over the Reef and offshore flows near the ends of the Reef.
6. For the staggered arrangement of units, there appears to be an effect termed herein as "bridging" which
reduces the seaward flow through the gaps associated with the staggering of the units to a greater degree
than anticipated. This phenomenon also appears to occur for the case of segmented breakwaters if the gaps
are relatively small.
7. Staggering of the Reef segments appears to provide some relief of the longshore currents created by the
pumping mechanism described herein.
8. The presence of gaps along the line of the Reef also appears to provide some offshore relief to the
9. Although only one wave height was tested in this program, the qualitative effects of different
heights can be determined from the results presented herein.
The following are recommended.
1. Consideration be given to segmenting and offsetting the breakwater in order to reduce the induced
longshore currents and associated loss of sand landward of the breakwater. This will entail placing units
in two different depths of water, affecting their wave attenuation characteristics. Subsequently, the amount
of settlement of the units will be important (see below).
2. Two related non-dimensional freeboard variables, f/h and f/Hi, have been discussed in this report. For
design, it is recommended that consideration be given to locating the units in a water depth that will, after
unit settlement, provide a relative freeboard value, f/Hi of greater than or equal to -1.3 based on the NGVD
placement and the design wave height chosen (if the relative freeboard is less than -1.3, the effectiveness
of the Reef will be reduced significantly). It follows that consideration be given to allowing for up to the
amount of settlement that the first 57 units have experienced at the Midtown Palm Beach installation
(average of 2.7 feet). Alternatively, consideration be given to finding a solution to the settlement problem,
such as adding a foundation under the units to prevent settlement.
3. Consideration be given to making the segments of the Reef as short as economically feasible, as the length
of the continuous structure affects the amount of pumping in its lee.
4. If the breakwater is segmented, recognition be given to the effects of the variation of wave heights
and directions along the shoreline and the resultant irregular planform of the shoreline. The use
of relatively small gaps in the planform will minimize this effect.
5. Consideration be given to the safety aspects of the system, especially the induced currents. With any
artificial structure addition to the surf zone, creation of actual or perceived adverse currents, i.e. rip
currents, etc., must be anticipated. This may be particularly relevant in view of the observed seaward
flowing surface currents in the model.
6. Following the completion of the numerical modelling, consideration be given to appropriate placement of
sand in anticipation of any potential adverse effects of the installation as identified in the numerical
Vernon Sparkman is acknowledged for his superb craftsmanship and assistance in the completion of this project.
Jim Joiner provided general assistance in the laboratory studies, and Mike Dombrowski contributed to the data
collection. In addition to contributing the foreword to this report, Don Donaldson also provided valuable assistance
in all aspects of the model study.
Ahrens, John P., "Characteristics of Reef Breakwaters," Coastal Engineering Research Center, Technical Report
CERC-87-17, 1987, 45 pp. plus 3 Appendices.
Coastal Tech, Inc., "Sebastian Inlet District Comprehensive Management Plan," 1988, 53 pp. plus 5 Appendices.
Dean, R.G., Dombrowski, M.R., and Browder, A.E., "Performance of the P.E.P. Reef Installation, Town of Palm
Beach, Florida, First Six Months Results," Coastal & Oceanographic Engineering Department, University of
Florida, UFL/COEL-94/002, 1994, 34 pp. plus 5 Appendices.
Seelig, William N., and Walton, Todd L., Jr., "Estimation of Flow Through Offshore Breakwater Gaps Generated
by Wave Overtopping," Coastal Engineering Research Center, CETA 80-8, 1980, 21 pp.
U.S. Army Corps of Engineers, "National Shoreline Study, Regional Inventory Report: South Atlantic-Gulf
Walton, T. L., "Littoral Drift Estimates Along the Coastline of Florida", Florida Sea Grant Program, Report No.
3, 1976, 39 pp. plus 3 Appendices.
TRAJECTORY / VELOCITY
___i. Ql3 ^I
I I______ S^ i i__
; 62 i
I ____ ~ *
'fl-H-- H -f-I-
-- a. i i -
1 II !
~ -- ""-n-i" ^
_J __ : __ i ^0
Figure I-1 Control and Type A Case, f/h = 0.0. Wavemaker is to the right. Units in ft/s.
ZI I ______ f HI
1. I __A
Figure 1-2 B and C Cases, f/h = 0.0. Wavemaker is to the right. Units in ft/s.
Figure 1-3 Type D Case, f/h = 0.0. Wavemaker is to the right. Units in ft/s.
Figure I-4 Control and Type A Case, f/h = -0.2. Wavemaker is to the right. Units in ft/s.
^^ 1 !
_^kJ_____ ; l I a
Figure I-5 B and C Cases, f/h = -0.2. Wavemaker is to the right. Units in ft/s.
-! j ^ s
Figure I-6 Type D Case, f/h = -0.2. Wavemaker is to the right. Units in ft/s.
Figure I-7 Control and Type A Case, f/h = -0.4. Wavemaker is to the right. Units in ft/s.
Figure I-8 B and C Cases, f/h = -0.4. Wavemaker is to the right. Units in ft/s.
Figure I-9 Type D Case, f/h = -0.4. Wavemaker is to the right. Units in ft/s.
Figure 1-10 Type E Case, f/h = 0.0. Offset Distances = 2w and 4w. Units in ft/s.
Figure I-11 Type E Case, f/h = 0.0. Offset Distance = 6w. Units in ft/s.
Figure 1-12 Type F Case, f/h = 0.0. Offset Distances = 2w and 4w. Units in ft/s.
Figure 1-13 Type F Case, f/h = 0.0. Offset Distance = 6w. Units in ft/s.
Figure 1-14 Type G Case, f/h = 0.0. Offset Distance = 4w. Units in ft/s.
FOR P.E.P. REEF MODEL
WAVE TRANSMISSION CHARACTERISTICS
P.E.P. Reef Wave Transmission Characteristics Based on Ahrens' (1987) Analysis
Calculations performed on laboratory scale for Vero Beach P.E.P. Reef Studies, UF COE 1994, AEB
i:= 0...200 j:= 1..7
T := 2 Wave period for all tests [sec]
g := 9.81 Acceleration due to gravity [m/secA2]
dsj := Water depth for freeboard ratios 0.0, -0.1, -0.2, -0.3, -0.4, -0.5, -0.6 (cm)
15 Use dispersion relationship to solve for
17.143 wavelength at each freeboard ratio
g-tanh k, 10
he := 12
At := 133.62
Hmo := 4
Wave numbers for each freeboard ratio [1/m]
Wavelengths for each freeboard ratio [cm]
Height of unit [cm] (note unit is actually 11 cm, but
floor is uneven, so 12 cm creates 0.0 freeboard)
Cross sectional area of unit [cm^2]
Zero moment incident wave height [cm]
Freeboard of structure crest, and corresponding
In order to fully use Ahrens equation, the typical dimension of the median stone must be found although
the P.E.P. Reef is not a rubble mound structure. Exclusion of this term in the equation results only
in less than one percent change in Kt for the P.E.P. Reef conditions.
wr := 2.4
W50 : At-22.86-wr
d50 := W 3
C1 := 1.188
C2 := 0.261
C3 := 0.529
C4 := 0.00551
d50 = 14.5094
Density of stone [g/cm^3]
Median stone weight [g]
Typical dimension of the median stone [cm]
Constants determined by Ahrens (1987)
Ahrens (1987) equation (13)
At \C2 Fj \ At2
I.LP- .exp C3 ) + C4-
sj -Lp Hmo d502LpJ
Transmission coefficients at each freeboard ratio