THE EFFECTS OF FORESHORE SLOPE ON

BREAKWATER ARMOUR UNIT STABILITY

Student: Imraan Bux

MSc Thesis (16102)
April 2007
Graduation Committee

Professor H J Verhagen DUT, Faculty of Civil Engineering: Coastal Engineering

Professor D Roelvink UNESCO-IHE, Coastal Engineering, Port Development

Dr R Hassan UNESCO-IHE, Coastal Engineering, Port Development

THE EFFECTS OF FORESHORE SLOPE ON
BREAKWATER ARMOUR UNIT STABILITY

Master of Science Thesis

Imraan Bux

Supervisor

Professor H J Verhagen (TU Delft)

Internal Supervisor

Professor D Roelvink (UNESCO-IHE)

This research is done for the partial fulfilment of requirements for the Master of Science degree at the
UNESCO-IHE Institute for Water Education, Delft, the Netherlands

Delft
April 2007
The findings, interpretations and conclusions expressed in this study do neither
necessarily reflect the views of the UNESCO-IHE Institute for Water Education, nor of
the individual members of the MSc committee, nor of their respective employers. The
use of trademarks in any publication of UNESCO-IHE does not imply any endorsement
or disapproval of this product by the Institute. Core-Loc is a registered trademark of the
US Army Corps of Engineers and Accropode is a registered trademark of Sogreah
Consultants,France.
 DEDICATION

This work I dedicate to my most beloved grandmother: for her love, support, and
inspiration throughout my life. Even in your absence you will continue to
inspire me.

i
 ABSTRACT

Experience has shown that the stability parameters relating to breakwater

armour unit stability vary considerably for steep foreshores when compared
to milder foreshore slope. Furthermore, the change from the deep water
spectra (Tm0) to the shallow water wave spectra (Tm-1.0) does not fully
explain the difference. Leading on from recent laboratory tests that were
undertaken at the Research Flume at the Delft University of Technology
indicate that the stability of breakwater armour unit is further dependant on
parameters that are not described by the wave spectrum at the toe of the
breakwater. The skewness of the wave is one such parameter.

From recent tests that were conducted at TU Delft’s Laboratory it was found
that for equal deep water waves, the damage to breakwater armour unit was
more for a steep foreshore slope than that of a milder slope. It was also
found that the wave spectra are identical but the damage to the armour unit
vary considerably, implying that the damage to the breakwater has to
depend on a wave parameter that is not represented in the shallow water
wave energy spectrum. This parameter has to be cognisant of waves
approaching over different foreshore slopes.

It has also been established from recent laboratory tests that velocity and
acceleration are both relevant for the stability of rock on slopes. Wave
peakedness and asymmetry have to be also considered as they relate to
armour unit stability. Within the context of this study an attempt was made
at looking at the influence of foreshore on breakwater armour unit stability.
It was found that a clear correlation exists between foreshore slope and
armour unit stability. More damage to the armour units was noticed for
steeper foreshores as when compared to milder slopes. Furthermore, an
attempt was made at finding a correlation between the van der Meer
formual and the laboratory experiments. Since the van der Meer formula
does not consider foreshore slope directly and his definition of damage area

ii
differed to that which was used in the study, made such comparisons
difficult. Wave peakness was also found to be an important parameter
contributing to stability together with stone density and porosity,
obliqueness and angularity together with a proper definition of damage area.

This study proved the importance of foreshore slope on breakwater armour

unit stability and for the importance of further research in understanding the
other contributing parameters.

iii
 ACKNOWLEDGEMENTS

I would sincerely like to acknowledge the following organisations and

individuals:

The Blom Foundation for having sponsored my studies;

UNESCO-IHE for having afforded me this incredible study opportunity;

My supervisor, Professor Verhagen for his invaluable guidance in bringing this

study to fruition;
Professor Roelvink for his internal supervision;
To all my Peers, respected colleagues and dear friends, I thank you for the
enriching social, cultural and professional interactions.

iv
 TABLE OF CONTENTS

DEDICATION .................................................................................................................. i

ABSTRACT ..................................................................................................................... ii

ACKNOWLEDGEMENTS ............................................................................................ iv

TABLE OF CONTENTS ................................................................................................. v

LIST OF TABLES .......................................................................................................... vi

Chapter 1: Problem Definition and Research Methodology .......................................... 10

1.1 Functional Requirements................................................................................ 12
1.2 Problem Definition ......................................................................................... 12
1.3 Research Objective ......................................................................................... 14
1.4 Report Outline ................................................................................................ 14

Chapter 2: Theory........................................................................................................... 16
2.1 Stability formula of Iribarren.......................................................................... 16
2.2 Stability formula of Hudson ........................................................................... 18
2.3 Stability formula of van der Meer .................................................................. 19

Chapter 3: Experimental Set-up ..................................................................................... 21

3.1 General experimental set-up........................................................................... 21
3.2 Breakwater...................................................................................................... 22

Chapter 4: Analysis ........................................................................................................ 29

4.1 Data Analysis.................................................................................................. 29
4.2 Description ..................................................................................................... 34
4.3 Damage area and void ratio calculation ......................................................... 35

Chapter 5: Results........................................................................................................... 44
5.1 Results of Laboratory tests ............................................................................. 44
5.2 Comparisons based on erosion area................................................................ 46

Chapter 6: Conclusion and Recommendation ................................................................ 51

BIBLIOGRAPHY .......................................................................................................... 53

v
 LIST OF TABLES

Table 1: Laboratory measurement results

Table 2: Test parameters relating to program number

Table 3: Final datasheet of test results

Table 4: Summary of relevant tests

vi
 LIST OF FIGURES

Figure 1: Typical section through a rubble mound breakwater structure

Figure 2: Definition of foreshore slope

Figure 3: Breakwater layout

Figure 4: Breakwater Cross-section

Figure 5: Side elevation of the front side of the breakwater

Figure 6: Definition of damage area on breakwater slope

Figure 7: Rip-Rap armour after testing

Figure 8: Description of damage level on breakwater slope

Figure 9: Initial plot of the actual damage of laboratory results

Figure 10: Damage level in terms of S

Figure 11: Van der Meer formula

Figure 12: Graph of all damage levels

vii
 LIST OF SYMBOLS

β Bottom steepness (slope angle) [-]

ξ Iribarren (breaker) parameter [-]
∆ Relative mass density [-]
α Slope angle of breakwater [-]
Ae Damage area [m2]
Ae erosion area [m2]
B width of flume [m]
D Stone diameter [m]
d stone thickness or axial breadth [m]
D15 sieve size where 15% of the stones fits through [m]
D50 sieve size where 50% of the stones fits through [m]
D85 sieve size where 85% of the stones fits through [m]
Dn50 Nominal mean diameter of stone [m]
E(f) spectral density [m2/Hz]
f frequency [Hz]
fm peak frequency [Hz]
g acceleration of gravity [m/s2]
H Wave height [m]
Hm0,0 Hm0 at deep water [m]
Hs Significant wave height [m]
KD Breakwater stability co-efficient (Hudson) [-]
l maximum axial stone length [m]
L wavelength [m]
L0 deep-water wavelength [m]
M stone mass [kg]
mD dry mass of a stone [g]
mU underwater mass of a stone [g]
N number of waves [-]
Nod number of displaced stones (normalised) [-]
NS number of displaced stones [-]
P Permeability factor [-]
S Damage level [-]
ρs Density of stone [g/cm3]
ρw Density of water [kg/m3]
s wave steepness [-]
Tm-1,0 wave period (Van Gent) [-]
Tp peak period [s]

viii
Vs volume of a stone [cm3]
Vs volume of a stone [m3]
W stone weight [N]
z sieve size (side of smallest square hole) [m]
α scaling factor (Pierson-Moskowitz) [-]
γ0 scaling factor (Jonswap peak enhancement factor) [-]
σ, σa, σb scaling factor (Jonswap peak enhancement factor) [-]

ix
 Chapter 1: Problem Definition and Research Methodology

This introductory chapter will focus on the role and functional requirements of
rubble mound breakwaters together with the problem definition and research
methodology adopted. The proposed outline of the study will also be explained
in this chapter.

1.0 General Introduction

The most important functional requirement of all coastal structures is their

ability to withstand forces caused by wave attack. In the case of ports the
breakwaters play a vital role in this process. Breakwaters1 are structures placed
offshore to dissipate the energy of incoming waves. They are designed to
typically absorb the energy of the approaching waves which can be done by
either using mass (e.g. with caissons) or by using a revetment slope (e.g. with
rock or concrete armour units). Many breakwaters in the world presently are
rubble mound structures which use the voids in the structure to dissipate the
wave energy with the armour units absorbing most of the incoming energy. The
main strength parameter for most breakwaters is the type and mass of the armour
units. The interlocking abilities of the armour units to the breakwater are also
important and contribute significantly to its overall stability. Rubble mound
breakwaters are structures specifically built of quarried rock or other stone
material. Generally, the larger rock armourstones are used for the outer layer
which must withstand forces due to wave attack. The rocks or stones placed in
the outer layer is normally undertaken with more care to obtain better
interlocking and consequently better stability. Figure 1 shows a typical section
through a rubble mound breakwater structure.

1
The term breakwaters shall refer to rubble mound structures throughout this document

10
Figure 1: Typical section through a rubble mound breakwater structure

Rubble mound structures typically consist of a core usually made of small stone
or rocks, a filter layer and an armour layer. As stated above, larger stones are
normally used for the armour layer (outer layer) in order to withstand forces due
to wave attack. A toe is constructed on both sides of the structure which
contributes to the overall stability. A heavy toe built of rock or stone is
constructed on the seaward side of the breakwater. A concrete crown wall or
splash wall is then placed at the top of the breakwater.

The wave forces acting on a rubble mound breakwater slope can cause armour
unit movement which is termed as hydraulic instability. According to van der
H
Meer (1988), rubble mound structures can be classified according to their
∆D
parameter, where H is the wave height, ∆ is the relative mass density and D is
the characteristic diameter of the structure armour unit. Statically stable
structures are those were little or no damage is allowed under design conditions.
Damage is referred to the displacement of the armour units under wave attack.
The mass of the individual rock units within the armour layer must be large
enough to withstand wave forces during design conditions. Traditionally
designed rubble mound breakwaters are statically stable structures with
H
an =1 − 4.
∆D

11
1.1 Functional Requirements

Breakwaters generally serve to provide for safe navigable access for vessels by
delineating the harbour entrance channel and limiting the amount of energy
penetration into the harbour basin. The strong currents and wave forces are
limited by the breakwaters thus allowing for the safe berthing and mooring of
vessels.

Other functional requirements of breakwaters could include namely:

• Protection of the ports access channel and turning circle for safe stopping
and manoeuvring;
• Reduction of wave disturbance within the channel thus allowing for tugs
to operate;
• Morphological function by blocking the longshore transport of sediments
that could silt up within the entrance channel thus increasing dredging
costs.

1.2 Problem Definition

Many methods for the prediction of stability of armour units based on physical
model tests have been proposed. However, these tests were conducted with
deep water wave conditions at the toe of the structure and most breakwaters are
generally located in shallow water. Furthermore, the design formulae do not
take into account directly the effects of the foreshore slope on the stability of the
breakwater. Breakwaters that are constructed on relatively steep foreshores are
problematic from a stability point of view with significant damage to the armour
units.

Coastal structures situated on steep foreshores suffer significant damage than

can be normally expected from the same given boundary conditions at deep
water. Design guidelines as suggested in the CUR report 169 (Hudson formula
KD) compensate for the decrease in stability. Recent laboratory research

12
(Hovestad, 2005) confirmed that for equal deep-water waves the damage caused
by steep foreshores is greater than that for milder slopes. It is for this reason that
many guidelines particularly the CUR report recommends a heavier class of rock
in this situation. Manufactures of single armour units like Accropode, Core-Loc
and Xbloc recommend a lower KD value in the case of steep foreshores.

A first attempt into a more systematic analysis of the problem into shallow water
and steep foreshores was undertaken by Van Gent et al (2003). The results of
the research revealed that the load on the structure can better be described by a
shallow water spectrum at the toe of the structure and because the longer periods
are more relevant than the shorter periods more attention should be given to the
low frequency part of the spectrum. The Tm-1.0 should be used instead of the Tp
or Tm0 in the stability formula as it yields more reliable results according to Van
Gent et al. Van Gent following a statement of Van der Meer (1998) stated that
for shallow water conditions the 2% wave height should be used instead of the
significant wave height and the value of ξ should be calculated with the spectral
wave period Tm-1.0.

From engineering practice it is supposed that the steepness of the sea bottom in
front of a breakwater or sea defense, as can be seen in Figure 2, denoted as the
angle β may have influence on the stability of the armour units on these
structures.

Figure 2: Definition of foreshore slope

13
 The stability of breakwaters on steep foreshores is still a problem which resulted
in a series of experiments being undertaken at the research flume of the Fluid
Mechanics Laboratory at Delft University of Technology. As was expected, the
tests revealed for equal deep water waves, the damage caused to the armour unit
was more significant for steep foreshores when compared to shallow foreshores.

1.3 Research Objective

Recent studies undertaken at TU Delft (Hovestad, 2005), indicate that the wave
spectra are identical but the damage to the armour units is clearly not identical.
This implies that the damage to the breakwater is also dependant on a wave
parameter which is not represented by the shallow water wave energy spectrum
and has to consider waves approaching over different foreshore slopes.

What is the parameter not included in the wave spectrum which seems to be
relevant for breakwater armour unit stability on steep foreshores? What is the
physical relationship between steep foreshores and armour unit stability?

The research shall not only focus on the statistical relationship of the problem
(curve fitting) but will endeavour to establish the physical relation between steep
foreshores and armour unit stability which is presently unclear.

1.4 Report Outline

Leading on from the tests that were undertaken by Hovestad, 2005, an attempt
shall be made to determine the physical relationship between steep foreshores
and armour unit stability of breakwaters. In order to realise this, it is imperative
to examine the existing theories pertaining to breakwater armour unit analysis
and design. Presently, the most common are that of van der Meer, Hudson and
Iribarren. All three theories shall be examined in chapter 2 in order to determine
which theory supports the realisation of the research objective closest. In
chapter 3 an explanation of the experimental process shall be described. The
experiments that were recently carried out at the research flume of the Fluid

14
Mechanics Laboratory situated at Delft University of Technology in order to
understand the behaviour of breakwaters on different foreshore slopes. Chapter
four focuses on the data analysis of all the laboratory test results and what would
be important for the purposes of this study. It also explains all the derivations
that were used in the study. In chapter five the results and findings of the
analysis are discussed. Chapter six closes with the conclusion and
recommendations resulting out of this study.

15
 Chapter 2: Theory

This chapter will review the existing theories that are relevant to breakwater
armour unit stability in shallow water. The most common are that of van der
Meer, Hudson and Iribarren. The theories shall be examined in order to
establish which theory would support the research objectives the closest.

Many prediction methods for the stability of armour units have been proposed.
Most of these tests were conducted with deep water conditions at the toe of the
structure.

2.1 Stability formula of Iribarren

The Iribarren number also known as the surf similarity parameter was
introduced by Iribarren in 1938. This parameter was introduced by Iribarren
and Nogales as an indicator whether breaking would occur on a plane slope. As
discussed by Battjes (1974)-, the derivation of the Iribarren formula suggests
the parameter gives the ratio of the structure slope steepness to the square root
of the wave steepness as defined by the local wave height at the toe of the
structure divided by the deep water wave length.

Battjes (1974)-, popularised the Iribarren number (surf similarity number) and
showed its application to a number of surf zone processes, namely, wave
breaking, differentiation of breaker types, wave set-up, wave run-up and run-
down and wave reflection

As was already established from the above theories, the permeability of the
structure and foreshore slope are important parameters governing breakwater
armour unit stability.

16
Assuming that the velocity in a breaking wave or wave breaking on a slope
is proportional to the wave celerity in shallow water with the wave height as
being representative of the water depth: µ ∝ gH , Iribarren proposed the
following formula in 1938:

ρ w gHd 2 ∝ (ρ s − ρ w )gd 3 (tan φ cos α ± sin α ) (2.7)

“drag” force resisting force slope correction
Where:

ρw Density of water [kg/m3]

ρs Density of stone [kg/m3]

g Acceleration due to gravity [m/s2]
u Velocity parallel to the breakwater [m/s]
H Wave height [m]
d Diameter of stone [m]
φ Angle of internal friction [-]
α Slope angle of the structure [-]

By raising all the terms in the above formula to the third power and considering
( )
the mass of the stone M ∝ ρ s d 3 the Iribarren formula becomes:

ρ sH 3
M ≥ (2.8)
∆3 (tan φ cos α ± sin α )
3

Where:

M Mass of stone [kg]

∆ Relative buoyant density of the material [-]

From the above analysis it seems as if the Iribarren theory would best explain
the research objectives of this study. It considers the velocity and acceleration
of the armour unit on a slope and would thus be best suited in determining the

17
 parameters related to armour unit stability of breakwaters on steep foreshores.

2.2 Stability formula of Hudson

Extensive research has been carried out on the stability of stones on a slope in
the case of breakwaters. In general, the linear wave theory is used in the
calculation of wave heights. Linear wave theory can calculate wave heights at
any given depth as a function of a given off-shore wave height and a wave
period. However, it may be limited in that it does not consider the steepness of
the bottom slope.

The Hudson formula which is based on the Irribarren formula (the formula as
proposed by Irribarren is discussed later on in section 2.3) is well known and
was commonly used for the prediction of rock armour size for structures under
direct wave attack. After extensive tests Hudson proposed the following
formula:

ρ c *H 3
W= (2.1)
K D * ∆3 * cot α

Re-writing the above equation in terms of weight reveals:

1
Hs
= ( K D cot α ) 3 (2.2)
∆D n 50

The Hudson formula was based on waves that did not break at the toe of the
structure and no overtopping was considered. The KD value is more like a
“dustbin” co-efficient taking into consideration all other variables pertaining to
stability. The Coastal Engineering Manual provides values for KD based on
specific circumstances and conditions. Various KD values have been derived for
a wide range of armour units and configurations and, due to its simplicity it was
widely used.

18
 However, the Hudson formula has limitations in that:

• Potential scale effects due to small scales under which most of the tests
were conducted;
• It uses regular waves;
• No consideration is given to wave period and storm duration;
• It does not describe the damage level;
• It does not consider wave overtopping and is related to permeable
structures only;
• It does not take into consideration foreshore slope.

The use of KD and cot α does not always best describe the effect of the
foreshore slope.

2.3 Stability formula of van der Meer

Based on the earlier work of Thompson and Shuttler (1975), van der Meer
undertook a series of extensive tests at the Delft Hydraulics Laboratory (1987).
The tests included a wide range of structures for different core and underlayer
permeabilities over a wide range of wave conditions. Van der Meer’s research
was based on a large data set and considered deep water conditions at the toe of
the structure for plunging and surging waves. Finally van der Meer proposed
the following formulas:

0.2
Hs  S 
= 6.2 P 0.18   ξ −0.5 Plunging waves ξ <ξ transition (2.3)
∆D n 50  N

0.2
Hs  S 
= 1.0 P −0.13   cot α ξ p Surging waves ξ >ξ transition (2.4)
∆D n 50  N

The transition from plunging to surging waves can be calculated using a

critical value of ξ .

19
 ( )
1
ξ transition = 6.2 P 0.31 tan α P + 0.5 (2.6)

Where:

Hs Significant wave height [m]

Dn50 Nominal mean diameter [m]
P Permeability factor [-]
S Damage level [-]
N Number of waves [-]
ξ Surf similarity parameter [-]

The reliability of these formulas depends on the difference due to random

behaviour of rock slopes, accuracy of measuring damage and curve fitting. P is
a measure of the permeability of the structure, S a measure of the damage and N
is the number of waves. The van der Meer formula seems to be more involved
when compared to the Hudson formula in that more parameters are included like
the Iribarren number (wave steepness), the porosity (permeability) of the
structure and the damage level, however it is still limited in the case of steep and
shallow foreshores. Furthermore, it considers deep water conditions at the toe of
the structure.

The slopes considered by Hudson and van der Meer were in the range 1:15 to
1:6 which can be considered as mild slopes. It can then be deduced that the van
der Meer formula would not be able to satisfy the research objectives of this
study were steep foreshore slopes and shallow water conditions are considered.
However, the van der Meer formula for the plunging case which is based on a
zero foreshore slope will be used later on in the analysis when making
comparisons to the laboratory experiments.

20
 Chapter 3: Experimental Set-up

For the purpose of completion, this chapter shall illustrate the experimental
process that was undertaken at TU Delft’s laboratory. The experiment as
described by Hovestad (2005) shall be used as the analysis in this study is based
on his data set.

3.1 General experimental set-up

The dataset is based on experiments that focussed on the different influences of

armour unit stability considering foreshore slope. As mentioned earlier, the
design of breakwaters on steep foreshores is problematic and the design
formulas do not take into account the effects of foreshore slope on armour unit
stability. The main experimental questions were to investigate whether armour
unit stability is affected by foreshore slope and if so, how large are these
influences?

In the experiments two foreshore slopes were considered, namely a 1:302 and a
1:8 foreshore slope. The values of the foreshore slope steepness were based on
practical considerations. The values had to also be as far apart as possible in
order to create the biggest possible effect. The 1:30 slope was chosen as the
mild case as the floor to the flume in the laboratory is permanently fixed to this
inclination. The 1:8 slope was a result of optimisation, it needed to be as steep
as possible, but making it too steep would result in the horizontal length of the
slope becoming too short in relation to the wave lengths to be used. If the
horizontal length became too short, this could impact on the ability of the wave
to shoal correctly. Thus the steepness was chosen in such a way that the length
of the slope was approximately twice the wave length (the wave length in this
case defined at the deepest part of the flume, based on the peak period). This
process resulted in the 1:8 foreshore slope.

2
1:30 shall refer to a mild foreshore slope and 1:8 for a steep slope throughout this document

21
 3.2 Breakwater

The tests on the breakwater were conducted in the “Lange Speurwerkgoot”-

(Long Research Flume) of the Fluids Mechanics Laboratory of the Faculty of
Civil Engineering and Geosciences at Delft University of Technology.

The dimensions of the flume were 80cm wide 42m long and had a maximal
depth of 100cm. The dimensions of the flume especially the width will be
considered later on when the damage level on the breakwater slope will have to
be expressed in terms of damage level S. The damage level in these tests were
expressed in terms of damage level N and, within the context of this study
damage level will be expressed in terms of damage level S in order to make a
comparison with the van der Meer equation.

Wave
1m

Gen.
50cm 1:30

42m 8.5m

Wave
Gen.
1m

50cm 1:8

Figure 3: Breakwater layout

The layout of the breakwater in the flume can be seen in figure 3. The
breakwater was in all series constructed at 50cm above the reference level with
the top of the breakwater just reaching the chain rail thus making it 45cm high.
The front slope was constructed at a 1:2 gradient and the rear slope to a 1:1.5
grade. The crest was 25cm wide. As can be seen from figure 4, the rock armour
layer did not extend all the way to the top of the breakwater as this was not

22
 necessary as most of the wave impact occurred exclusively in the lower zone.
Figure 4 and figure 5 show the cross-section of the breakwater in the flume.

25

1:2
1:1.5
45

16

183

Figure 4: Breakwater Cross-section

Wave gauge at toe

Armour

Toe
Filter

Core

Figure 5: Side elevation of the front side of the breakwater

The selection of the stones was done by measuring a sample of stones with a
slide gauge. The definition of the stone dimensions was according to the CUR
manual, 1994, where:

z sieve size (stone that fits through the small sieve hole)
l maximum axial length
d thickness or axial breadth

23
The definition was also used in the selection process were the CUR manual
advises to limit the amount of stones with a ratio of l/d >3 to an amount of 3 to 5
percent.

After an iterative process the D50 of the stones (ie. the sieve size through which
50% of the stone passes) was found to be 2.02cm and the D85 was 2.22cm. The
D15 was 1.78cm thus yielding a D85/D15 ratio of 1.25cm.

From the above weights the volume of the stones were then calculated. The Dn50
and the stone density were calculated according to the following expression:

mD
ρs = ρw (3.1)
mD − mU

and

Dn = 3
(mD − mU ) (3.2)
ρw
ρw was taken at a constant 1000 kg/m3

where:

Vs volume of a stone (cm3)

ρs density of the stone (g/cm3)
ρw density of water (g/cm3)
mD dry mass of the stone (g)
mU underwater mass of the stone (g)

The stones that were chosen, had a Dn50 (i.e. a median Dn) of 1,57cm.

In the calculation of the damage area on the breakwater slope, the area will be
expressed in terms of the Dn50 of the stone. Stone densities varied from
2630 kg/m3 to 2980 kg/m3. The weighted average was thus taken as 2780 kg/m3.

24
Due to the differences in densities, the damage estimation may be affected as the
stability of the stones is heavily influenced by the density. The thickness of the
armour layer was according to CUR guidelines which recommend a layer of at
least 2dn50. As large numbers of damage was expected in these tests the armour
layer was doubled resulting in a layer thickness of 6cm.

For the purposes of completion it is worth noting that in the wave height
calculation used in this study which is based on the experiment conducted,
irregular waves were used. This was done in order to simulate a “real” sea state
in front of the breakwater so that the results could be translated to prototype
situations. The waves were generated according to the standard Jonswap-
spectrum which describes a young sea state as opposed to the Pierson-
Moskowitz spectrum which describes a fully developed sea state which would
hardly occur in nature.

The standard Jonswap-spectrum is described by:

 
 5  f  −4  exp  − 12  fσ−f f m  
E ( f ) = αg (2π )
2 −4
f −5
exp −   γ 0   m   (3.3)
 4  fm  
 

with:

E spectral energy density [m2/Hz]

α scaling parameter (Pierson-Moskowitz) [-]
f frequency [Hz]
fm peak frequency [Hz]
γ0 scaling parameter (Jonswap peak-enhancement factor) [-]
σ scaling parameter (Jonswap peak enhancement factor) [-]

The last part of this equation is called the peak enhancement factor.

25
The value of σ changes according to the frequency:
σ = σa if f < fm
σ = σb if f > fm

For the standard Jonswap spectrum, the values in the peak-enhancement factor
are:
γ0=3,3
σa=0,07
σb=0,09

These values were used in all the experiments. The theory on wave spectra is
explained according to BATTJES (1992).

H
The wave steepness was defined as s =
L
with:

s wave steepness [-]

H wave height [m]
L wave length [m]

From the above analysis a deep water wave height of 12cm was used based on
different wave steepness’s. The measured wave heights at the toe and in deep
water together with the wave steepness shall be used later in the calculation of
the damage area to the breakwater on different foreshore slopes.

It is important to note that during this experiment all the stones that were not in
their original position as part of the coloured profile were counted. This causes
a problem when comparing the results to the van der Meer equation who defines
damage level as the area of the erosion in the cross-section of the breakwater
divided by the square of the stone size.

26
 Ae
S= (3.4)
d n250

With:

S damage level [-]

Ae Erosion area [m2]

It is imperative to understand this definition within the context of this study as

the results of the experiment based on a 1:30 and 1:8 foreshore slope shall be
compared to the van der Meer formula. As can be seen from figure 6 if a stone
is removed from its original band but remained within the erosion area it will be
counted as damage if just the displacements of the stones are counted. However,
according to the definition of damage by van der Meer, which is based on the
erosion area of the cross-section, this will amount to no damage. The counting
of the stones method will thus over-estimate the damage level when compared to
the van der Meer formula. A comparison will be made in the following chapter
in order to see the relationship based on the experimental tests and that of the
van der Meer formula.

Also important to note is that according to the van der Meer and van Gent
formula the damage should grow with the 5th power of the wave height. Both
formulas have a term H ∝ S 0.2 or inversely S ∝ H 5 . This expression becomes
important in the next chapter when the van der Meer equation has to be
expressed in terms of damage level S.

Figure 6: Definition of damage area on breakwater slope

27
All the tests conducted in the experiment were undertaken using N=1000 waves
which is common in breakwater research. It is also important to note at this
point, as this will influence the comparisons being made in the following
chapters: the tests that were conducted on the stone armour layer (rip-rap)
showed a strong influence on the wave steepness and the target wave height was
used enabling the stones that were moved from their original band to be counted.

The normal displacement of stones that would be expected to occur under calm
wave conditions did not occur resulting in a higher damage level when
compared to the van der Meer formula. The definition of damage is thus critical
in understanding the comparisons being made. Figure 7 shows the damage on
the armour layer after being subjected to testing. The damage can be seen on the
lower part of the breakwater slope.

Figure 7: Rip-Rap armour after testing

The majority of the tests were undertaken to make comparisons on steep

foreshores. Four different wave steepness’s were used: so=0.030, 0.044, 0.058
and 0.086 respectively. The wave height at the board varied between
Hmo=11.4cm and Hmo=12.1cm. These values will be used later when calculating
the van der Meer equation for damage.

28
 Chapter 4: Analysis

This chapter shall focus on the analysis of the results undertaken in the
laboratory when compared to the van der Meer formula for breakwater armour
unit stability, in order to determine what correlation exists between damage level
and foreshore slope.

4.1 Data Analysis

During the laboratory testing undertaken by Hovestad (2005), large amounts of

data were available which was undertaken for damage of breakwater armour
layer based on the two different foreshore slopes mentioned earlier. The focus
of this research was to find that parameter not included in the wave spectrum
which seems to be relevant for armour unit stability. As previously mentioned,
in the case of coasts with steep foreshores, coastal structures experience more
damage than can be normally expected from given boundary conditions at deep
water.

It is primarily for this reason that most design guidelines and manufactures of
armour units suggest using a heavier class of rock or a lower Kd value in the
case of steep foreshores. As stated in the introduction a first step into the more
systematic approach to addressing the problem associated with shallow water
and steep foreshores was introduced by Van Gent et al (2003), were the results
revealed that the load on the structure can best be described by a shallow water
wave spectrum at the toe of the structure. Furthermore, since the longer period
waves are more significant in this case than the shorter period waves, more care
should be given to the lower frequency part of the spectrum.

As a first step into the data analysis, all the data was reviewed in order to
establish which data files contained complete records that would be used to
answer the research question in this study. Various tests were conducted but not
all had a complete set of records relevant to this study. All the tests that were

29
performed were tabulated in terms of time series and frequency series with their
associated program number. As a first step each test number was sorted to
match its corresponding times series and frequency series. The test results
included the test number, the program number and the damage level expressed
in terms of Ns and Nod. Table 1 shows a typical result summary table indicating
all the relevant parameters. The values of tan α, Dn50, the flume width B etc
shall be used in the damage area calculation later. The permeability of the
structure has an influence on the stability of the armour layer and the notional
permeability factor was taken as P=0.6 (no filter and no core) in order to fully
understand the displacement of the stone within the armour layer. This was
based after the notional permeability factors as described by van der Meer
(2003).

30
 dn50 0.0157 (m) tanα 0.50
B 0.8 (m) ∆ 1.78
P 0.6
N 1000 (T01040/41/42: 2000)

Testnumber Program Wh Rd Bl Pk Ye Gr Rd+ Ye+ Gr+ Bk NS Nod

T00010 M05D66 4 15 15 8 4 46 0.90
T00011 M06D66 6 24 51 41 23 1 146 2.87
T00012 M07D66 8 56 100 74 43 6 287 5.63
T00013 M08D66 15 73 138 123 113 35 20 517 10.15
T00014 M09D66 18 125 192 185 209 150 23 902 17.70
T00015 M10D66 20 238 224 211 285 261 57 1296 25.43
T00016 M11D66 31 324 344 225 298 324 55 30 1631 32.01
T00020 M05D66 2 19 23 13 1 58 1.14
T00021 M06D66 6 35 59 37 10 1 148 2.90
T00022 M07D66 10 55 108 82 59 4 318 6.24
T00023 M08D66 13 86 158 147 123 40 1 568 11.15
T00024 M09D66 16 116 191 192 186 127 11 839 16.47
T00025 M10D66 24 223 224 227 286 211 42 21 7 1265 24.83
T00026 M11D66 36 273 257 258 425 303 54 29 13 1648 32.34
T00100 M05D66 9 26 43 9 87 1.71
T00101 M06D66 14 49 96 82 8 249 4.89
T00102 M07D66 26 80 164 183 82 6 541 10.62
T00103 M08D66 33 123 249 346 266 75 1092 21.43
T00104 M09D66 39 148 295 415 391 284 9 1581 31.03
T00110 M05D66 1 18 38 7 64 1.26
T00111 M06D66 9 51 83 51 9 203 3.98
T00112 M07D66 15 94 190 207 101 7 614 12.05
T00113 M08D66 22 115 238 311 231 126 1043 20.47
T00114 M09D66 30 156 286 386 364 254 3 1 1480 29.05
T00120 M05D66 5 36 31 5 77 1.51
T00121 M06D66 9 70 81 39 1 200 3.93
T00122 M07D66 18 131 164 162 60 3 538 10.56
T00123 M08D66 30 196 237 303 231 78 1075 21.10
T00124 M09D66 37 228 296 410 387 256 3 1617 31.73
T00130 M08D66 10 145 214 260 210 30 869 17.05
T00131 M08D66 10 138 200 262 162 19 791 15.52
T00132 M08D66 12 155 240 255 163 6 831 16.31
T00140 M08B66 18 175 357 261 339 258 6 1414 27.75
T00141 M08B66 7 176 258 371 337 275 1 1425 27.97
T00142 M08B66 14 163 251 371 364 292 10 1465 28.75
T00150 M08C66 15 171 239 323 271 122 1141 22.39
T00151 M08C66 21 207 221 274 214 74 1011 19.84
T00152 M08C66 10 215 229 329 268 113 1164 22.84
T00160 M08E66 14 162 160 165 37 1 539 10.58

Table 1: Laboratory measurement results

As can be seen from table 1, each test had a unique set of parameters that had to
be related to their program number thus obtaining their wave period, wave
height and wave steepness. The tests were then matched to their relevant
programme numbers in order to obtain the relevant variables pertaining to this

31
 study. Wh (white), Rd (red), Ye (yellow) all refer to the colours of the stones
that were removed from that band during testing. As can also be seen the
damage level is expressed in terms of Ns and Nod.

For the purposes of this study the damage level taken from table 1 shall be
transformed and expressed in terms of damage level, S. Table 2 shows a typical
layout of the tabulation of the test results corresponding to their program
numbers. Relating each test to their relevant test number and program numbers,
the dataset could thus be completed.

Program M05D66 M06D66 M07D66 M08D66 M09D66 M10D66 M11D66 M08E66 M08F66
T (s) 0.89 0.98 1.06 1.13 1.2 1.27 1.33 1.01 0.92
Hm0,b (m) 0.071 0.084 0.098 0.111 0.124 0.137 0.151 0.112 0.113
hb (m) 0.66 0.66 0.66 0.66 0.66 0.66 0.66 0.66 0.66
Hm0,0 (m) 0.071 0.086 0.101 0.116 0.131 0.147 0.162 0.114 0.114
s0 (-) 0.057 0.057 0.057 0.058 0.058 0.058 0.059 0.072 0.086

Table 2: Test parameters relating to program number

Each test number from table 1 was compared to the corresponding programme
number from table 2 in order to complete the data set that would include all the
parameters relevant to this study. The tests were done for both foreshore slopes.
Table 1 shows a typical compilation of the test information pertaining to one set
of test only; this is shown for illustrative purposes as the complete data set in not
entirely relevant. From this the data set could now be matched for their
associated time series and frequency series as can be seen in table 3.

32
 Time Series Frequency Wave height Foreshore slope
Test Wave Damage
Number
number (File name) (File name) Toe Deep water period 1:30 1:8 level Ns
1 T00010 0.071 0.071 0.89 46
2 T00011 0.084 0.086 0.98 146
3 T00012 0.098 0.101 1.06 287
4 T00013 0.111 0.116 1.13 517
5 T00014 0.124 0.131 1.2 902
6 T00015 0.137 0.147 1.27 1296
7 T00016 0.137 0.147 1.27 1631
8 T00020 0.151 0.162 1.33 58
9 T00021 0.084 0.086 0.98 148
10 T00022 0.098 0.101 1.06 318
11 T00023 0.111 0.116 1.13 568
12 T00024 0.124 0.131 1.2 839
13 T00025 0.137 0.147 1.27 1265
14 T00026 0.151 0.162 1.33 1648
15 T00100 R00100DN S00100DN 0.071 0.071 0.89 87
16 T00101 R00101DN S00101DN 0.084 0.086 0.98 249
17 T00102 R00102SN S00102DN 0.098 0.101 1.06 541
18 T00103 R00103DN S00103DN 0.111 0.116 1.13 1092
19 T00104 R00104DN S00104DN 0.124 0.131 1.2 1581
20 T00110 R00110DN S00110DN 0.071 0.071 0.89 64
21 T00111 R00111DN S00111DN 0.084 0.086 0.98 203
22 T00112 R00112DN S00112DN 0.098 0.101 1.06 614
23 T00113 R00113DN S00113DN 0.111 0.116 1.13 1043
24 T00114 R00114DN S00114DN 0.124 0.131 1.2 1480
25 T00120 R00120DN S00120DN 0.071 0.071 0.89 77
26 T00121 R00121DN S00121DN 0.084 0.086 0.98 200
27 T00122 R00122DN S00122DN 0.098 0.101 1.06 538
28 T00123 R00123DN S00123DN 0.111 0.116 1.13 1075
29 T00124 R00124DN S00124DN 0.124 0.131 1.2 1617
30 T00130 0.111 0.116 1.13 869
31 T00131 0.111 0.116 1.13 791
32 T00132 0.111 0.116 1.13 831
33 T00140 0.110 0.120 1.6 1414
34 T00141 0.110 0.120 1.6 1425
35 T00142 0.110 0.120 1.6 1465
36 T00150 0.110 0.118 1.31 1141
37 T00151 0.110 0.118 1.31 1011
38 T00152 0.110 0.118 1.31 1164
39 T00160 0.112 0.114 1.01 539
40 T01000 R01000DN S01000DN 0.110 0.120 1.60
1482
41 T01001 R01001DN S01001DN 0.110 0.120 1.60
1417
42 T01002 R01002DN S01002DN 0.110 0.120 1.60
1386
43 T01010 R01010DN S01010DN 0.110 0.118 1.31
948
44 T01011 R01011DN S01011DN 0.110 0.118 1.31
990
45 T01012 R01012DN S01012DN 0.110 0.118 1.31
1011
46 T01020 R01020DN S01020DN 0.111 0.116 1.13
653
47 T01021 R01021DN S01021DN 0.111 0.116 1.13
664
48 T01022 R01022DN S01022DN 1.13
717
49 T01023 R01023DN S01023DN 0.111 0.116 1.13
694
50 T01030 R01030DN S01030DN 0.113 0.114 0.92
284
51 T01031 R01031DN S01031DN 0.113 0.114 0.92
295
52 T01032 R01032DN S01032DN 0.113 0.114 0.92
277
53 T01040 R01040DN S01040DN 1.13
1072
54 T01041 R01041DN S01041DN 1.13
1056
55 T01042 R01042DN S01042DN 1.13
1005
56 T02000 R02000DN S02000DN 0.110 0.12 1.6
1786
57 T02010 R02010DN S02010DN 0.110 0.118 1.31
1455
58 T02011 R02011DN S02011DN 0.110 0.118 1.31
1394
59 T02020 R02020DN S02020DN 0.111 0.116 1.13
1102
60 T02030 0.113 0.114 0.92
632
61 T02050 R02050DN S02050DN 0.096 0.102 1.22
973
62 T02051 R02051DN S02051DN 0.096 0.102 1.22
947
63 T02060 0.105 0.112 1.31
1311
64 T02070 R02070DN S02070DN 0.096 0.104 1.31
1082
65 T02071 R02071DN S02071DN 0.096 0.104 1.31
1106
66 T02080 R02080DN S02080DN 0.098 0.102 1.13
915
67 T02081 R02081DN S02081DN 0.098 0.102 1.13
890
68 T02090 R02090DN S02090DN 0.096 0.105 1.6
1390
69 T02091 R02091DN S02091DN 0.096 0.105 1.6
1466
70 T02100 0.099 0.1 0.92
458
71 T02110 S02110SN 0.094 0.095 0.92
395
72 T02111 0.094 0.095 0.92
359
Table 3: Final datasheet of test results

33
 Tests numbers that begin with 00 indicate preparatory tests that were undertaken
to obtain the best configuration and layout of the breakwater. Test numbers that
begin with the 01 indicate measurements that were undertaken for a mild
foreshore slope and tests that begin with the numbers 02 indicate tests on a steep
foreshore slope. After this was completed and the results tabulated in table 3,
the tests that have complete records that would be beneficial to this study can be
seen. The tests that were used in this analysis for the different foreshore slope is
ticked in table 3. In table 3 a link was created between the time series, the
frequency series and the test data in order to obtain a complete picture of the
testing procedure and data. Once this picture of complete and relevant records
was obtained the analysis in terms of damage level and comparisons can now be
undertaken.

4.2 Description

In order to obtain a more reliable and accurate correlation between the measured
laboratory results and the van der Meer equation which is based on a zero
foreshore slope, the void ratio of the stones had to be calculated and the damage
level on the breakwater slope expressed in terms of dn50. The erosion area had to
be calculated and expressed in dn50 considering the width of the flume and the
van der Meer equation re-written in terms of damage level, S.

Ultimately, it is envisaged to plot three curves, the first curve will be for a mild
foreshore slope, the second for a steep foreshore slope based on the laboratory
experiments and the third curve will be the van der Meer formula. In order to
make an adequate comparison with the van der Meer formula the results in the
laboratory will have to be expressed in terms of damage level, S. For this
comparison, the erosion area on the breakwater slope will have to be calculated.
The erosion area should consider as is as practicable as possible the void ratio of
the stones, the porosity and density of the stones, the packing density and the
obliqueness.

34
4.3 Damage area and void ratio calculation

The damage level as defined by van der Meer is given by the expression:

Ae
S= (4.1)
d n250

The damage of the armour layer can be given as a percentage of the displaced
rocks related to a certain area (the whole or part of the layer). In general it is
difficult to compare various structures as the damage figures are related to
different totals for each structure and test condition. Another possibility is to
describe the damage by the erosion area around the still water level. When this
erosion area is related to the dn50 of the rock, a dimensionless damage level can
be obtained which is independent of the size, slope angle and height of the
structure. This damage level is defined by equation 4.1 and the extent of Ae can
be seen in figure 8.

Erosion area
Original
Ae
Cross
Section

Ve

Ae

Figure 8: Description of damage level on breakwater slope

35
The damage should consider both the settlement and displacement of the rocks.
Another description of S is the number of cubic stones with a side of Dn50 that is
eroded within a Dn50 wide strip of the structure, the approach that will be
adopted in this study. The actual number of stones eroded will be dependant on
the porosity, the stone grading and the obliqueness of the stone (shape of the
stone).

The erosion area on the breakwater slope will now be calculated in order to
apply equation 4.1 and express the damage level in terms of S. All the test
results will then be factored by this and a new plot generated in terms of wave
period and damage level, S. The outputs of the different plots shall be discussed
in the next chapter entitled results.

Considering the erosion area Ae in figure 8 and expressing in terms of dn50

reveals:

Since the stones are spherical and not all the stones have the same shape, the
erosion area shall be expressed in terms of dn.

4 3
Volume of sphere = πr
3

W ≡ ρ x dn503

Calculating in terms of dn, volume equals

3
4 1  1
π  d  = π d 3 ≈ 0 .5 d 3
3 2  6

⇒ d n50 = 3
0.5 d n 50 (4.2)

= 0.81d 50

36
 The value of d will have to be incorporated into the damage level area
calculation in order for it to be expressed in terms of S.

The void ratio of the stones will now have to be considered.

Va

VT

Vs

VT = Va + Vs

Va
Void ratio e =
Vs
Va = air + water.

In the laboratory experiment the layer thickness was taken as 6cm which is
double when compared to the Manual on the use of rock in coastal and shoreline
engineering (CUR, 1991) which advises a layer thickness of at least 2dn50. The
layer thickness was doubled in the experiment as large scale damage was
anticipated.

37
In the calculation of the void ratio, the dimensions of the flume especially the
width (80cm) used in the laboratory testing will have to be considered.

The total volume of rock in the armour layer is given by:

t × 0.8 × L (see illustrative diagram above)

0 .8
Now the number of stones per metre width of flume = = 37 stones
d 50

dn

Volume of cube V = dn 3
dn = 0.81 d50 (from expression 4.2 above)

Substituting dn = 0.81 d50 into the volume of the cube we get:

38
V = dn 3

= (0.81 d50)3 This is the volume of 1 stone.

An assumption is made that the stones are spherical but in reality this is not the
case, hence the co-efficient will be somewhat greater than 0.81.

L
Number of stones = 2 x 37 x (the armour layer consisted of 2 layers)
d n 50

Total volume of stones = number of stones x volume of 1 stone

 l  
Vstones =  2 x 37 x  x (0.81 d 50 )3  (4.3)
 d n 50  

In order to calculate the total volume we need the volume of voids as the total
volume is equal to the volume of stone + volume of voids. VT = Vv + Vs.

We now need to calculate the volume of the space (voids).

Vspace = B x t x L

Volume of space Vspace = 0.8 x 0.06 x L (4.4)

Where 0.8 is the width of the flume, 0.06 is the layer thickness (2 layers) and L
is the length of the section, assumed to be 1m in this case.

39
Total volume VT = Vv + Vs

= (Vspace x e) + Vs

= Vs (1 + e) volume of the erosion area on the breakwater

Vv = Vspace x void ratio

Having worked out this expression we now need to calculate the void ratio e in
order to compute the total volume.

Vv V space − V stone
Void ratio e = = (4.5)
Vs V stone

Vspace = B x t x L from 4.4 above

 L  
Vstone =  2 x 37 x  x (0.81 d 50 )3  from 4.3 above
 d n 50  

Substituting Vspace and Vstone into expression 4.5 reveals:

Vv V space − V stone
Void ratio e = =
Vs V stone

 L  
(0.8 x 0.06 x L ) −  2 x 37 x  x (0.81 d 50 )3 
 d n 50  
= (4.6)
 L  
 2 x 37 x  x (0.81 d 50 )3 
 d n50  

40
From expression 4.6 we can now calculate the void ratio as we have the d50 and
L.

VT = Vs (1 +e)

VT = (0.81 d50)3 x N (1+e)

Equating the (erosion area x the width of the flume) to VT we get:

Width of flume x A = volume of 1 stone x number of stones + the void ratio.

This reveals,

(0.8 x A) = (0.81 d50)3 x N (1+e)

Re-writing this expression in terms of A

0.81
d 50 (1 + e) N
3
A=
B

0.81
d 50 (1 + e) N
3
A= (4.7)
0 .8

In the above expression B equals the width of the flume, d50 the stone size, e is
the void ratio (expression 4.6) and N is the damage level from the laboratory
experiments.

We have now arrived at an expression for the damage area A and equation 4.7
will now be used to calculate the areas and then expressed in terms of damage
level, S. We know the value of N (from laboratory tests), the value of d50 and
the void ratio (expression 4.6). Equation 4.7 will thus be used to transform all
the tests performed in the laboratory and express them in terms of damage level,
S, the tests that are ticked in table 3 shall only be considered. Once the areas are
calculated equation 4.1 will be used to express them in terms of S. These new

41
values of S will be plotted against wave period and the various output graphs
shall be discussed in the following chapter.

We have now expressed all the laboratory tests in terms of damage level, S and
in order to make a comparison to the van der Meer formula we need to express
the van der Meer formula in terms of S. Van der Meer proposed the following
formula for the plunging case. This formula will have to be re-written in terms
of S. Van der Meer described conditions for the surging and plunging case,
within the context of this study the plunging case will only be considered as the
laboratory experiments were conducted for the plunging case.

0.2
Hs  S 
= 6.2 P 0.18   ξ −0.5 Plunging waves ξ <ξ transition (4.8)
∆D n 50  N

where:

Hs significant wave height ≡ H1/3 [m]

M
dn50 nominal diameter ≡ 3 50 [m]
ρs
M50 median stone mass [kg]

P notional permeability [-]

A
S damage number ≡ 2e [-]
d n 50
Ae cross section of the erosion area [m2]

N number of waves [-]

ξα0 Iribarren-parameter using deep water steepness and breakwater slope

tan α
≡ [-]
H0
L0

42
Re-writing expression 4.8 in terms of S reveals:

0.2
Hs  S 
= 6.2 P 0.18   ξ −0.5
∆D n 50  N

0.2
Hs ξ 0.5  S 
x =  
∆D n 50 6 .2 P 0.18
 N 

5
 Hs ξ 0.5  S
 x 0.18 
=
 ∆D n50 6.2 P  N

5
 Hs ξ 0.5 
 x 0.18 
x N = S
 ∆Dn 50 6.2 P 

( N)
5
 H ξ 0.5
0.2

S =  s
 (4.9)
∆
 n 50
D 6.2 P 0.18 

This derivation supports the theory of van der Meer who states that the damage
should grow with the 5th power of the wave height.

We have now expressed the van der Meer formula in terms of damage level, S.
This can now be plotted with damage level, S against period and comparisons
can now be made to the laboratory tests on different foreshore slopes. A curve
of the van der Meer formula is plotted against period and superimposed on the
same system of axis as the other tests and will be discussed in the next chapter.

43
 Chapter 5: Results

The focus of this chapter shall be on the results of the analysis undertaken in the
previous chapters. Comparisons will be made with the laboratory tests on
different foreshore slopes when compared to the van der Meer formula which is
based on a zero foreshore slope.

5.1 Results of Laboratory tests

Test Tp N Hm0,toe Hm0, Test Tp N Hm0,toe Hm0,

No (s) (m) deep No (s) (m) deep
T01000 1.6 1482 0.11 0.12 T02000 1.6 1786 0.11 0.12
T01001 1.6 1417 0.11 0.12
T01002 1.6 1386 0.11 0.12
T01010 1.31 948 0.11 0.118 T02010 1.31 1455 0.11 0.118
T01011 1.31 990 0.11 0.118 T02011 1.31 1394 0.11 0.118
T01012 1.31 1011 0.11 0.118
T01020 1.13 653 0.111 0.116
T01021 1.13 664 0.111 0.116 T02020 1.13 1102 0.111 0.116
T01023 1.13 694 0.111 0.116

Table 4: Summary of relevant tests

Table 4 is a summary of all the relevant tests that have complete records. The
test numbers in table 4 are extracted from the comprehensive dataset in table 3.
All the comparisons in this study were based on the dataset as reflected in table
3. A first step would be to plot a graph showing the exact damage in terms of
damage level Ns against wave period without calculating the “damage area” and
factoring the Ns values. Figure 9 shows the relationship between damage level
Ns from the laboratory experiments and wave period.

44
 Laboratory tests
Number of stones moved Ns

1800
1600
1400
1200
1000
800
600
400
200
1.1 1.2 1.3 1.4 1.5 1.6 1.7
Tp (s)
1:30 slope 1:8 slope

Figure 9: Initial plot of the actual damage of laboratory results

Figure 9 shows the actual damage as calculated in the laboratory experiments

when plotted against wave period. The damage in the laboratory experiments
were based on Ns which is the total number of stones moved.

The basic settings for this experiment were:

Wave height (at the wave board) Hm0,0 : 12.1 cm
Wave period Tp : 1.6 seconds
Water depth : 66 cm
Rock size Dn50 : 1.57 cm
ρ : 2780 kg/m3

In order to simulate actual conditions and make reliable comparisons, in the first
series of tests the wave height at the wave maker was kept identical and in the
second series the wave at the toe was kept identical. This was done in order to
keep the spectra the same were the incoming spectrum at the toe of the
breakwater was exactly the same for both the 1:8 and 1:30 foreshore slope.

As can be seen from figure 9, for equal deep water waves, the damage level in
the case of a steep foreshore slope is more when compared to a mild slope.

45
 Furthermore, with the tests that had identical waves at the toe of the breakwater
the results indicated a significantly higher damage for the steeper slope
especially when the wave steepness is considered.

This yields interesting results: because the spectra are identical but the damage
is clearly not identical when comparing the cases of the different foreshore
slopes means that there is a parameter which is not represented by the shallow
water wave energy spectrum for waves approaching over different foreshore
slopes. One observation is clearly evident that the damage level is greater for a
steep foreshore slope when compared to a milder slope even when the damage
level is based on the actual number of stones displaced.

It is also important to note at this point that when a plot of ξ (surf-similarity

parameter) is generated against increase in damage, one notices the damage level
increasing with increasing alpha values. This means that higher ξ values yield
lower damage and vice versa. Furthermore, important to note is that most
graphs are generally plotted in terms of ξ , but the graphs in this study are
plotted in terms of wave period Tp. The primary reason for this is that the slope
is constant in this case and Hm,0 is kept the same at the toe and deep water,
therefore, plotting the graphs in terms of period is the same as plotting them in
terms of ξ .

5.2 Comparisons based on erosion area

Having explained the actual damage level from the laboratory experiments and
expressed their relationship in terms of damage based on foreshore slope, the
damage level is now calculated based on an erosion area as explained in the
previous chapters. The reason for expressing them in terms of area is to make a
comparison with the van der Meer formula which is based on the erosion area as
previously explained.

46
 New damage in terms of S

100

80
Damage level S
60

40

20

0
1 1.1 1.2 1.3 1.4 1.5 1.6 1.7
Period Tp (s)
1:30 slope 1:8 slope

Figure 10: Damage level in terms of S

Figure 10 shows the damage level of the laboratory experiments expressed in

terms of S. This leads to the conclusion that if the wave heights at the toe are
equal, the trend is the same for a given bottom steepness. Equation 4.7 was used
to transform the damage level from the laboratory experiments expressed in
terms of N as can be seen in figure 9 and, equation 3.4 was used to express them
in terms of S as shown in figure 10.

In order to make a comparison with the van der Meer formula, expression 4.9
was used to generate the van der Meer curve as can be seen in figure 11. The
surf similarity parameter was based on the dataset as shown in table 4.

van der M eer in terms of S

150
Damage Level (S)

120
90
60
30
0
1.1 1.2 1.3 1.4 1.5 1.6
Wave Period Tp

van der Meer Linear (van der Meer)

Figure 11: Van der Meer formula

47
The blue line in figure 11 is a linear baseline fit to the van der Meer formula.
When comparing figure 10 with figure 11 there seems to be no correlation
between the measured results for a 1:8 slope and a 1:30 slope when compared to
the van der Meer formula. One expected the damage to be less significant. This
could be as a result of the following:

1. In the calculation of the damage area on the breakwater slope, the

porosity of the stone together with the void ratio is ignored.
2. Furthermore, it is assumed that all the stones have the same shape and
are packed one next to each other with the same packing density.

Also important to note is the definition of damage according to van der Meer.
Van der Meer describes damage as the area of the erosion in the cross-section of
the breakwater divided by the square of the stone size. The damage in the case
of the experiments were based on the actual number of stones moved, which
might not necessarily be out of the “erosion area” as defined according to van
der Meer. This means that the counting of stones method structurally
overestimates the damage level hence the van der Meer equation cannot be
applied easily. Furthermore, an important consideration is that during the testing
conducted by Hovestad (2005), the target wave height was immediately used in
order to reduce the gradual build up of wave heights. This was done to aid the
counting process and save on time, which resulted in the initial settlement of
stone that usually take place under calm wave conditions, not taking place, so a
higher damage level could be expected especially when comparing to the van
der Meer formula.

After having expressed the damage level on the slope of the breakwater in terms
of S, a new plot was generated as can be seen in figure 12 superimposing the van
der Meer equation onto the same system of axis as the other tests. This was
done in order to find the correlation of damage level between the experiments
conducted in the laboratory based on a slope of 1:8 and 1:30 as compared to
the theoretical damage level as described by van der Meer, which is based on a
zero fore-shore slope.

48
 All damage curves

150
Damage Level (S)

120

90

60

30

0
1.1 1.2 1.3 1.4 1.5 1.6 1.7
Wave Period Tp

1:30 slope-old 1:8 slope-old

Linear (van der Meer) Linear (1:8 slope-New)
Linear (1:30 slope-New)

Figure 12: Graph of all damage levels

Figure 12 shows all the test results superimposed onto the same system of axis
and the original test results based on actual measured damage has been included
to complete the picture. The blue and yellow dots are the original laboratory
experiments. Expression 4.9 was used to obtain the damage in terms of S, the
van der Meer formula denoted as the pink line in figure 12. Expressions 4.7 and
4.1 were used to transform the laboratory results based on the different foreshore
slopes denoted as the green and blue lines in figure 12. The blue, green and pink
lines in figure 12 are a baseline fitted through the points in order to see the
differences clearer.

This is not what was expected, after calculating the damage area and expressing
in terms of S, it was expected that the van der Meer formula (expression 4.8)
would lye below the measured laboratory results but in actual fact it lies above.

49
Furthermore, there is no information on the shape (angularity and elongation) of
the rock that Hovestad (2005) used in his research. He could have used very
angular rocks and that could be a possible explanation for the over-estimation in
the damage level of the standard van der Meer equation which does not correct
for angularity of the stone. Also when the van der Meer formula was
transformed using angular rocks in the area calculation, it was observed that the
damage becomes less, I thus concluded that Hovestad (2005) could have used
angular rocks in the damage area calculation conducted in the laboratory. The
non-uniform density of the stones used in the research could also be a possible
explanation for the over-estimation of the damage level conducted in the
laboratory when compared to the van der Meer equation which is based on
empirical research and curve-fitting.

50
 Chapter 6: Conclusion and Recommendation

This research has clearly shown the importance of foreshore slope on breakwater
armour unit stability. There exists a clear correlation between damage level and
foreshore slope: increasing foreshore slope resulting in greater armour unit
damage for equal off-shore conditions. When comparing wave heights on
different sea bottoms, average waves on a steep sea bottom can become higher
when compared to milder foreshores. This is as a result of shoaling before the
waves actually break. The larger wave heights will consequently result in a
greater amount of damage to the armour units of breakwaters. The longer period
waves seem to reflect more damage over both foreshore slopes.

The research has further showed that there could be other parameters affecting
armour unit stability namely:

• Wave skewness or peakedness;

• Porosity and Density of the stones;
• Void ratio;
• Angularity and elongation of the stones;
• Velocity and acceleration of the waves on a slope.

More research would have to be undertaken in order to find a clear correlation

between the van der Meer formula and results based on different foreshore
slopes. Although some consideration is given to foreshore slope by way of the
surf-similarity parameter it does not however consider it directly. Another
important factor would be the definition of damage area as defined by van der
Meer when making comparisons to empirical studies. The porosity, density and
void ratio of the stones would have to be considered in the laboratory process.
Once a clear correlation can be found between the application of the van der
Meer equation on steep foreshores based on different parameters and laboratory
experiments of armour units on steep slopes, a correction factor could then be
applied to the formula that would consider structures on different foreshore

51
 Hs
slopes. As a preliminary result, the value of in the van der Meer
∆D n 50
formula has to be reduced by a “factor” when considering structures on steep
foreshores.

In general structures that are to be constructed on steep foreshores should be

designed with more consideration given to armour unit stability. In the absence
of practical design guidelines relating specifically to steep foreshores, extensive
physical model studies should be undertaken for structures on steep foreshores.

Recommendations

• More research should be conducted on breakwaters with steep foreshores;

• Investigation into other parameters other than foreshore slope, like wave
peakness, velocity and acceleration;
• Specific definition of damage area for armour unit stability;
• As a possible recommendation when calculating the damage area for the
van der Meer equation to be expressed in terms of S, the amount N can
be normalised to the number of displaced units by the following
expression:

N s d n 50
Nod =
B
Where:

Nod number of displaced units [-]

NS number of stones that were removed from their original band [-]
B width of the measured section [m]

Theoretically, the volume of the erosion can then be derived using Ve = Ae B

and with Ae Dn502, Ve = SBdn50. The porosity has to be considered for reliable
comparisons to be made.

52
 BIBLIOGRAPHY

Battjes, J.A., “Surf-Similarity”, Proc. of the 14th conference on coastal

engineering, Copenhagen, Denmark, 1974

CUR, “Manual on the use of Rock in Hydraulic Engineering”, Cur Publication

169, 1994

Hovestad, M., “Breakwaters on steep foreshores”, MSc Thesis, Delft University

of Technology, 2005

Terrile, E., “The threshold of motion of coarse sediment particles by regular

non-breaking waves”, MSc Thesis, Delft University of Technology, 2004

TROMP, M., “Influences of Fluid Accelerations on the Threshold of Motion”,

MSc Thesis, Delft University of Technology, 1994

USACE, “Coastal Engineering Manual”, Fundamentals of Design, U.S Army

Corps of Engineers, 2005

USACE, “Shore Protection Manual”, Fourth edition, U.S. Army Engineers

Waterways Experiments Station, Coastal Engineering Research Center, 1984

Van der Meer J.W., “Rock slopes and gravel beaches under wave attack”,
Doctoral Thesis, Delft University of Technology, 1988

Van Gent, M.R.A., Smale, A., Kuiper, C., “Stability of rock slopes with shallow
foreshores”, Portland, ASCE, 2003

LIST OF WEBSITES

http://chl.erdc.usace.army.mil/cem

53