~

Hydraulics Research
Wallingford

Hydraulic Pe rf o rmance of Breakwa t e r

Cro wn Walls

A P Bradbury, N W H Alls op & R V S t e phens

Report No SR 146
March 1988

Wallingford, Oxfordshire OX! 0 8BA.

Registered Office: Hydraulics Research limited,

Telephone: 0491 35381. Telex: 848552

This repo r t describes work funded j ointly by the Department of the
Environment and the Ministry of Agriculture Fishe r i es and Food .

The Depar tment of the F.nvironment contract number was PECD 7 / 6 / 5 2 for which
the nomi n a t e d officer was Dr � P Thorogood .

The M i n istry of Agriculture Fisheries and Food contribut ion was w i t hin the
Research Commission (Marine Flooding) CSA 557 for which the nomi n a t e d
o f f i c er was Mr A J All ison .

The work was c a r r ied out in the Mar i t ime Engineering Department o f
Hydra u l i c s Research, Wa l l ingford under the management o f Dr S W Hu n t i n g to n .

The report is published on behalf o f both DoE and MAFF, but any optntons
expressed in this report are not necessa r i l y those of the funding
Depar tments.

«) Crown copyr ight 198R

Published by permission of the Controller of Her Majesty's S t a t ionery

Of f i c e .
 Hydraulic performance of breakwater crown walls

A P Bradbury, N W H Allsop & R V Stephens

Report SR 146, March 1988

Hydraulics Research, Wallingford

Abstract

A breakwater crown wall can increase the overall effectiveness of the

structure in limiting wave overtopping. In so doing, it will contribute to
a reduction in the volumes of material required and hence the cost to
achieve a given level of performance. Current design methods are unreliable
in their prediction of the effectiveness of different crown wall armour
crest configurations in reducing overtopping. Similarly very little
information is available to support the estimation of wave forces on the
front face of the crown wall.

This study has addressed two of the major aspects of the design of
breakwater crown walls: the efficiency with which such walls deal with wave
overtopping; and the forces imparted to the front face of the crown wall.

This report draws together information from previous studies, together with
results from a series of random wave model tests. The overtopping discharge
and the impact force have been quantified for a range of wave conditions and
crown wall configurations. The effects of the main wave and structure
variables have been described by dimensionless parameters. Empirical
formulae have been derived allowing the data presented to be used for design
purposes for a wide range of conditions. A series of recommendations for
good practice are made based npon the results of the review and model
tests.

The results of this study will allow the designer of many configurations of
crown wall to determine the overtopping performance, and to quantify the
factor of safety against sliding failure, with a much higher level of
certainty than hitherto.
CONTENTS

Page

1 INTRODUCTION 1

1.1 Background 1
1.2 Out l i n e o f this study 1
1.3 Ou t l i n e o f this report 2

2 PERFORMANCE OF CROWN WALLS 3

2. 1 Previous exper ience 3

2. 2 Wave overtopping 4
2.3 Wave forces on crown wa l l s 11
2. 4 Phys ical modelling of crown wal l s t abi l i ty 12
2.5 Summary o f factors i n fluencing crown wall per formance 13

3 DESIGN OF MODEL TEST PROGRAMME 14

3. 1 Aims o f the model t e s t s 14

3.2 S e l e c t i o n o f mod e l test param e t e r s 14

4 TEST PROCEDURE AND MEASUREMENTS 16

4. 1 T e s t fac i l i ty 16
4.2 Wave cal ibra tion 17
4.3 Cons t ruct ion o f model test sect ion 17
4.4 Overtopping meas uremen t s 1 il
4.5 P r e s s ure measurements 19
4.6 Force measurements 20

5 ANALYS IS OF RESULTS 22

5.1 Overtopping 22

5. 1 . 1 Emp i r ical r e l a t ionships 22

5 . 1. 2 E f fectiveness o f cres t geome try 26

5.2 Forces 29

5 . 2. 1 Analysis procedure for random wave t e s t s 29

5.2.2 Re sul t s 30
5 .2.3 Calcu l a t ion o f horizo n t al wave forces on the
crown wal l 33

6 RECOMMENDATIONS 34

6.1 D e s i g n calculat ions 34

6 . 1 .1 Over topp ing 34

6. 1 .2 Wave forces on a crown wa l l 34
6. 1 .3 Sl iding 35

6.2 Recommenda t ions for good pract i ce 35

CONTENTS ( CONT / D )
Page

7 ACKNOWLEDGEMENTS 36

8 REFERENCES 37

TABLES

1. T e s t cond i t ions
2. Te s t s e c t ion construc t ion
3. Summary o f empirical coe f f i c i e n t s for various crown wal l
conf igurat ions
4. Re sul t s of comparative wave loading

FIGURES

1. Geome try o f breakwater c ross s e c t ions

2. Overtopping of breakwa ter crown wal l s - a fter Grav e s en & Sorensen
3. Deep r andom wave f l ume
4. T e s t sec t ion 1 , 2, 3 and 1 4 ( smooth s lope s )
5. Test s e c t ions 4-7 & 1 1
6. T e s t sec t ions 8 , 1 0 , 1 2 , compar i s on of armour configura t i o n at
c re s t
7. Te s t sect ion 1 3 - recurved wal l
8. Force table
9. T e s t s e c t ion 3 R* vs -Ln 0*
10. Te s t s e c t ion 4 R* v s -Ln O*
11. Te s t sec t i on 8 R* vs -Ln 0*
12. Te s t s e c t ion 9 R* v s -Ln 0 *
13. T e s t s e c t ion 3 F ' vs Q
1 4. Te s t s e c t ion 4 F ' vs Q
15. Te s t s e c t ion 8 F ' vs Q
16. Te s t s e c t ion 9 F ' vs Q
17. Te s t s e c t ion 1 -LnF* vs -LnO*
18. T e s t s e c t ion 2 -LnF* vs -LnO*
19. Te s t s e c t ion 3 -LnF* vs -LnO*
20. Te s t s e c t ion 4 -LnF* v s -LnO*
21 . Te s t sec t ion 5 -LnF* vs -LnO*
22 . T e s t s e c t ion 6 -LnF* vs -LnO*
23 . T e s t s e c t ion 7 -LnF* vs -LnO*
24. T e s t s e c t ion 8 -LnF* vs -LnO*
25 . Te s t sec t ion 9 -LnF* vs -LnO*
26 . T e s t s ec t ion 1 0 -LnF* vs -LnO*
27 . T e s t s e c t ion 1 1 -LnF* vs -LnO*
28 . Test s e c t ion 1 2 -LnF* vs -LnQ*
29 . T e s t sec t ion 1 3 -LnF* vs -LnO*
30. Compar i son o f measured and pred i c ted d i scharges
31 . Example of force table output s ignals
32. Example of derived total force t ime s e r i e s
33. Example o f mul t iple threshold cro s s ing with raw d a t a
34. Influence o f s e c t ion geometry on impact r a t io
CONTENTS ( CONT / D )

FIGURES ( CONT / D )

35. I n fluence o f s e c t ion geome try on impact ratio

36. Influence o f s e c t ion geometry on f i l tered 5% exceedence force
37. Inf luence of s e c t ion geometry on f i l t ered 5% exceedence force
38. Influence of sect ion geometry on f i l tered maximum force
39. Influence of sec t ion geome try on f i l tered max imum force
40. Influence of T on f i l tered max imum force - constant � s
m
4 1. Influence o f T on impac t rat i o
m - constant H5
42. Infl uenc e o f H 5 on fil tered max imum force - constant Tm
43. Influence o f H5 on impac t rat io - constant Tm
44. Influenc e of T on f i l tered max imum force - constant s
m
45. Influence of T on imp a c t rat io
m - constant s
46. Compar�son be tween measured forces and those pred�cted
• •

47. Measured forces presented in the form proposed by .Tens e n

48. Compar i s on between measured forces and pre d i c ted force s us i ng
Jens e n ' s best f i t l ine
49. Force data presented by Jensen ( R e f 28)
so. S p a t i a l d i s tribut ion of maximum wave pressure on crownwa l l
- a f t er Jensen

PLATES

1. Force table showing proof ring d e t a i l s

NOTATION

A, B Empirical coef f ic ients

a, b ••

A E levation o f armour crest rel a t ive to s t a t i c water level

c
B Struc ture wid t h , in direction normal to fac e , see Fi gure 1
c
c, c c c. Empirical or shape c o e f f ic ients
l' 2' 1
c Coef f i c i ent of ref l ec t ion
r
d Typical dimension of prototype s truc ture
p
d Typ i c a l dimens ion o f model s tructure
m
D Part i c l e s i ze or typ i c a l d imension
D Nominal part i c l e d i ameter
n
E E l a s t i c modulus
E. Incident wave energy
1
F Pro j ec t ion o f crown wa l l above armour cres t , see Figure 1
c
F To t a l depth-int egrated horizontal force on crown wal l per
H
unit width
F Horizont a l force exceeded by 5� of force peaks
H5
F Maximum horizontal force
Hmax
F* Dimensionless freeboard parameter , defi ned in Equa t i o n 5 . 4
G Width of horizontal armour crest berm , see Figure 1
c
g Grav ita tional acceleration
H Wave height , from trough to cres t
H O f f shore wave hei ght , unaffec ted by sha l low water proces ses
0

H S i gnificant wave height , average of highest one- third of wave

s
heights
H Maximum wave height in a record
max
h Water depth
h Height of front face of crown wal l , over which wave forces
f
may ac t
h Water depth in front o f s tructure
s
J Geometric parameter , rear fac e , see Figure 1
Ir Iribarren or surf s im i l arity number
Ir' Mod i f ied Iribarren number
IR R a t io between wave impact period and mean wave period ,
T. /T
1mp m
K Geometric parameter , rear fac e , see Figure 1
 Damage coe f f i c ient in Hudson formula
Wave number , 2n/L, al s o armour layer packing c oe f fi c ient
Wave l ength
Deep water or o f fshore wave l enRth , gT2/2n
L Deep water wave length of peak wave period
p
L Wave l ength of peak per iod in water depth in front o f
ps
structure
M Armour unit mas s
N Number o f armour units , o n the s l ope , o r i n an area o f the
a
( test) section
Number o f armour units d i s p l a ced
Number o f armour units rocking

N Number of waves in a storm , record or test

w
n Poros ity , usua l l y taken as n
V
n Volumetric poro s ity , v o l ume o f voids expressed a s proportion
V
o f total volume

n Area porosity
a
Q Overtopping dis charge , per unit length o f sea wal l

Q* Dimens ion l e s s overtopping d i s charge , d e f ined in equation 2 . 2

Volume o f overtopping , per wav e , per unit length o f
structure
Supe r f i c ial v e l oc ity , or spe c i f i c d i s charge , d i s charge per
unit area , usua l l y through a porous matrix
R Run-up l ev e l , r e l ative to static water l eve l
R Mean run-up level
R Crown wal l freeboard , rel ative to static water l e v e l
c
R* Dimens ionle s s freeboard , d e f i ned i n F. q uation 2 . 3
Run-up l evel o f s ign i f ic ant wave
Run-up level exceeded by only 2! of run-up crests
Run-down leve l , b e l ow which only 2% pass
Roughness value , usua l ly relative to smooth s l o p e s
s. Incid ent spectral energy d en s ity
1
s Re f l e cted spectr a l energy dens ity
r
s Wave steepnes s , �/L

s Steepness o f mean period 2n H / g T 2

m s m
s Steepness of peak period, 2n H /g T 2
p s p
T Wave period

T Mean wave period

m
T Spec tral peak period , inverse o f peak frequen c y
p
T Duration of s t orm , s e a s t a t e o r tes t
R
T. Mean period between wave impac t s on crown wal l s
1mp
u, V Flow veloc i t i e s , o ft en orthogonal compone n t s o f v e l o c i t y

S t ructure front s l ope ang l e

s Angle of wave at tack

p Mass density , usua l l y o f fresh wa ter

Mass den s i t y of sea wa t e r
Mass density of rock
Mass dens i ty of conc r e t e

Rel a t ive den s i ty , (_E. - 1 )

p
pw
P r o t o t ype to model ratio o f a charac t eri s t ic s t ruc tural
d imension
1 INTRODUCTION

1.1 Backg round

Rubble mound breakwa t e r s or sea wa l l s us ual ly
incorporate a c rown or parapet wall to reduce the
seve r i ty and l im i t the e f f ec t s , o f any waves
overtopping the s t ructure . This contr ibutes t o the
overall effec t iven e s s o f a breakwa t e r in reducing the
transmission o f wave energy , and o f a s e a wall in
reducing erosion and flooding. �y increas ing the
per formance o f the s t ructure , a crown wall w i l l a l l ow
a reduc t ion in the volume o f rubble reQui red , and
hence the c os t , to achieve a given s t andard o f
pro t e c t ion. Crown wa l l s are o f ten a l s o des igned to
carry and pro t e c t pipel ines and other s e rv ices from
berths in the l e e o f the breakwa ter ; t o p rov i d e acce s s
along, and t o the outer end o f the s t ructure ; and t o
contain and direc t any over topping t o avoid damage o r
f looding o f vulnerable areas .

Two main unce r t a i n t i e s a f fe c t the d e s ign o f a c r own

wal l in the cal cul at ion o f :

a) the e f fect of the c r own wa l l on wave

overtopping ;
b) the forces appl ied to the wal l .

Present des ign methods do not a l low the d e s c r i p t ion o f

wave-induced flow over the front face and at the c r e s t
of a rubble s truc t ure w i t h acceptab l e c e r t a in t y . The
d e s i gn of crown wal l s gene r a l l y rel ie s on the resul t s
o f a few, s i te spec i f i c , hydraulic model s tud ies ;
toge ther with the l o c a l knowl edge and experience o f
the des igner . The consequence may b e over-d e s ign and
hence increased co s t , or und e r- d e s i gn wi t h the
a t t endant r isk of f a i lure . There have been many
examples of damage to breakwa ter crown wa l l s by s t o rm
a c t ion. In several in s t ances this has contributed t o
the failure o f lengths o f the s t ructure . Some notab l e
examples o f crown w a l l f a i lure have been a t Sine s ,
Diab l o Canyon , Arzew, Tripo l i and Ant a lya . Despite
the s e failures there h a s b e e n very l i t t l e r e s e arch
e f fo r t d irected to the d e s ign of crown wal l s , and
the ir response t o wave f l ows and forces has been
l i t t l e unders tood .

1.2 Out l in e o f this

s tudy
A s tudy of wave overtoppin� and wave for c e s on crown
wal l s was ins t i t u t ed as part of an ove r a l l progr amme
o f research on the des ign and performance of rub b l e
mound s tructure s . The s t udy was conduc ted i n three
s t ages .

1
 The l i terature available on hyd r aul i c per formance o f
c r own walls and des ign practice was reviewed to
identify the data and method s ava i l ab l e , and the maior
areas of uncertainty. The re sults o f th i s rev iew were
then used to set the parameters to be determined ,
d e s ign the model test procedure s , and to identify
po s s ib l e empirical methods for the ana l y s i s of test
r e s ults .

A comprehens ive series o f model tests were conducted

in the large random wave flume at Hydraul ics Research.
A base test section with a 1 :2 front s lope was
mod i fied to give 13 d i f ferent test sections . The main
structural parameters to be var ied were the c rown wall
he ight and freeboard , and the r e l ative armour crest
l ev e l . The tests were conducted at 2 d i f ferent water
levels using 10 wave cond itions . During testing ,
measurements were made of wave overtopping d i s charges ,
wave pressures and wave forces on th e front face of
the wa l l . Video recordings were made o f wave flows
over the wa l l , and an attempt was made to quant i fy
overtopping veloc ities us ing v i d eo image process ing
techn iques .

The results o f these measurements , and o f the other

stud ies revi ewed , were then ana l ysed to �ive
appropriate des ign guid anc e . A number o f empirical
formulae we re used to d e s c r ibe the data , and to a l l ow
the general i sation of the te st r e s ults for use in
des ign.

I t may be useful to the reader c oncerned with the

d e s ign and performance o f rubble mound structures to
note that the proj ec t , o f wh ich thi s study was a part ,
has a l s o addressed : -

a) the hydro-geotechn ical per formance of large

rubble mound s (Ref 1);
b) the d e s ign and performance o f concrete armour
units for coastal structures ( R e f 2);
c) the d e s ig n , performance and dur a b i l ity of rock
armour (Refs 3-5).

1.3 Outline o f th i s
report
Th i s report may be cons idered in three parts . The
review of in formation on the per formance o f c rown
wal l s in Chapter 2 draws together the results of s ite
and l aboratory experience to ident i fy the main
variab le s , and suggests pos s ib l e des ign metho d s . The
d e s ign and execution o f model stud i e s conducted in
this project are reported in Chapters 3 and 4. The
r e s ults o f the test measurements are analysed and
d is c u s s e d in Chapter 5 . Conc lusions and

2
 recommend a t ions drawn from both the review and the
model t e s ts are d e s c r ibed in Chapter 6 .

2 PERFORMANCE OF
CROWN WALLS

?. . 1 Prev ious
experience
Much may b e learn t of the performance of breakwa ter
crown wa l l s from example s o f their fai lure . The
principal modes of failure may be d iv ided into two
c a t e gor i e s : functional and / or s t ru c t ural fa i l ure .
Func t ional f a i l ure occurs when the breakwa ter , or the
e l ement con s idered , f a i l s to per form its main task . A
s truc tural failure occurs when an e l emen t is broken,
or s i g n i f ic a n t l y d i s pl a c ed , such that it no longer
serves its or ig inal purpose .

At D i ablo Canyon, Ca l i forn i a , the two breakwa t ers

protec t ing the cooling wat e r basin were armoured with
l arge Tribars and surmounted by a very s imple crown
wal l s l ab ( R e f 6 ) . During a s torm peaking on 2 7 - 2 8
January 1 9 8 1 the outer s e c t ion of the we s t breakwa t e r
was severe ly damaged , and 4 o f t h e 3 0 0 t o n capping
b l ocks s l id into the s e a . Model t e s t s conduc ted t o
s t u d y the reasons for the fa i lu r e sugge s t tha t the
loss of the crown wall s e c t ions was pre c i p i ta ted by a
local fail ure of the armour , fol l owed by a p rogr e s s ive
f a i l ure of armour up to the crown w a l l . D i r e c t wave
a t tack on the crown wa l l sec t ions caused s u f f i c i e n t
movement to release i t s weight from underl ayer s tone
benea th4 Con t i nuing wave a t t ack progre s s ively removed
the underlayer support , a l l owing the w a l l to f a i l in
8 . 7m long s e c t ions .

In D ecember 1 9 7 1 the new breakwater at Antalya harbour

Turkey was very near completion , lacking only the
p l acement o f some rock armour behind the crown w a l l
s e c t ions . A severe s torm o n 1 0 - 1 1 December lead to
the f a i lure of the outer 600m of the bre akwa t e r .
Gunbak & Ergin (Ref 7 ) d e s c r ib e de t a i l s of the
cons truct ion and damage . They d e s c r ibe c a l c u l a t ions
of wave force on the crown wal l , and conc lude tha t
s l id ing of the 2 5 0 tonne c rown w a l l s e c t ions was the
primary f a i lure mode . As at Diablo Canyon, the crown
w a l l had been c a s t d i rec t l y onto the 2-6 tonne
und e r l aye r , rather than onto the l e s s permeab l e core
m a t er i a l . In the ir c a l c u l a t i on s , Gunbak & F.rgin note
that a mean value for the coe f f ic ie n t of f r i c t ion
� = 0 . 7 be tween the c rown wall and the underlayer
a l lowed crown wa l l failure at the wave cond i t ions
e s t im a t ed for the s torm, H = 6m , T = !Os .
s m

3
 A l lsop & Steele ( R e f 8 ) report the resul t s o f tes t s on
a l ternative breakwater c r o s s -sec tions in a water depth
of h = 2 3m with a 1 : 100 year design o f fshore s torm of
H 8 . 7m . One cros s-sec tion was armoured with 1 6m 3
s
=

Tetrapods a t a front slope o f 1 : 1 . 5 t o a crest level

around 7 . 8m above s t a tic water level. The o t her used
1 6m 3 Antifer cubes at a s lope of 1 : 2 . 0 . The u p s tand
on the crown wall reached around 6 . 7m above water
level . For both cross-sections , the fir s t design for
the crown wal l , weighing around 1000 tonnes in 1 5m
lengt h s , and laid with a small key onto 3-6 tonne
underlayer , star ted to slide backwar d s under waves of
H s = S . Om . The final , s tabl e , c rown wall sec tion was
approximately 30% heavie r .

The failure o f the north wes t main breakwater t o

Tripoli harbour , Libya , a l s o involved the failure of
the c rown wall ( Refs 9, 1 0 ) . In this ins tance
s t ruc tural failure of lengths of the wa l l was
precipitated by breakage and removal o f the Tetrapod
armour , and the consequent impac t forces . �any
sec tions of the crown wall u p s t and sheared or were
bent backwards from the base pl a t e . Gunb ak ( F e f 9)
e s t imates a number of a l t ernative combinat ions of wave
conditions and s liding coef ficien t s , between
� = 0 . 5 -0 . 9, that would lead to crown w a l l movement .
A t Tripoli the c rown wall was c a s t in lOm leng ths ,
each weighing around 500 tonnes , on t o 2-4 tonne
underlayer . The ups t and of the crown wal l pro j ec ted
approxima tely 3 metres above the c r e s t of the armour
layer .

At Arzew and Sines the main c au s e o f failure appears

t o have been the rela tive fragility o f 48 tonne
Tetrapods and 42 tonne Dolo s s e res pec tively ( Refs 2 ,
1 1 , 1 2 ) . As a consequence of the removal o f the
primary armour the c rown wall was severely damaged on
both s t ru c t ures . At Sines the ups t and of the crown
wal l projec ted 3 metres above the original crest o f
the Do l o s s e armour . The crown wal l was c a s t in 1 5m
long sec tions weighing around 4000-5000 tonnes
( R e f 1 3, 14 ) .

The c rown wall a t Akranes , Icel and , survived a severe

s t orm which removed much of the rock armour in front
( Re f 1 5 ) . In this ins t ance the c rown wal l was
construc ted as part of a cais son used to form the rear
part of the breakwa ter.

2.2 Wave overtopping

The main purpose of a breakwa ter c rown w a l l is to
al low a savin� in materia l s by reducing the wave
over topping. I t wil l be noted t h a t the s tochas tic

4
nature of s torm waves imp l ie s that a crown wa l l would
have to be uneconom i c a l l y l a rge to prevent a l l
over toppin g . I t i s there fore we l l accepted that some
probab i l ity of overtopping should be a l l owed for in
d e s i gn , perhaps by l im i t ing the mean expected
d i s charge for a g iven re turn per iod event to an
acceptable leve l . The l ev e l o f ove rtopping perm i t ted
w i l l vary wide l y , depending upon the c r e s t and rear
s l ope prot e c t ion ; the frequency of use of berths in
the lee of the breakwa t e r ; and the construct ion and
use of any other structures c l o s e l y behind the
br eakwater . The d e s ign overtopping should be much
l e s s i f the area behind the bre akwa ter is to be
r e c l a imed , than i f open water is to be maintaineO.

R e l a t iv e l y l i t t l e guidance is ava i l ab le to the

d e s igner on the level o f d i scharge that may be
perm i t ted . It w i l l be i n f l uenced by the importance of
three d i f ferent aspec t s :

a) I nconvenience or danger to persons or veh i c les

us ing the crown wal l ;

b) Damage to e l ements of the c rown w a l l s t ructure ,

or leeward prote c t i on ;

c) Wave d i s turbance i n the l e e of the b reakwa ter .

The l im i t ing d i s charges for use of the area behind an

overtopping sea wa l l have been s tudied by Japanese
r e s e archers Fukud a , Uno & I r ie ( Re f 3 6 ) . Their work
h a s heen d i scussed by Owen ( Re f 1 6 ) and Jensen & Juhl
( Re f 34 ) . Owen has summarised their recommendat ions :

1) For a person to walk immed i a t e l y beh ind the

s e awa l l with a l i t t l e d i s comfor t ,

2) For a per son to walk immed i a t e l y beh ind the

s e awa l l with l i t t l e danger

3) For an automob i l e t o p a s s imme d i a t e l y beh ind the

s e awa l l at high speed

4) For an automob ile t o p a s s imme d i a t e l y beh ind the

s e awa l l at low speed

5
5) For a house located immed i a t ely behind the
seawall to s u f fer no d amage ,

6) For a house loca ted immed ia tely behind the

s e awa l l to suf fer no sub s t an t i a l flooding or
damage, a l though experiencing p a r t i a l damage t o
windows and g lazed doors ,

It may be noted that these sugg e s ted l imi t s are, as

yet , not supported by tes t s or f i e l d measuremen t s
e l sewhere . It i s recommended that the reader should
con s u l t the or ig inal re ference before using the values
in des ign .

Aga in , rel a t ively l i t t l e informat ion i s ava i l able on

the effect of overtopping d i s charge on cre s t and rear
s lope protec t ion . In genera l , proposed solut ions
shou ld be mod e l tes ted to give acceptable cer t a i n ty .
Goda ( Ref 37) has sug�e s t ed l im i t s for certain l ev e l s
o f protection to cre s t o r rear s l op e s o f sea wal l s : -

S t ructure Max value

of 5 : m 3 / s . m

Paved ( conc r e t e ) c r e s t 0.2

unpaved ( grassed) crest 0 . 05
Crest and rear s l ope paved 0 . 05
Cres t , paved and rear s lope unpro t e c t ed 0.02
Cr e s t and rear s l ope gra s s e d only 0.005

I t may be noted that the l a s t value compares wel l w i t h

the his torical Du tch l imi t for grassed sea wa l l s o f
0 . 002m 3 fs . m .

No informa t ion i s ava i l ab l e on the l im i t i ng

over topping con d i t ions that may be tolera ted by
v e s s e l s against or c l ose to the leeward face o f the
breakwa ter . The as s e s sment of t hi s , and of the degree
of wave d i s turbance caused by overtopping that may be
tolera t ed , are gener a l l y treated as s i te spec i f i c
problems us ing p hy s i c a l model t e s t s .

The c a l cu l a t ion of overtopping d i s charge under random

wave s has been ad�re s s ed by r e l a t ive l y few
researchers , and the general appl i c a t ion o f those
resu l t s ava i l ab l e to breakwa t e r crown wal l s i s
somewhat uncer ta i n .

6
Owen (Ref 16-lR) has developed an empirical method for
the calculation of overtopping discharges for simple
seawalls, based on a series of hydraulic model tests
under random waves. The test used plain and bermed
sea wall sections with smooth faces and no crown wall.
Owen derived an empirical equation relating a
dimensionless discharge, 0*, to a dimensionless
freeboard, ll*:

0* = A exp ( -R R* /r ) ( 2. 1)

where

0* 5/T g (2.2)
m Hs
=

RC s -'-
m t =
R* = ( ) R c/T m (g Hs)'
r.s ( 2 . 3)

Owen presents values of the empirical coefficients A &

F for a range of slope angles and berm configurations.
This method was not developed for walls with complex
crest details. However, two modifications have been
considered. They involve the definition of an
efficiency factor to describe the effect of the crown
wall element in relation to the simple slope. In
defining an efficiency factor for a given crown wall
detail, a hypothetical discharge may be useful. This
is defined as the discharge that would occur for the
same wave conditions over a simple slope to the crest
level considered. Steele & Owen (Ref 19) have defined
an efficiency factor E :
f

(2.4)

where 01 is the predicted discharge at the crest of

the armour, without the crown wall, and o, is that
with the crown wall. The efficiency factor will
depend upon the crown wall geometry, principally Fe,
G and A (see Fig 1), as well as the incident wave
c c
conditions, H and T . In use, values of the
. s m . .
d1scharge needed, 02, m1ght be calculated us1ng a
modified version of Owen's expression:

( 2 . 5)

where

(2.6)

7
It should be noted that the freeboard used to
calculate R � is that of the crest of the armoured
slope, not of the crown wall.
F. will therefore
f
depend strongly on the projection of the crown wall
R A .
c - c

An alternative ef ficiency factor, W , may be defined

f
in terms of a discharge, 0* , predicted at the crest
1
of the equivalent simple slope continued up to the
level of the crown wall crest:

( 2. 7 )

Again, if using Owen's general form of expression:

A W exp (-B R*/r) (2.R)

f

where R* is defined as before in equation 2.3. It may

be noted that W may in turn depend upon R*.
f

Ahrens & Heimbaugh (Ref 20) discuss a series of random

wave tests for a sea wall in relatively shallow
water. The sea wall incorporates a rip -rap armoured
revetment slope, and a crown wall with various
geometries. They derive an expression that appears
similar to Owen's:

0 = 00 exp(C F') (2.9)

1

where 00 is a coef ficient having the dimensions of

discharge rate per metre run; c1 is a dimensionless
coef ficient; and the dimensionless freeboard
parameter, F', is defined in terms of the local wave
height, H • and wave length, L :
si ps

F' = R
C
I (H
st
. 2 L
ps
) 1I3 (2.10)

In their work, Ahrens & Heimbfugh, define H in terms

= 4 m 0' .
si
of spectral energy, H In deep water
si
L = g T 2 12 11 , .but in shallower water the following
p p
approx1mat1on may be used:
• •

(2.11)

Generally the remaining information on breakwater

overtopping either relates to structures without crown
walls, or is based on regular wave testing only.
Allsop (Ref 21) measured wave transmission over
rock-armoured low crest breakwaters without crown
walls. A good description of the coefficient of wave

8
transm i s s i on was given by the d imen s i onless freeboard ,
R*. Jensen & Sorensen ( Ref 2 2 ) presen t a set o f
equa t i o ns , based on s i te spe c i f i c mode l t e s t s al lowing
the c a l c u l a t ion of the i n t en s i t y of overtopping water
as a func t ion of d i s tance behind the breakwa te r :

( 2 . 12 )

and

q ( x) - q 1 0 -x/b (2 . 1 3 )
o

Where : 0 is the t o t al overtopping d i s charge

(m 3 /s per m)
b is a cons tant equal to the d i s tance
behind the b r e akwater, in me tres, for
wh i c h the ove r t opping decreases by a
factor o f 1 0
q ( x ) i s the overtopping inten s i ty a t a
d i s tance x a l ong a normal to the rear o f
the breakwa ter (m 2 fs per m)
is the over topping in ten s i ty immed i a t e l y
behind the breakwater ( i e a t x = 0 )
(m 2 fs per m )

These equat ions g ive an ind ic a t ion o f the decrease i n

overtopping inten s i ty w i t h d is t ance behind the
s tructur e . It i s l ikely however that these s p a t i a l
varia tions w i l l b e s ig n i f i c a n t l y l e s s important than
the variat ions of d i s charge wave by wave , where peak
d i s charges may be ord ers of magnitude greater than the
mean value . Un fortuna t e l y l i t t l e data is ava i l able on
the tempo ral var i a t ions o f overtopping di scharges
under random waves . Jensen & Juhl ( R e f 34) report
resul t s o f the measurement o f d i s charge over the 5
10 waves g iving the h ighes t overtopp ing . They present
r e sul t s of their measuremen t s graph ic ally, f i t t ing a
l ine o f general equa t ion :

(2.14)

Where : Q i s the average d i scharge over N wave s

q w i s the d i scharge for the s ingle l arges t
over topping wave in N waves
p is the probab i l i t y of occurence, 1/ N
A and E are emp i r i c a l coe f f i c ie n t s

Example values t h a t may be deduced from t h e graph a s

shown below :

9
 N p qw /6

25 0 . 040 1.5
100 0 . 01 0 33
200 0 . 00 5 87
500 0 . 00 2 243

The e f fect of the shape o f the front face of the crown

wall has been ad dressed by Vera-Cruz (Re f 23). Using
regular waves o nl y , Vera-Cruz d e f ined an e f fec t iveness
parameter for a curved wa l l in terms of the r a t i o of
wave heights at the onset of overtopping for the
curved wa l l r e l a t ive to a s imple vertical wal l .
Values o f t h i s e f fec t ivene ss parameter o f around
80-85% were determined , sug�est ing that under random
waves any s m a l l change in wa l l shape w i l l have
r e l a t ively l i t t l e influen c e .

Generally l i t t l e d a ta is ava i l able t o de s c r ibe the

e f fects o f d i f ferent crown wa l l con figur a t ions on the
overtopping performanc e . Some experience from the
measurement of wave run-up leve l s , and of ove r topping
o f s imple sea wal l s , may s t i l l be helpful .

In pred i c t ing overtoppin� d i s charges o f s imple s e a

wal l s using Owen ' s method , a simple r e l a t ive run-up o r
roughness f a c to r , r , is u s e d t o d e s c r ibe the
influence o f roughness and permeab i l i ty o f the front
fac e . Value s o f r were a s s umed from the r e s u l t s o f
previous inv e s t i g a t ions o f run-up under regular waves .
It was impl i c i t l y as sumed that values o f r were
constan t for a g iven struc ture s . More recen t l y A l l s o p
e t a l ( Re fs 24, 2 5 ) have examined wave run-up on smooth
and rubb l e s lopes under random wave s . From the s e
t e s t s i t may be conc luded that the value o f the
roughnes s c o e f f i c i e n t r , varies with the Ir ibarren
numb e r , I r .

The e f fect o f the angle o f inc idenc e , S, o n run-up and

overtopping has been stud ied by Owen ( R e fs 16-lR) and
Tautenhaim e t a l ( Re f 2 6 ) . In both instances
inc reases i n the res ponse measured , over topping

for ang l e s o f i n c idence around B = 10-2 0°, over those

d i scharge and run-up l ev e l s respect ive l y , were noted

fore = 0°. Wh i l s t n o t i c e ab l e , these increases were

no t s u f f i c i e n t l y severe to outwe igh many o f the o ther
uncer taint i e s . A more complete review o f the e f fe c t
o f wave ob l iq u i t y has been given previous l y by Al l s o p
( Re f 27 ) .

In the des ign o f a c rown wall the v e l o c i t y and path o f

waves over topping the struc ture w i l l b e o f conc e r n .

10
 Often the parape t wal l wi l l be pos i t ioned s o a s to
throw overtopping water c lear o f the rear face armour .
An exampl e o f such a d e s ign i s shown by Jensen
(Ref 2 8 ) , c i t ing e a r l ier work by Gravesen & Sorensen
( R e f 29 ) , and is i l lu s t r a ted by Figure 2 .

2.3 Wave for c e s on

c rown wa l l s
Wave forces act ing on a crown wa l l sec t ion wi l l
princ i p a l ly act on the front fac e , and on the
underneath caus ing upl i f t . In both inst anc e s
hydraul ic model t e s t r e s u l t s are l iable t o s c a l e
e f fec t s . Wave impac t pressure again s t a wal l may
reach very l arge values for very short dur a t ions .
These short dura t i o n impa c t pressures are unl ike l y t o
exc i te any s igni f i cant r e s ponse in crown w a l l s e c t ions
weighing hundred s , or thousand s , of tonnes . Thi s is
for tunate as it is the brief impact pres sure that i s
most a f fected by s c a l e e f fe c t s in the en t r a inment o f
a i r . Momentum and quas i-hyd r o s t a t i c forces gener a l l y
s c a l e corre c t l y in we l l d e s igned hydrau l i c mod e l s , s o
the wave forces o n the front face c au s i ng s l id ing o r
overturning wi l l be exp e c t ed to b e reproduced by mod e l
te s t s . Upl i ft pre s s ures o n the und e r s ide o f the c rown
wa l l are l e s s e a s y to reproduce corre c t l y due to the
uncertainty in the s c a l ing o f v i scous f l ow e f fec t s ,
particularly unde r cond i t ions o f air en tra inmen t .

The s c a l ing o f s teady s ta t e f l ows to correct for any

v iscous e f f e c t s has been d iscussed prev ious l y by
A l l sop & Wood ( Re f 1 ) and by Jensen ( Re f 2 8 ) . In such
c i r cums tance s flow vel o c i t i e s can be correc ted by
us ing a d i s tor ted scale for the mode l l ing of the
·
porous laye r s . very l i ttle data is av a i l a b l e to cover
cond i t ions o f revers ing f low with h igh l ev e l s o f
entrained a i r . In des i gn work the pressure
d i s tr ibut ion i s gener a l l y a s s umed to be r e c t angu l a r ,
trapezoidal , or tr i angu l a r , w i th the max imum pre s sure
on the und e r s ide equal to that a c t ing a t the b o t tom o f
the front fac e .

In one o f the more comprehens ive p i e c e s o f ana l y s i s o f

forces on crown wal l s , Jensen presents resul t s o f wave
force measureme n t s , and d i s c u s s e s the main d e s ign
as sumpt ions (�e fs 28,30 ) . The max imum hor i zo n t a l
force i n 1000 wave s , p e r m e t r e r un , FH , i s made
d imensionle s s by d iv i d ing by p g h f L , where h is the
f
he ight of the front face o f the c r owK wa l l . Thi s
d imen s i o n l e s s force i s then p l o t ted agains t a r e l a t iv e
wave he ight , H s /A , and a s tr a ight l ine drawn through
the res u l t s for e �ch s tr u c t ur e , imp l y ing

11
 ( 2 . 1 5)

where the d imens ionle s s coe ffic ients a and b are

spec i f i c to a par t ic u l ar crown wall con f i gur a t i o n .
Jensen argues that the influences of water l ev e l anrl
wave period are given by A and L respec t ive l y . The
c
e f fect of wave obliquity was exam¥ned by t e s t s a t
a= o• , 22• and 4 5 • . Over t h i s range the force
decreased w i th ang l e . The decrease was most marked at
the shorter wave periods , be ing equivalent to a
reduc t io n fac t o r , k , o f around 0 . 3 3 at a= 4 5• . For
8
the longer wave per1od the reduc t ion fac to r was near er
0 . 5 a t a = 45• .

Jensen notes that wave forces are o f ten c a l c u la ted by

determining the cond i t ions for s l id ing , and s ug g e s t s
that for concrete on quarry s tone , a c o e f f i c ient o f
fr i c t io n � = 0 . 50-0 . 55 i s appropr iate . Where a
downward key , or heel is incorporated into the crown
wa l l , values for � up to 1 . 0 might be u s e d .

Other ana l yses o f wave p r e s s ures and forces on c rown

wal ls have been reported bv Gunbak & Gokce ( R e f 3 1 ) ,
and Gunbak & Ergin ( R e f 7 ) . A tr iangu l ar pres sure
d i s t r ibut ion on the front face is po s t u l a ted by Gunbak
& Gokce , but this relates primar ily to the b r i e f
dur a t ion impact pre ssur e s . The wor s t case f o r s l id ing
or over turning o f the wall w i l l occur when the wave
has reache d , or is near , i t s full run-up h e i gh t .
Gunbak & Ergin u s e a very s i mpl i s t ic c a l c u l a t ion of
run-up to e s t imate a maximum deoth of wat e r over the
armour . A t o t al force is c a lculated by summing an
impact forc e derived from the local wave c e l e r i ty , and
the hydro- s t a t ic for c e . This approach appears t o b e
considerably l e s s certain than one based o n t e s t
resu l ts , a s d e s c r ibed by Jensen ( Re fs 2 8 , 3 0 ) .

2.4 Phy s i c a l mod e l l ing

o f crown wall
s t ab il i t y
A rigorous mathema t ical model of the wave proc e s s e s
involved i s not ye t a t t a inab l e . Thus i n any
inve s t i g a t ion of the se phenomena for des ign purpo s e s ,
i t i s neces sary to rely on physical mode l l in g . The
scal ing laws for wave for c e s act ing d i r e c t l y on
vertical wall breakwa ters have been d i s cu s s ed by
Lundgren ( R e f 32) . However the s i t ua t ion c on s i d e r ed
in this s tudy i s a l i t t l e d i fferent in that the wave
may break onto a rubble mound s l ope and then
subsequently run up the s lope to impact w i th the crown
wal l . This proc e s s may produce a shock load ing ,
comparable w i th the ven t i l a t ed shock pro f i l e d e s c r ibed

12
 by Lundgren , whereby a l l , or nearly al l , o f the air
b e t ween the run-up front and the parapet i s able to
e s c a pe upwa rds . For this ven t i l ated shock , Lundgren
conc luded that b o th the pressures and the r e s u l t ing
impul s e forces could be scaled us ing Froude ' s l a w .
However , it should be noted that the concentrat ion o f
entrained air within the run-up front wi l l be much
greater in pro t o t ype than in the mod e l ; thus the mode l
w i l l have a relatively higher fluid den s i ty . This may
result in l ocal pressures derived from the mod el be ing
a l i t t l e conservat ive , although the impu l s e forces
w i l l be sub s t an t i a l l y una f f e c t ed .

2. 5 Summary o f fac tors

i n f l uencing
crown wal l
per formance
The primary var i ab l e s a f fe c t ing the over topping
performance of a breakwater crown wall are :

a) inc ident wave cond i t ions , g iven by H , Tm and/or

5
Tp ;
b) t o t a l s truc ture freeboard , R ;
c)
c
armour s l ope , a, armour un�t s1ze , D0, and layer
0 0

thickne s s , t ;
8
d) geometry of crown wal l / armour c r e s t det a i l , given
by A c , F e , and G c .

A number o f empirical expr e s s ions have been sugge s t e d

to r e l a t e the mean overtopping d i s charge , Q, t o some
of the primary var iabl e s . Those b a s ed on random wav e s
b y Owen and by Ahrens & Heimbaugh u s e exponen t ia l
expr e s s ions and d imen s ionless freeboard parame ters R*
or F ' . It may be noted however that valu e s o f the
empirical c o e f f i c ients for these expr e s s ion have o n l y
b e e n derived for a v e r y l im i ted set o f con f i gurat ions ,
none d ir e c t l y appropriate to breakwa ter c rown wal l s .

Previous work sugge s t s that run-up levels and

overtopping d i s c harges on s imple s l o p e s may increase
at angl e s o f wave a t t ack a= 10-20°. The r e l a tive
increase is not great , and no data i s ava i l ab l e to
pred i c t the e f fect of a crown wa l l . Test r e s u l t s
reported by Jensen sug g e s t s that the wave for c e s
decrease s ig n i f i c an t l y a t angles of a � 20°.

Wave forces on a c rown wall s e c t ion depend primar i l y

o n the v e l o c i t y and volume o f the wave arriving a t the
wal l . Very high impact pressures o f very short
dura t ion may be measured in hydraul i c mod e l s , but such
measurements w i l l usually s u f fer from s c a l e e f fe c t s .
Short dur a t ion impact pre s sures are unl ike l y t o have
any s tructural s ignif i c ance for the d e s ign of large
concrete crown wal l sect ions , o f t en we ighing many

13
 hund reds o f tonnes . Those forces that may cause
s l iding failur e o f the wal l must per s i s t for long
enough to overcome the over a l l s l id ing res is tance.
These forces are gener a l l y c o r r ec t l y s c a l ed in a
wel l designed hyd raul ic model . The review has
iden t i fied a s imple emp i r i c a l relat ionsh ip from
previous measuremen ts wh ich may al low the des c r i p t i o n
of the horizontal wave forc e , F H , in t erms o f
parameters des c r ib ing the inc ident wave cond i t ions and
crown wa l l / armour configurat ion .

3 DESIGN OF MODEL
TEST PROGRAMM E

3.1 Aims o f the

model tes t s
The l i terature review iden t i fied a number of areas o f
hydrodynamic des i g n o f crown wal l s presen t l y sub j ec t
to uncertainty which can be reduced by the present
research . The main areas of per formance selected for
detai led s tudy in these tests wer e :

a) Over topping d i scharges ;

b) Wave forces .

It was intended that the s t udy should permit

development of methods of pred i c t i on of over toppin�
dis charges and forces on crown walls , using emp i r i c a l
frameworks derived from mod el tes t i n g . In both c ases ,
the emphas is on the model tests would be on geome t r i c
variation o f the cres t deta i l , both to armouring and
to the shape and height of the crown w a l l .

3.2 Selec t i o n o f model

test parameters
The hyd raul ic effic iency of the breakwater is o ften
assessed on its per formance under a g iven des ign
storm, defined in terms of wave and water level
cond i t ions . These two parameters , a l ong with wave
d i r ec t i o n , s truc t ur e geometry , cons t ruc t i on type, and
foreshore gradient, are the most impor tant factors
a f fec ting the level o f overtopping and the wave for c es
a c t ing on a s truc ture.

A s tudy o f the hydraulic performance and s t ab i l i ty o f

breakwater crown walls should therefore take account
of as wide a range of these parameters as pos s ib l e . A
wide range of wave cond i t ions were selec t ed for use in
this s t udy, thus a llowing a good des c r i p t ion to b e
made o f the hydraul ic performance o f the s truc t ur e .
The following e f fects have a l l been considered :

a) Constant wave period and varying s i g n i f icant wave

height;

14
b) Constant signific ant wave height and varying wave
period ;
c) Constant wave steepn e s s - various values o f �5 ,
and T p for a sea steepness o f 0 . 04 .

Since a breakwater with a crown wall s up e r s tructure is

of rel atively complex geome try , it might reasonab l y be
expected that the overtopping discharge would be given
by a more complex function than that suggested by Owen
for s imple s loping seawa l l s ( Re f 1 6 ) . The e f fects o f
water level relative to the crest o f the seaward s l ope
of the breakwater and the c r e s t of the breakwater
super structure were there fore examined .

The e f fects o f foreshore g radient have not b e en

investigated in this study and a sing l e foreshore
gradient of 1 : 5 2 was used throughout this stud y .
E f fects o f any uncertainties in the wave
trans formations near the s tructure , c aused by wave
breaking , shoaling and refractio n , were minimised by
measurement of wave conditions at the site o f the
structure in the calibration stage o f testin g .

Whilst it i s acknowledged that the e f fects o f oblique

wave attack on wave fo rces and overtopping o f the
structure may be important , the e f fect o f this
parameter was not investigated in this study .

The structure geometry and construction type o f

breakwaters and their crown wal ls varies quite
considerably . Factors such as s e award s lope o f the
breakwater ; porosity, �ermeabil ity , and roughne s s o f
the front s l ope armouring ; po sitioning o f the
armouring rel ative to the crown wal l ; e l ev a tion and
geometry of the wall ; a l l have significant e f fects o n
the hydraulic per formance . The test sections were
d e s igned to incorporate as many o f the s e parameters a s
practicable . The ef fects o f sl ope roughn e s s were
examined by comparing rock armoured structures with
smooth s l ope s . In all case s the core was imperme ab l e ,
representing a worst case for wave run-up . A single
seaward slope o f 1 : 2 was used throughout the stud y .
Whil st the s l ope will a f fe c t the form o f the wave
striking the crown wal l , it was felt that a 1 : 2 s l ope
was reasonably typical , and a l s o gene r a l l y represents
the most severe case for wave run-up ( � e fs 24 , 2 5) .

Geometric changes to the structure were c o ncentrated

around the crest area , a t the inter f a c e between the
crown wall and the armouring , and at the crest of the
crown wal l . The fol lowing geometric parameters were
varied :

15
 a) Freeboard;
b) Height of crown wall;
c) Level of armouring and berm width of armouring
relative to crown wall;
d) Profile of crown wall.

A detailed description of the wave conditions, water

levels, and test sections used in the model tests is
given in Chapter 4.

It will be noted that the test programme was not

designed to measure uplift pressures on the underside
of the crown wall. The flow of air and water in the
region below a crown wall will depend critically upon
the detailing of the local geometry and on the
effective permeability of the rock layers in this
region of the structure. Problems in the scaling of
these flows have been discussed previously, by Allsop
& Wood ( Ref 1) and elsewhere. It was clear that it
would not be possible to reduce present uncertainties
in the calculation of uplift pressures until results
of another research project on the hydro-geotechnical
behaviour of rubble mounds were available. Uplift
forces were not therefore addressed further in this
study.

4 TEST PROCEDURES
AND MEASUREMENTS

4. 1 Test facility
The model tests were conducted in the deep random wave
flume at Hydraulics Research, Wallingford. This
flume, shown in Figure 3 is 52m long, and is divided
for much of its length into a central test channel,
ending in a finger flume, and two side ahsorption
channels. Splitter walls of graduated porosity are
designed to minimise the level of re-reflected waves.
The flume has a range of working water depths between
1.3-1.7m. Two water depths at the paddle, of l.Sm and
1. 4 m, were used for this project. The wave paddle is
a buoyant sliding wedge, driven by a douhle acting
hydraulic ram. The random wave control signal is
supplied by a BBC micro computer using software
written at Hydraulic Research.

An approach beach, at a slope of 1:52, was moulded in

cement mortar, in the central channel of the flume.
The slope extended offshore from the test section
around 18 metres into deep water, where it was
truncated by a smooth curved transition slope into a
1: 10 slope to the floor of the flume.

16
4.2 Wave cal ibrat ions
Before t e s t ing o f the var ious crown wall sec t ions
commenc ed , a s e r i e s of wave calibra t ions were carried
out wi th the moulded seabed in place , but with no test
s e c t ion . A wave ab sorb i ng beach was in s t a l l e d
landward o f the s i t e o f the t e s t s e c t ion to l im i t wave
r e f l e c t ions from the end wall o f the f l ume . Wave
cond it ions were measured in deep wa ter o f fshore and at
the s i te o f the s tructure .

Waves were recorded us ing twin wire r e s i s tance type

wave probe s . The analogue s ignal was d i g i t ized and
analysed on l ine us ing a PDPll / 7 3 m i n i c omput er . Wave
data was ana l ysed us ing a spectral anal ys i s prog ram
and the s igni f i c an t waye he ight d e f ined u s i ng the
approxima t ion H s 4m ' . Resul t s o f the wave
0
=

c a l ibrations are given in Table 1 . JONSWA P wave

s pec tra were u s ed throughout the s tudy .

4.3 Construc t ion of

Model Te s t
Sec t i ons
The test s e c t ion was con s t ruc ted on a f l a t floor in
the gl azed sec t ion o f the finger flume , w i th the toe
o f the s truc ture approxima tely 46m from the wave
pad d l e . An impermeab l e core s e c t i on was construc ted
in p l ywood , with a seaward s l ope of 1:2.

A total o f 1 3 t e s t s e c t ions , with d i f ferent c re s t

geome tr i e s , were constructed for the ove r t o p p ing
tes t s . The t e s t s e c t ions are d e s c r ibed b e l ow and are
shown in Figures 4-7 .

A l l t e s t s e c t ions were cons tructed u s i ng the same

impermeab l e core sect ion . T e s t sec t ions 1 - 3 were not
armoured , wh i l s t s e c t ions 4 - 1 3 were a l l armoured wi th
ro ck . The rock armour ing was des igned to be
s t a t i c a l l y s t able under the mos t severe t e s t
cond i t ions and was not rebui l t during te s t in g .

The l evel o f the s l ope break a t the top o f the smooth

1 : 2 s lope was fixed for a l l t e s t s except for te s t 3 .
For t h i s t e s t the smooth p l a in s lope was e x tended
upwards to the same level as the ver t i c a l crown wal l s
on s e c t ions 1 & 2 , thus a l l owing d ir e c t l y comparative
measurements of s tructures with and without crown
wa l l s . The e f fe c t o f a narrow berm a t the base o f the
crown wal l was t e s ted in s e c t ions 1 and 2. The s e
s e c t ions had the s ame cre s t e l ev a t ion and a smooth
s lope , with and without a berm a t the toe of the wal l .
Sec t ions 4 - 1 3 were rock armoured . Sec t ions 4-7 had
a narrow berm at the toe o f the ver t i c a l face of the
c rown wal l . The h e ight o f the v er t i c a l c rown wall was

17
 varied in these test sec t i ons, t o examine the e f f e c t
o f freeboard , Fc , o f a vert ical wa l l on over topp ing .
.
T e s t sec t1ons 6 , 8 , 9 , 1 0 and 1 2 were a l l con s t ructed
with the vertical faced crown wall at the same l eve l .
The e f f e c t o f varying the level o f rock armour
r e l a t ive to the cre s t o f the crown wa l l , A , was
inve s t igated in these t e s t s . The e f fec t o r a wider
armour berm was tested in s e c t i o n 1 1 , where the crown
wall level was comparable with t e s t sec t ion 6 . A
recurve wal l was used in s e c t i o n 1 3 . Deta i l s o f the
geometry of all t e s t s e c t ions are g iven in Table 2 .

F o l l owing t e s t s to measure overtopp ing d i s c harge s ,

t e s t sec tions 1 , 4 , 8 , 10 and 1 2 were mod i f ied by
removing the c rown wa l l sec t ion . This was replaced by
a force tab l e , to al l ow force measurements to be
record e d , us ing the s ame rock armour c o n f i gurat ions .
The se t e s t s are descr ibed in d e t a i l in s e c t i o n 4 . 6 .

4.4 Overtopping
meas uremen t s
A cal ibrated volume tric tank was in s t a l l ed landwards
of the model test s e c t ion . The tank was f i t ted with a
hinged chute which could be l i fted and lowered onto
the rear of the c rown wall s e c t ion thus a l l owing wa ter
d i s charged over the crown wa l l to be c o l l e c ted and
measured when required. The tank was o f rec tangular
p l an shape . It was subd ivided into s e c t ions of
d i f ferent p l an are a , to a l l ow accurate measurement o f
low d i s charge . Ra f f l e s were incorporated to reduce
wave movement in the tanks . Water l ev e l s in the tank
were recorded using a magne t o s t r i c t ive float
transduc er , which produced a vol tage s ignal
proportional to the leve l , and hence volume of wa t e r ,
in the tank. The min imum v o l ume that could be
measured accur a t e l y was 0 . 1 5 l i t r e s . Wa ter was
d irec ted into a tank of sma l l p l an area , which
over s p i l led suc c e s s iv e l y into each of the three
larger tanks when fu l l .

A s t andard procedure for measurement o f overtopping

d i s c harges , was used throughout this s t ud y . This
procedure 1s outl ined b e l ow .

No . o f Ope r a t i on
wav e s ( Tm )

a) 0 Remove overtopp ing chute from rear

of crown wal l and set wave
generator running for r e qu i red wave
cond i t ions .
b) 300 Lower wa ter c o l l e c t ion chute onto
rear of crown wa l l , s t art to

18
 co l l e c t over topping d i s charge
( s ample 1 ) •

c) 400 Remove ove r topping chute from rear

of c rown wa l l , rec ord l ev e l in
cal ibrated tanks .
d) 600 Sample 2 a s b .
e) 700 As c .
f) 900 Sample 3 as b .
g) 1 00(1 As c .
h) 1 200 Sample 4 a s b .
i) 1 300 As c .
j) 1 5 00 Sampl e 5 as b .
k) 1 600 End o f t e s t .

This procedure a l l ows a s ta t i s t ic a l d e s c r i p t ion o f the

var i a t ion of d i s charge , over 5 batches of 100 wave s .
The overtopp ing t e s t s were run with the wave genera t o r
s e t t o produce very long sequence o f wave s without
repe t i t ion of the sequenc e , thereby reproduc ing
corre c t l y the Rayl e igh probab i l i t y of wav e he ights
found in natural sea wave s .

An extens ive s e r i e s o f t e s t s for var ious c omb inat ions

of wave height , per iod and water level were run on
each of the 1 3 t e s t s e c t ions d e s cr ibed in sec t ion 4 . 3 .
A t o t a l o f 247 overtopping te s t s were carried out .
The f u l l range o f wave and wa ter l evel cond i t ions
tes ted are d e s c r ibed in Table 1 . The ful l range o f
s tructur e parame ters are g iven i n Table 2 .

4.5 Pressure
measurements
It was intended ini t i a l ly that the hor i zontal
component of force act ing on the breakwater should be
calcula ted from the output o f a s e r i e s of pressure
transducers mounted into the s e award face o f the c rown
wal l . Ey us ing transduc e r s with an appr o p r i a t e
frequency respon s e , i t would be p o s s ible to c al c u l a t e
both h i g h frequency wave impac t pre s sures and al s o
quas i-hydro s ta t i c ine r t i a l wave p r e s s ur e s . The
pressure dis t r ib u t i on over the ver t i c a l face o f the
crown w a l l could be d e s c r ib e d , and the to tal force and
moment c a l c u l a ted from the transducer analogue time
s e r i e s output . A set of six pre s s ure tr ansducer s ,
were purcha s ed . The transducers had a ceram i c
d i aphragm o f 40mm d i ame t e r . It was noted that they
might there fore be sub j e c t to erroneous output due to
par t ial immersion o f the r e l a t iv e l y l arge pres sure
sensor head . The pressure transduc e r s were cal ibrated
by measur ing the output v o l tages for a range o f
constant depths o f imme r s i o n .

A t e s t progr amme was commenced w i th the i n t en t io n o f

measuring pre s s ur e d i s t r ibutions on the crown wal l ,

19
 for a wide range of wave cond it ions and cre s t
geome t r ie s . In i t i a l l y , very short t e s t s were run , to
iden t i fy the most appropr iate d i g i t i z at ion rate for
the output from the pre s s ure transduc ers . As t h i s
study was prima r i l y designed to measure ine r t i a l wave
forc e s , which might resul t in backward s l id ing o f the
c rown wal l , as opposed to impact fo rces causing
de forma t ion o f the mater i al s , i t was n � t n e c e s sary to
measure the peak of the very fast edge o f the s i gnal
resul t ing from the i n i t ial impact on the wal l .
Spec tral ana l y s i s of the pressure s ignal s , d i g i t i zed
at 500Rz ident i f ied l i t t l e ener�y above f r e q uenc ies o f
about 50Hz . I t was therefore decided to d i g i t i z e the
s ignal at a rate of 1 00Hz . It was a c c e p ted that the
measured peak impac t force might be rather lower than
i f a h i gher d i � i t i zat ion rate had been us ed .

On s t art ing t e s t ing however , a l arge zero s ignal d r i f t

occurred from a l l o f the pre s s ure transduc e r s . In
many cases the d r i ft was greater than the f u l l range
of peak to zero impac t measuremen ts . Th i s was
a t t r ibuted to the heat ing and rapid cool ing , at the
air/water inte r fac e , of the e l ec tronic e l emen t s
at tached t o the d i aphragm, caus ing var iable output
from the transducers . As the drift was not l inear , i t
was not p o s s ible t o f i l t er the data for further
analys i s . Ad d i t ional l y, the transducers were
operat ing over a very narrow range of their in i t ia l
cal ibrat ion range , thus reducing accuracy o f
measurements and requir ing very large ampl i f ier
gains , which resulted in further he a t i n g of the
transducers . Pr ior to t e s t ing however , the range o f
pre s s ur e s t o b e measured was not known , there fore
selec t ion o f a transducer with an appropr i a t e range
was extreme l y d i f f i c ul t . The se problems r e s u l t e d in
s igni f i c an t changes to this part of the s t udy . The
use of pre s sure transducers was abandoned in favour o f
force measurements recorded b y a s imple force t ab l e ,
d i scussed be low.

4. 6 Force
measuremen t s
The force table was d e s i gned to mea sure horizontal
wave load ing s . Details o f the instrument are shown in
Figure 8 and �late 1 . The force tabl e element was a
rigid l ightweight aluminium channel s e c t i o n ,

A c l earance o f approxima t e l y lmm was a l l owed

c an t i l evered from a supporting p la t e by four proo f
rings .
be tween the bot tom fl ange of the channel s e c t ion and
the wooden t e s t sect ion base . A s im i l a r c l earance was
a l l owed b e t ween the end of the force table and the
flume wal l s . In th i s way , the force tab l e e l emen t was
free to move under wave l o ading and to de form
elas t i c a l ly the supporting proof ring s . The pro o f

20
rings , each of d iame ter 70mm, were constructed from
l. Smm thick aluminium tube. Foil strain gauges were
resin-bonded to the outer face of each r ing at the
points at max imum fl exure. Each pa ir of strain gauges
were connec ted in a fu l l bridge c irc u i t to produce an
output vol t age propor t ional to the appl ied force . A
prec1s1on strain gauge ampl i f ier un i t was used to
power the gauges and to ampl i fy the o u tput signal.

The c a l ibrat ion procedure was to c l amp the suppor t ing

plate such that the force tab l e e l emen t was
hor i zon t a l . Loads were appl ied incremen t a l ly to the
po int o f a t tachment to the force tab l e e l ement for
each proof r ing in turn . For each appl ied l o ad , the
output vol t age was recorded . Output vol t age was found
to vary l inearly with applied load over the
cal ibration range .

The force t able was not expected to respond well to

very high frequency forces induced by wave impac t.
This was due in part to the natural d amping o f the
system and also to the d i fficulty of e f f e c t ively
depth-averaging a wave pressure sign a l , the phase o f
which varies with height up the crown wal l . The
instrument described here is therefore considered most
sui ted to assessing the total depth- in tegrated
hor i zontal force imparted on a c rown w a l l by the
wave . Tt is this parame ter which has been considered
in the subsequent data analysis.

Max imum expe c ted load ing on the force t ab l e ,

c a l c u l a ted using a simpl ist ic design procedure for
wave forces on a ver t i cal ca isson ( Re f 3 3 ) , was
approxima t e l y SOON , under the most severe test
cond i t ions. Each proof r ing was there fore designed t o
withstand a 140l'l' load and remain wi thin its elas t i c
l imit.

No a t tempt was made to measure upl i f t forces on the

crown wal l. There are substan t i al probl ems asso c iated
with the accurate model scal ing o f f l ow through
permeab l e underl ayers and such upl i f t forces cannot he
simu l a ted r e l iably in a small scale f l ume mod e l .

The test programme was designed to assess the

inf luence o f the fol lowing parame ters on wave-induced
crown wa l l load ings :

a) Inc ident wave cl imate . Influence o f wave he ight ,

wave period and wave steepness ;
b) Wa ter leve l ;
c) Armour geometry.

21
 The wave cond i t ions app l i e d to each t e s t s e c t i on were

pre sented 1n s e c t i o n 4 . 2 and Table 1 .

iden t i c al to those used for overtopp ing measuremen t s ,
Tes t s were
conducted a t two water d e p ths , h 0 . 4m and h
= = O . Sm
re s pe c t ivel y , at the toe of the sl ope .

The var ious armour geome t r i e s t e s t ed are d e s cr ibed in

Figures 4 to 6 . These s e c t i ons were des i gned to be
repre s e n t a t ive of those c ommonly used on c o a s t a l
s t ructure s . In a l l cas e s , the model armour was
angular rock . In a l l of the force t e s t s , the he ight
o f the crown wal l , h , remained con s t an t a t l l Omm . It
f
should be noted that the crown wal l e l ement used for
overtopping t e s t s on s e c t ions 6, 8- 1 2 was s l igh t l y
sma l l e r at 8 5 mm high .

For each t e s t the amp l i f ied analogue vol t a �e s ignal

from each proo f ring o f the force table was d ig i t i sed
and the four data channels were logged s imul taneous ly
onto a PDP l l /7 3 m i n i c ompu ter . A t r i a l wave s e q uence
was imposed on the test r ig and the force table output
was logged . Th i s procedure was repeated several times
us ing the s ame wave s equence but varying the
d i g i t i s a t i on rate in order to determine that mo s t
appropriate for the te s t s . A rate o f 1 0 0Rz was
chosen ; this gave adequate represen t a t ion o f the
output s i gnal wi thout s i g n i f icant l o ss of d e t a i l s . A
higher frequency wou l d be required to d e s c r i b e wave
impact forces but these were not rel iably measured
by the force tab l e .

Each random wave t e s t was o f duration 1 000Tm where T m

i s the nominal mean wave zero-cro s s i ng p e r i od .

5 ANALYSIS OF
RESULTS

5.1 Overtopp ing

5.1.1 Emp i r i c a l r e l a t ionships

The mean and s t andard devia t ion o f the over t o p p ing

d i scharge was c a l c u l a t e d from the five samp l e s each
for 1 00 waves , c o l l e c ted during each t e s t . This rl a t a ,
together w i th the other input parame ters , was ana l ysed
us ing a number of d imen s ionless parameters d e r ived
from earlier work. The se methods are d is c u s s e d
b r i e f l y b e l ow .

Owen ( R e f 1 6 ) has used an e q ua t ion o f the form :

0* = A exp (-BF * / r ) ( �. l)

A and B are coe f fi c ients for d i f ferent s lope an� l e s

c o t a and r is a re l a t ive run-up or roughness

22
coe f f ic ient . The study from wh ich the above formula

a roughne s s coe f f i c ie n t r = 1 . A large prop o r t i on o f

is derived was conducted on smooth faced s eawalls with

this s tudy was however conducted us ing permeable rock

armou r , with an impermeab l e core . It is reasonable to
expect there fore , that the roughness c o e f f ic ient r
wi l l be s ig n i fi c a n t l y les s for the s truc tures in thi s
s tudy . A roughness c o e f f i c ient o f r = 0 . 5 -0 . 6 has
been sugge s t ed (Ref 16) for two or more layers of rock
armour. The relat ive roughness for smooth and rough
s l ope s , measured in th is study is d i s c u s s e d in Se c t ion
5. 1.2.

The data c o llec ted in this study was f i t ted to the

d imen s i on l e s s parameters sugge s t ed by Owen .
The coe f f i c ient o f regre s s ion for the f i t o f the d a t a
t o t h i s relationship was no t par t i c u l a r l y good , even
for s truc t ures o f the s imple s t geome try . Results from
the present sturly have been compared u s i n g th i s method
and examp l e s are shown in Figures 9-1 2 .

Ahrens & ?e imhaugh ( R e f 2 0 ) pre s en t over topping data

for a number o f s t ruc tures in a d i f ferent form. The y
sugge s t t h a t d i s charge is not well d e s cr ibed by
dimens ionl e s s parame ter s , but describe a d imen s ionle s s
freeboard parameter that takes account o f the l o c al
wave leng th . Th i s al lows a bet ter d e s c r i p t i o n o f
s tructure per formance i n r e l a t iv e l y shallow wa ter , and
is appropr i a t e to the shallow water te s t s in Ahrens &
He imbaugh ' s s t udy . Wh i l s t this method does appear t o
ind i c a t e some improvement on Owen ' s method , i t h a s the
d i s advantage of not be ing truly d imen s i on l e s s . Ahrens
& Heimbaugh derived equa t ions of the form given
b e l ow :

( 5 . 2)

Where 0 is a c o e f f i c ient with the s ame u n i t s as 0

( volume 7un i t time per me tre run o f wall) , c 1 i s a
d imen s ionle s s coe f f i c ient , and Ahren s ' d imen s ionl e s s
freeboard i s def ined in terms o f the l o c a l wave length
of the peak per iod , L ps '

RC
F' =
( 5 . 3)
(Hs 2 L ) 1/3
ps

Compa r i s on o f the results o f this s t udy w i th a

s impl i f ied version of the above equ a t i ons ( u s ing the
s h a l low water wave wavelength 19 = Tm/gh ) , sugge s t a
b e t ter r e l a t ionship than that d t s c u s s e d by Owen . The
measured data has been pre sented , for s e l e c ted te s t
s e c t ions a s -ln Q against F ' in Figures 1 3- 1 6 .

23
On careful examinations o f graphs o f R* against 0* ,
for values measured in t h i s s t udy , i t was noted that
there was a stronger dependence on d imen s ionl e s s
freeboard R c /H than o n wave s t eepne s s . A
5
d imens i o nl e s s r e l a t i onsh i p incorpora t i ng th is func tion
was derived and an e q u a t ion for rlimen s ionless
freeboard deve l oped :

F* = R* (tf)
R
s
= ("") (�l"
c 2
( 5 .4)

It is a l so sugge s ted that an equat ion o f the form:

B
Q* = A F* ( 5 . 5)

g ives a s l igh t l y b e t ter d e s c r i p t ion o f the

relat ionship o f Q* to F* than does an equa tion o f
exponen t ial form. Coe f f ic ients A and B have been
calculated for each t e s t s e c t i o n . Th e s e are given in
Table 3 . The e f fe c t of incre a s i ng the we ight ing o f
the func tion R c / H draws the data c l o s e r to a
�
regres s i on l ine , 1mproving the corre l a t ion coe f f i c i e n t
s ig n i f i c an t l y , par t i c u l a r l y for higher d i s charge
events .

The r e l a t ionship be tween ln Q* and ln F* is

demon s t rated in Figur e s 1 7 -2 9 . It shoulo be noted
that the results of t h i s s t udy were der ived in
r e l a t iv e l y deep water cond i t ions . Ahren s ' pred i c t i on
me thod , which was based l argely on r e s u l t s in shallow
wa t e r , may be more appropr i a t e at lower water leve l s ,
because o f the local wave l en�th e f fe c t s cons idered by
the inc lus ion of Lp s in the equ a t ion for F ' .

None o f the analy s i s methods d e s cr ibed above ,

expl i c i tly take account o f the s tructure geome try .
The use o f emp i r i c a l l y d e r ived coe f f ic ients involves a
s i g n i f i c ant simp l i f i c a t i o n o f the d e s c r i p t i o n o f the
overtopping proce s se s . The e f fe c t s o f berm width ,
armour cre s t po s i t ion and v e r t i c a l wa l l freeboard mus t
each have a n e f fect o n d i scharge , ins ide o f c e r t a in
threshold leve l s . It seems l ik e l y that the geome t r i c
var i a t ions d e s c r ibed above w i l l have t h e m o s t
s i g n i f i c an t e f f e c t f o r a l im i ted range o f
d imensionless freeboard v a l ue s . When the freeboard i s
such tha t the crown wal l i s inundated , r e l a t iv e l y
small g e ome tric v a r i a t ions a t the c r e s t are l e s s
s i g n i f i c an t and wi l l have n o no t ic e a b l e effec t o n
d i scharg e . For very h igh d i s ch a rge event s , i t s e ems
l ik e l y that the r e l a t i v e l y s im p l e rel a t i onship o f
d imen s ionless freeboard , F* , t o d imens io n l e s s
overtopping , Q* , d o e s prov i d e a n ade q u a t e d e s c r i p t i o n
o f the overtopp ing per formance o f t h e s tructur e . The

24
threshold for wh ich t h i s r e l a t ionsh i p h o l d s is however
not e a s i l y d e f in e d . The r e s u l t s from t h i s s tudy
B
suggest that an equat ion o f the form AF* =
0* i s
v a l id where 0* > 2 x l o - s . Th i s arguement i s d e s c r ibed
in more detail b e l ow .

S im i l arly the crown w a l l g e ometry i s l e s s s igni f icant

for low d i s charge cond i t ions , when the freeboard is
very large. There i s however a range of events for
wh ich the crest geome try plays a s i g n i f i c ant par t in
the overtopping per formance of the s tr uc ture . The
graphs of -lnO* aga i n s t -lnF* , shown in F igures 1 7-29
s ugge s t that the r e l a t ionship be tween F* and 0*
becomes weaker for values o f O* < 2xl o - s . Th i s may be
due to two fac t or s . Firs t l y , the c rown w a l l geometry
may have s u f f i c i e n t e f fec t on overtoppin� to d i s t o r t
t h e r e l a t ionsh i p . Second l y , when 0 * i s sma l l , the
mean overtopping dis charge wil l gener a l l y be low and
there fore s ub j e c t to s ig n i f i c an t var i a t io n by
occas ional large wav e s . The confidence in the use o f
measured d i scharges o f re l a t iv e l y sma l l s amples of
waves , for pred ic t i on o f over topping , i s there fore
much reduc e d . The l arge c o e f f i c ient of var iat ion o f
the five sampl e s measured in each t e s t in th i s s tudy ,
for low mean di scharge cond i t ions , bears out this
p o in t . Thi s e f fe c t can be compensated for in one of
two ways . E i ther a we ight ing fac tor , proport ional t o
d i s charge , can b e appl ied to t h e d a t a , o r va lues of 0
b e l ow a certain l im i t can be d i s c oun ted in analys i s .
In this study , d i s charge events resul t ing in values o f
0 l e s s than 0 . 05 l i t r e s per second per me tre (mode l )
h ave not been included in the c al c u l a t ion o f
c o e f f i c i e n t s A & B , a s such low mean d ischarges are
s ub j e c t to large random var i a t ions . The impor tance o f
exclusion o f invalid data from a data s e t is
demonstrated in F igure 3 0 , which shows measured 0 *
against predicted 0* , using the who l e data s e t for a
s i ngle test sec t ion . E l im i n a t ion o f the low d i s charge
values resu l t s in a stronger correl a t ion between
predic ted and measured d i s c h arge . More impor tan t l y
though , the slope o f the pred i c t ion l ine and thus the
coe f f i c ients of A and B c an be a l tered q u i t e
s ignifican t l y b y changing the range . Coe f f i c ie n t s A
and B , for the r e s t r i c ted d a t a s e t , are g iven in
Table 3 .

The s c a t ter in the d a ta se t , 0* � 2 xl o - s , may however

r e s u l t from the geometry of the c rown wal l and armour .
B
I t i s poss ible that the r e l a t i onship 0* = AF* is too
s imple to d e s c r ibe the over topping o f a struc ture over
the range O* < 2 xlo - s . The f o l l owing r a t ios might
b e considered to have an e f f e c t on overtopping
performance o f a s tructur e . The r a t i o F / A d e s c r ib e s
c c

25
the freeboard in terms o f the ratio of the pro i ec t ion
of the crown w a l l above the armour cre s t , aga i n s t the
e l evation of the armour cre s t relative to s t a t i c wa ter
level . Th i s r a t i o may help to descr ibe the r e l a t ive
e f f e c t of various comb inations of wal l geome try and
armour on d i scharg e . S imilarly . the r a t io F c / G c may b e
used to d e s c r ibe the e f fect o f the pro j e c t ion o f the
crown wa l l above the armour cre s t , agai n s t the width
o f the hor i zontal armour cre s t berm. The geometry o f
the seaward face o f the crown wal l i s a l s o l ik e l y t o
have a s ig n i f i c an t e f fe c t o n the d i scharge for a range
of dime n s i o nl e s s freeboard s .

5. 1 . 2 E f fec tivene s s o f c r e s t geometry

An add i t ional method , for the compari son o f s tructures

with a crown wa l l with a smooth plain s l o pe , has a l s o
been used to d e s c ribe the rel a t ive performance o f
crown wa l l s e c t i o n s . Compar i sons o f the r e l a t ive
performance of d i f ferent cre s t geome t r i e s are
d e s c r ibed b e l ow .

The influence o f the geometry o f the te s t s e c t ions can

be described by a r e l a t ionsh ip of the form :

ll
0* 2 A2 F * 2
w =

f 0* 1 8
A I F* 1

w A F* B 3 (5 .6)
f 3

wher e , W f i s the e f f ic iency fac tor ; A 1 , B 1 , are

coe ffic ients for the performance of a smooth p l a i n
s lope and A , B a r e c oe f f ic ients d e s c r i b ing the
2 2
performance of a s tructure of more c omp l ex geome t r y .
The per formance o f each t e s t s e c t ion c an there fore be
related to a smooth pl ain s l ope , by sub s t i tu t ion of
the c o e f f i c i en t s g iven for each s tructure in Table 3 ,
for any value o f dimen s io n l e s s freeboard .

The r e l a t ive per formance o f each of the t e s t s e c t ions

i s desc ribed below, and the e f fe c t s of geome t r i c
var i a t ions o f t h e c re s t d e t a i l o n over topp ing
per formance d i s c us s e d .

Compar ison o f the per formance o f te s t s e c t ions 1 , 2 &

3 a l l ows the re l a t ive per formance o f impermea b l e
s tructures , of the s ame he ight , t o b e d e s c r ibed , The
presence of a narrow berm at the toe of the crown wa l l
in s e c t ion 1 , c ompared with s e c t ion 2 which has no
berm, appears to reduce the d ischar�e when F* i s
larger than about 0 . 1 3 5 . Di scharge i s however
s l ightly increased when F* is sma l l e r than 0 . 1 3 5 .

26
This may be explained by examining the effective
ver tical h eight of the crown wall. The r a t i o
F /G i s very close to 1. This ratio combined with
c c
the slope angle, cota = 2, reduces the effec tive
ver tical height of the wall, by 50% by extending the
slope traj ectory to the wall. When inund ation oc curs
the angle between the wall and the sl ope will fill
with water thus causing a ramping effect, and
increasing discharge. The vertical wall wi th no berm
however re turns wave action over a larger vertical
range, thus reducing overtopping at higher discharge
levels.

The smooth slope with no crown wall,, test section 3 ,

performs significantly worse than both test sections 1
and 2 , for all wave condit ions, indicating that the
crest geometry does have a significant effec t o n
discharge, for structures with the same c rest level.

The performance of a rock armoured structure with the

same profile as a smooth impermeable structure was
compared, by relating sec tions 4 and 1, which had the
same cross-section geometry. The rock armouring has a
marked effect on the disch arge. Section � has an
efficiency factor of Wf =
2
3. 6 x l o-2 F*- 1 · whilst
section 4 has an f ic 1ency factor o f
Wf
r �1
= 4. R xl0 - 4 F*- · . This difference represents
the effect of the roughness coefficient (r ) in Owen ' s
equation (Ref 16) .

The effect o f ext ension o f the c rown wall above the

armouring, increasing both and the ratio F /A
R was
c
examined by c omparing test s � c t i ons 4, 5 , 6, 7 and �
(shown in Fig �) . As expected, a reduction in
discharge results from increasing freeboard. In
addition however there is a general trend indicating a
slight reduction in discharge as a result of
increasing the ratio F /A . This is shown by
comparison of Figures �0, �1, 2 2 , 2 3 and 25 . This
indicates that a vertical c rown wall at the c r est of a
1: 2 slope is more efficient at reducing overtopping
than a plain 1 : 2 slope of the same crest level, for
the range tested, confirming the conclusion d rawn by
comparison of sections 1, 2 and 3.

The effect of widening the armour berm at the toe of

the crown wall reducing the ratio F /G is quite
. . . c c .
marked, resulttng 1n a constderable reductton 1 n
.

discharge f o r all events measured. This is

illustrated by comparison of Figures 2 3 and 2 5 with
27.

Variation of both A and Fe was investiga t ed by

c

27
c ompari son o f test se c t i ons 6 , 8 , 1 0 and 1 2 (Fig 6 ) . A
cons tant berm width was mainta ined in each t e s t
sect ion , as was the c rown wa l l leve l , but the level o f
the c r e s t o f the armour was var ied . The trend
observed on test sect ions 6 , 8 and 1 0 , shown by
comparing Figures 2 2 , 24 and 2 6 , sugge s t s that the
higher berms reduce the d i s charge for R* values
greater than about 0 . 06 . H i gher d i s c h arges were
however noted for values of R* less than about 0 . 06 ,
as the berm l evel incre a s ed . This may be expla ined b y
the fac t that the permea b l e mound becomes ful l y
saturated under severe c o n d i t ions , caus ing the rock
mound to act as a ramp , over wh ich the waves run .
Th i s once again sugge s t s that the crest freeboard
parame � er A /F may be a s i g n i f icant fac tor in the
c c
ana l y s 1 s of crown wal l overtopping .

The e f fect o f al ter ing the geometry o f the crown wal l ,

wh i l s t maintaining the same freeboard parame ter s , was
inv e s t igated by comparing a ver t ical wall with a
recurved wal l . The reduc tion in d i s charge brought
about by introduc in� a wave return rec urve on the
crown wall is quite drama t ic , as is shown by
comparison of Figure 29 with Figures 2 2 & 2 5 . It
should however be noted tha t the fit o f the data to
the regre s s ion l ine is not part icul arly good ,
sugge s t ing strongly that a more complex re l a t i o n s h i p
may be requ ired t o de scr ibe the per formance o f
recurved crown wal l s .

The repeatab i l i t y o f the t e s t i ng proced ure was

inv e s t i?ated by compar i son o f t e s t s 6 and 9 which were
carried out on an iden t i cal test s e c t ion . The re s u l t s
o f these t e s t s are shown in Figures 2 2 and 2 5 , and
suggest that the procedure adopted provides
repeatable resu l t s over the high di scharge range . The
regr e s s ion l ines are q u i t e c l o s e l y al igned for both
tes t s . There was however some consid erab l e scatter
observed over the low d i scharge range , where the
re sul t are very sen s i tive to individual d i scharge
even t s . This has resul ted in a s ig n i f i c an t v a r i a t ion
in correl a t ion c o e f f i c i e n t s for the s e two te s t s .

The resul ts o f overtopping coe f fi c i ents for all t e s t

sec t ions , are given i n Table 3 . These may b e used to
e s t imate the overtopping per formance for s tructures o f
the geometry t e s ted , for a wide range o f values o f
�-

28
5.2 Forc e s
5.2.1 Ana l y s i s procedure for random wave t e s t s

Typical s ignal outputs from each o f the proof r i n g

channe l s are shown i n Figure 3 1 . The l o c a t i o n s o f
each proof ring are d e s c r ibed i n Figure 8 . It i s
intere s t ing to note that the upper r ings experienced
less load ing than the lower one s . Some l a t e r a l
var iat ion i n load ing was a l s o eviden t ; t h i s wa s
a t t r ibuted to local d i f ferences in rock armour
pl acement d e t a i l . For each te s t , the four r e s u l tant
force time serie s were s imply summed and d iv i d ed by
the test s e c t ion width to g ive an e q u iv a l en t t o t a l
hor i zontal force per un i t width o f c rown wal l , �H .
Typ ical raw total force t ime s e r i e s are p r e s e n t ed in
Figure 3 2 ( a ) . It can be seen from thi s exampl e tha t
there i s a c er t a in amount o f no i s e s u p e r imposed o n the
s ignal . Although the s i gnal to no i s e r a t i o i s
rel a t ively h igh , the s ignal perturbat ions comp l ic a t e
any thresho l d cro s s ing type anal ys i s , par t i c ul ar l y
when the peak force l ev e l i s co inc ident w i th the
s e l e c ted event threshold l ev e l . This can be s e e n
c l early i n F igure 3 3 ( a) where mu l t i p l e thr e sh o l d­
cro s s ings would be pred ic ted from the raw d at a .
Add i t iona l l y , i t i s d i f f ic u l t to d e f ine a peak load
from the raw data . Qn many o f the force peaks , there
exist h igh frequency components which would not be o f
great struc tural s ig n i f i cance because o f the i r very
short durat ions . The inab i l i t y of the force t ab l e to
resolve high frequency wave impac t load ings h a s been
d i s c ussed prev i o u s l y in s e c t ion 4 . 6 .

It was dec ided to f i l ter the total force d a t a pr ior to

s ta t i s t i c a l analys i s , in order to remove unwan ted high
frequency compon en t s . The s e l e c t ion o f an approp r i a te
low p a s s f i l ter was somewh a t sub j ec tive . I f the
frequency c ut-o f f was too high , the s ignal c o u ld not
be smoothed ade q ua t e l y . I f the cut-off was too l ow ,
the inherent character i s t ic s o f the s ignal would be
mod i fied . The f i l ter selected was a low p a s s
Eu t terworth t ime domain f i l ter w i th c u t-o f f frequency
5Hz , appl ied over 5 pas s e s . Exam p l e s o f the r e s u l t ing
f i l t ered t ime s e r i e s are presented in Figures 3 2 ( b ) ,
and 3 3 ( b ) . It should be noted that appl i c a t ion o f the
f i l ter induces an e f fec t ive d e l a y of 0 . 1 s econds in
the resultant time s e r ie s ; but does no t a f f e c t the
s ta t i s t ical v a l i d i t y of the r e s u l tant d a t a .

For the c r o s s ing anal y s i s , an event threshold l ev e l o f

9N/m was se l e c ted . The s e l e c tion o f t h i s l e v e l a g a i n
required a s omewhat subj e c t ive j udgement . The l ev e l
chosen was j u s t above the peak o f zero l ev e l
fluc tuations .

29
The fo l l owing parame ters were derived from the d a t a :

a) UNFILTERED TIME SFR I E S :

i) Maximum recorded force (N/m run )

ii) 1 % exceedence force from entire d a t a s e t (N /m
run)

b) FILTERED TIME SERIE S :

i i i ) Maximum f i l tered force FHmax (N/m run) .

iv) 5 ak, exceedance force from thre sh o ld-cro s s ln�
peak data s e t F 5 (N/m run)
�
v) Mean threshold up-cro s s ing per iod T 1m
. p

v i ) Mean impa : t rat io IR = T imp /Tm wh ere :m i s

( second s )

_
the mean 1nshore zero up-cro s s 1ng per i od i n
second s .

In some t e s t s , for the large st incid ent wave

cond it ions , the force tab l e exh i b i t e d a zero l e v e l
instab i l i t y . Par t icularly large i n c i d e n t waves would
induce a permane n t o f f s e t to the recorded zero force
leve l . Th i s might be a t t r ibuted to a small p l a s t ic
deformation of the proo f ring s ; re l a t ive movem e n t s
be tween t h e r e s i n bonded s train gaug e s and the pro o f
ring s ; or some r e l a t ive movement between the proof
rings and t h e ir f i x ing bl ocks . The th ird explan a t ion
was considered to be mos t l ike l y . De s p i t e the induced
zero l evel o f f se t s , the c a l ibra t ion c oe f f i c i en t s for
each proof ring remained qui te constant throughout the
test programme . Where severe zero l evel in s t ab i l i t i e s
were experienc e d , a low pass f i l tered time ser i e s with
frequency cut-off 0 . 2Hz was subtrac ted from the raw
data prior to further ana l y s i s . Th i s procedure
improved s igni ficantly the qual i t y o f the measured
forc e s , but was t ime-consuming to apply and d id not
completely resolve ind ividual zero o f f se t s . It was
therefore only applied to bad l y corrupted d a t a .

5. 2 . 2 Resul t s

The t e s t r e s u l t s were a s s e s sed in three ways .

Fi r s t l y , for each o f the te s t s e c t ions , s ub j e c ted to a
c ommon random wave sequenc e , the d e r ived param e t e r s
for wave loading on the c rown wa l l were c ompared .
Th i s procedure was carried out for two o f the mos t
severe test wave cond i t ions ; in each c a s e r e l a t ive
per formanc e s o f the var ious rock armour con f i gur a t i o n s
were derived . Second l y , for a s ingle armour g e ome t r y ,
a more in tensive study was made o f the influence o f
various wave parame ters o n crown wal l load ing .
Final l y , where pos s ib l e , t e s t resul t s wer e compared
with those presented by o ther worker s .

30
The resul t s o f the comparative wave load ing analys i s
for each o f the test sect ions are presen ted in
Figures 34 to 39 and Tab l e 4 . The impa c t ratio , IR ,
d e s c r ibes the mean period between succe s s ive wave
impac t s on the c rown wa l l , relative to the wave zero­
cro s s ing per iod , The greater the v alue of TR, the
l e s s freq u e n t l y waves hit the wa l l . The two force
parame ters cons idered both relate to the f i l tered
total horizontal peak wave forces on the wal l ,
respec t iv e l y a s a 5% exceedence leve l , and a max imum
recorded leve l . The 5� exceedence value would be
expected to be more s table than the s ingul ar max imum
po in t . lloth the impac t ratio and the two force
parameters have been used to a s s e s s the r e l a t ive
sever i t y of wave attack on the crown wal l to al l ow

I n Tab l e 4 the parame ters I R , FHS and FH

compari son between the per formance o f each t e s t
sec t ion .
. . max
are a l s o presen ted as proport�ons of the e q u 1valent
parame ter for test sec t ion Bf. This enab l e s a
s impl i s t ic extens ion o f the informa t ion d e r ived for
sect ion 8f , concerning the influence of incident wave
cond i t ions on c rown wa l l loading . Sec t ion 8 f i s
similar to the structure for wh ich Jensen pre s en t s
d a t a ( Re fs 2 8 , 30 ) . The u s e of the force r a t io s ,
FHs /FH S and F
s f Hmax /FHmaxSf ' presen ted in this study
may be used to extend the scope of Jensen ' s work .

In each o f the compar a t ive wave load ing t e s t s , the

l e a s t severe wave loading , in terms both of IR and FH ,
occurred for s e c t ion 1 2 f . The second l e a s t severe
load ing occurred cons i s tently for s e c t i o n lOf . These
results would be expected intui t iv e l y a s the rock
armour comp l e t e l y protec ted the crown wal l face in
each c a s e . Se c t ion 8 f , with rock armour extend ing
hal f way up the crown wal l , was cons idered to prov ide
the next most e f fe c t ive armour pro tec t io n . S e c t ion
l l f had a wide rock berm with its c r e s t at the base o f
the crown wa l l . This was found to b e e q u iv a l ent to
s e c t ion Rf in terms of wave impac t occurrence on the
c rown wa l l , but induced force magnitudes were be tween
20% and 30% greater . The two most sever e l y loaded
test s e c t i ons were 4f , with a narrow rock berm a t t h e
b a s e o f t h e crown wal l , and 1 4 f w i t h a narrow
impermeable b e rm at the base of the wal l . A s might be
expec ted , s e c t ion 14 f , with a smooth impermeab l e
s lope , experienced the highe s t inc idence o f wave
imp a c t on the crown wal l . However , magni t ud e s o f wave
forces a c t ing on sec t ion 4f were approximat e l y twice
those for s e c t i o n 1 4 f .

The r e s u l t s o f the s tudy o f wave parame ter i n fluence

on crown wa l l loading s , for tes t sec t io n 8 f , are
pre sen ted in Figures 40 to 4 5 . The r e s ul t s are
considered in terms of fil tered peak forc e , F and
Fmax '

31
mean impac t period ra t i o , IR , with respe c t to wave
height , wave period and wave s t eepne s s , respe c t ive ly .
The resul t s pr esented r e l a t e to a single water depth
at the toe of the armour s l o pe , h = 0 . 5m . It is
sugge s ted from Figures 40 and 41 � hat there i s an
approximat e l y l inear dependence of peak force on both
wave height and wave per iod . However , further data
would be required to sub s t a n t i a t e th i s asser t i o n . A
d e t a i led anal y s i s o f the r e s u l t s shown in Figures
40-45 is presented and d i s cu s sed in Appendix A . An
al tern a t ive technique to that o f Jensen for the
e s tima t ion of wave fo rces on crown wa l l s is sugg e s ted .
However , further work i s required to r e s o lve
limit a t ions of the method and to extend its range o f
appl icab i l i ty .

The threshold wave cond i t ions below which no waves h i t

the c rown wa l l is no t c l e a r l y de fined from the d ata
analysed in this s tudy. The thre shold is not of great
signi ficance in the s t ruc tural O e s ign o f the crown
wa l l but if requ ired , is perhaps b e t ter a s s e s sed us ing
wave run-up rel a t ionsh ips such as those which appear
in Reference 2 5 .

The mean wave impact per iod T imp g ives in forma t ion
about the frequency with which the c rown wall is h i t
b y waves . It does no t , however , de fine the dur a t ions
of quas i-hydro s t a t ic l oad ing . Further anal ysis would
be required to study the load ing duration parameter .

The max imum wave forces measured in each of the t e s t s

o n sec t ion 8 f have been p l o t ted i n the non-dimen s ional
format proposed by Jensen in Feferences 2R and 3 0 .
� e parame ter FHmax / pg� f Lp i s p l o t t ed aga inst Bs /Ac in
F1gure 47 . The best f 1 t l 1ne _ from Jensen ' s resu l t s
( Figure 4 9 ) i s a l s o compared with data from this s t udy
in Figure 4 7 . There is reas onable agreement between
the d a ta s e t s , in terms of the s lope of the best f i t
l ines . The scatter o f d a t a from th i s study , about the
best f i t l in e , s ugge s t s that the r e l a t ionship may be
more complex than is sugge s t e d by Jen sen . Closer
examinat ion of the c l u s t e r o f po ints in the centre o f
the graph reveal that the s e s c a t tered po ints have a
c ommon s igni f i c an t wave h e igh t but var ied mean wave
period . A s trong linear dependence o f force on wave
period is s ugge s ted by the trend shown in Figure 4 0 .
Th i s dependence i s no t shown in the relat ionship
sugge s ted by Jens e n . A comparison be tween measured
and predic ted forces ( us in g Jensen ' s pred ic ted line
from Figure 49) i s shown in Figure 4 8 . This ind ic a t e s
that Jensens pred i c t i o n is within ±30% of the resul t s
g iven in this s t udy .

32
5.2.3 Calculat ion o f hori zon tal wave forces on the
crown wa l l

In o rder t o a s s e s s hori zontal wave l oadings o n a c rown

wall i t is nec e s s ary to make use of resul t s from
hydraul ic ph ysical model t e s t s , conduc ted for
structures of s im i l ar geome t r i c a l configuration to
those o f intere s t . No pro totype mea s urements have
been reported in the l i terature . In general , the
hydrodynam i c s of the problem, for an armoured
struc tures , are too complex to model rel i ab l y u s ing
numer ical or analytical technique s .

Dimens ionl e s s empirical rel a t io n s h i p s are pre s e n ted b y

Jensen ( R e fs 2 8 & 30) for the pred i c t ion o f max imum
wave forces on three d i f ferent type s o f c o a s t a l
s tructur e . Jensen suggested tha t , f o r a given
s tructural form , there e x i s t s a l in e ar r e l a t ionship o f
the form :

a + b ( 5 . 6)

Where a and b are emp ir ical coe f f i c i en t s and FHmax i s

the predic ted max imum hor i zontal force per me tre wh i c h
might be expected to ac t on a c rown wal l during a
random sequence o f 1000 inc ident waves o f g iven
s i g n i f icant wave height H and mean wave per iod Tm .
s

This method provides a valuable f i r s t e s t imate o f wave

forc e . Howev e r , it appears tha t the method by wh ich
the r e l evant parameters have been non-dimen s ional i s ed
is no t compl e t e l y val id , and tha t the in fluence o f
wave period on c rown wal l force is not represented
adequa te l y . It is suggested that wave force s , FH ,
pred i c t ed us ing Jensen ' s best f i t l in e s are gen er a l l y
accurate t o ± 3 0 % . Fur thermore , the i n f l u enc e o f
armour geometry i n reducing wav e load ings has n o t been
addre ssed .

The present s t udy has addre s sed the i n f luence o f

armour geometry on crown wall l oad ing . The armour
c o e f f i c ients Hmax /Hmax Sf quoted in Table 4 may be
used to extend the H
max values pred i c ted by Jense n ' s
r e l a t io n s h i p to structures o f d i fferent c r e s t armour
detail .

An al terna t ive approach i s sugge s ted in Appendix A ,

but t h i s i s s t i l l under developmen t . Further work i s
required t o re s o lve s ome o f the l im i t a t ions and to
extend the range of appl i c ab i l i ty be fore i t can be
used with c o n f idence .

33
6 RECOMMENDATION S

6.1 Rec ommend at ions

for d e s i g n
c a l cu l a t ions
6.1.1 Overtopping

The ove rtopping per formance of a breakwater c rown w a l l

c a n b e described by a n equa tion of the form

B
Q* = AF * (6. 1 )

Coe f f ic ients for A and B for the c rown wa l l

c on f igura t i ons tes ted a r e given in Tab l e 3 . This
method of pred ic t ion prov ides a bet ter d e s c r i p t ion o f
overtopping than equa t io n s o f the form suggested by
Owen for s imple s lopes ( Re f 1 6 ) and Ahrens & H e imbaugh
( R e f 2 0 ) for a revetment and wave wal l .

Coe f f ic ients of A = 7 x l 0 4 and B = - 1 . 85 are sugg e s ted

for a smooth slope in equa t ion 6 . 1 .

The model test confirms that the fol lowing fac tors
wi l l reduce wave over topping :

a) Increas ing the freebo ard o f the vert ical wa l l

( Fe) ;

b) Increas ing the rock armour berm width (G c ) ;

c) Concave seaward faces o f the crown wall w i l l g iv e

a better per forman c e than vert i c a l crown wal l s o f
the same he ight ;

d) Incre a s ing the freeboard o f the rock armour ( A c )

and reducing the r a t io F c / A , w i l l reduce
. . �
overtopp1ng for mo s t cond 1 t 1ons , except for
cond i t ions tha t lead to par t icularly high
d i s charges .

6 . 1 . 2 Wave forces on a c rown wa l l

The me thod o f Jensen ( Re f 2 8 and 30) to a s s e s s wave

fo rces on crown wal l s i s d e s c r ibed in section 2 . 3 .
For a se l e c t ion o f struc tural geome t i e s Jensen
presents r e l a t ionships of the form:

( 6 . 3)

where a and b are emp i r i c a l c o e f f i c i en t s .

The FHmax values pred i c t ed us ing Jensen 1 s method may

34
 now be extended to structures wi th d i f ferent c r e s t
armour geome tries us ing the armour coe f f i c i en t s
presented i n Tab le 4 .

6. 1 . 3 S l id i ng

In d e t e rmining the weight o f a c rown wall to r e s i s t

s l i d ing a value for the coe f f i c ient o f fr i c t ion � =

0 . 5 should generally be used unl e s s :

a) the crown wa l l s l ab is keyed down into the layers

below, when � up to 0 . 7 may be appropr i a t e ;
and / o r
b) te s t s have confirmed a d i f ferent value .

Where the crown wall s i t s on underlayer or s e c ondary

armour , the pr es sure d i s t r ibution on the under s i d e o f
the crown wall s l ab may gener a l l y be a s s umed to be
triangul ar , varying from a maximum at the front to
zero at the rear , drained , face .

6.2 R e c ommendations for

good prac t i ce
A s e r i e s of b a s i c guidel ines have been d e r ived from
the phy s i c a l model s tudies and from the o ther s t ud ie s
d i scussed in this report . These gui d e l i n e s are
general in nature and may be inappropr i a te in c e r t a i n
circums tance s . They do however prov ide the reader
with s ome b a s i c guidel ine on pre l iminary
cons iderat ions for the des ign of crown wa l l s .

Where poss ible the des ign o f the crown wal l

cro s s-sec tion should ensure tha t :

a) the shape o f the crown wall w i l l throw any

over topping water c lear o f vulnerab l e par t s o f
the rear s l ope ;
b) the upstand is kept as low as po s s ib l e
c ommen surate with per formance ;
c) the crown wa l l s l ab should b e c a s t o n the l e a s t
permeable ma terial where poss ibl e , to prevent the
transm i s s ion o f large volumes o f wat e r and
entrapped a i r through to the l e e - s i d e ; and/o r
d) the crown wa l l should b e keyed i n t o the mat e r i a l
be low b y a ' he e l ' o r ' down s t an d ' a t the s e award
side;
e) the crown wal l should be cast a t a s u f f i c ient
level above s ta t ic water l ev e l , to a l low
cons truction without casting ope r a t ions being
hampered by water ingre s s .
f) pre- c a s t parpet sec t ions should gene r a l l y be
avoided owing to the ir low coe f f i c i en t o f
fr i c t io n against s l i d ing .

35
7 ACKNOWLEDGEMENTS
The work a t Hydraul ics Rese arch covered by this report
was conducted by memb ers of the Co a s t a l Structures
Sect ion in the Marit ime Eng ineer ing Depar tment . The
authors are grate ful to nr K A Powe l l for his work on
the l i terature review , Mr A R Channe l for a s s i s tance
in the tes t s ; and to the i r c o l l e ague s in techn ical and
support s e rvices for a s s i s tance in the analysis and
produc tion of th i s repor t . The authors are also
grateful for the adv i c e and comments o f 0 J Jensen o f
the Danish Hydraul ic Ins t i tute .

36
8 REFERENCES
l. A l l sop N W H & Wood L A . "Hydro-geo technical
per formance o f rubble mound breakwa t e r s " . Report
SR 9 8 , Hydrau l ic s R e s e arch , Wa l l ing ford , March
1987.

2. A l l sop N W H . " Concrete armour u n i t s for rubble

mound breakwa ters and sea wal l s : recent
progre s s " . Report SR 1 0 0 , Hydr aul i c s R e s e arch ,
Wa l l ingford , March 1 9 8 8 .

3. Bradbury A P , A l l sop N W H , La t h am J-P , Mannion M

& Poole A B . "Rock armour for rubb l e mound
breakwa ter s , sea wa l l s , and reve tmen t s : recent
progre s s " . Repo r t SR 1 5 0 , Hyd r a u l i c s �e se arch ,
Wa l l ingford , March 1 98 8 .

4. Bradbury A P & A l l sop N W H . " Durab i l i t y o f rock

on c o a s t a l s truc tur e s '' . Proc 2 0 t h Coas t a l Fng
Conf , Ta i pe i , November 1 9 8 6 . ( Al so as Hydrau l i c s
Research Conference Paper N o 1 . )

5. A l l sop N W H & Latham J-P. " R ock armouring to

unconventional breakwa t ers : the d e s i g n
impl i c a t ions for rock durab i l i ty " . Proc . Seminar
on unconvent ional rub b l e mound breakwa ters .
NRCC, Ottawa , Sep tember 1 9 R 7 . ( Al s o as
Hydrau l i c s Research Conference Paper No 2 . )

6. L i l l evang 0 J , R a i c h l en F R , Cox J C &

Behnke D L . "A d e t a i l e d model s tudy o f damage to
a l arge breakwa t e r , and model v er i f i c a t ion of
conce p t s for repair and upgraded s t r ength" . Proc
1 9 th Coas tal Eng Con f , H ous ton , S e p t ember 1 9 84 .

7. Gunbak A R & Erg in A . " Damage and repair of

An talya Harbour breakwa ter" . In De s i gn and
Cons truc t ion o f mounds for breakwa ters and c oa s t
protec tion , E l sev ier , Ams terdam , 1 98 5 .

8. Al l sop N W H & S t e e l e A A J . " ll shdod por t

development-north breakwater s t ab i l i ty s tudies11 •
Report EX 1 1 5 � ( Re s tric t e d ) , Hydraul i c s R e s e arch ,
Ha l l i ng ford , May lq8 5 .

9. Gunbak A R . "Damage to Tr i p o l i harbour north

we s t breakwa ter11 • In De s i g n and con s truc t ion o f
mounds for breakwa ters and coa s t al pro t e c t ion ,
F.l sevier 1 9 8 5 .

10. Barony Y S , Mal i ck V D , Gushy M & Sehery F .

11Tr ipo l i harbour north west breakwa t e r , and i t s
probl ems" . P r o c 1 s t COPEDEC Conf , C o l umbo , March
1983.

37
ll. Abdel baki A & Jensen 0 J. " Study o f prov i s ional
repair o f the breakwater in Port d ' Arzew El
D j edid" . Proc 1 s t COPF.DEC Con f , Co l umbo , March
1 983.

12. llurcharth H F . "The l e s sons from r e c en t

breakwatr fa ilure s , developments i n breakwa ter
d e s ign". Pre s e n t ed to Technical Congr e s s , World
Federation of Engineering Organ i s a t ions ,
Vancov er , May 1 9 87 .

13. Port S ines Inve s t igat ion Pane l . " Fa i l ure o f the
breakwater a t Por t Sine s , Portug a l " . Amer ic an
Soc iety o f Civil Engineer s , New York, 1 9 8 2 .

14. Zwamborn J A . " An a l y s i s o f c a u s e s o f damage to

Sines breakwa teru . Proc Coa s t a l Structures 7 9 ,
Alexandr i a , ASCE , March 1 9 7 9 .

15. llruun P (Ed) . " De s ign and Con s truc t ion o f mounds
for breakwa ters and coast pro t e c t ion" . E l s ev ier ,
Amsterdam, 1 9 85 .

16. Owen M W . "Des ign o f seawa l l s a l l owing for wave

overtopping" . lleport EX 9 2 4 , Hydra u l i c s
Research , Walling ford , June 1 98 0 .

17. Owen M W. " The h ydraul ic d e s ign o f sea wa l l

pro f i l e s" . Proc Con f Shor e l ine Pro t e c t i o n , ICE ,
Southampton, Sep tember 1 9 8 2 .

18. Owen M W . "Overtopp ing o f s e a de fences" . Proc

Conf Hydrau l ic Mod e l l ing of Civil Eng Struc ture s ,
llHRA, Coventry , S e p t ember 1 9 8 2 .

19. Steele A A J & Owen M W . "Pre s t a t yn coast

de fenc e : pred i c t i o n o f d i s charge over a pro p o s e d
seawa l l " . Report E X 1 3 3 5 , Hydrau l ics R e se arch ,
Wal l ingford , 1 98 5 .

20. Ahrens J P & Heimbaugh M S . " Irregular wave

overtopping o f sea wal l s " . Proc Conf Oceans ' 86 ,
IEEE, \olashington, Sep tember 1 9 8 6 .

21. All sop N \ol H . ''Low cre s t breakwa t er s , s tu d i e s in

random wav e s " . Proc Coas t a l Structures ' 8 3 , ASCE
Arlington, March 1 9 8 3 .

22. Jensen 0 J & Sorensen T . "Over s p i ll ing/

overtopping of rubble mound breakwa t e r s . �esul t s
o f studies , u s e ful in d e s ign procedur e s " .
Coastal Eng Vol 3 , No 1 , E l sevie r , Ams terdam ,
1 97 9 .

38
23. Vera-Cruz D . "Over topping o f rubble mound
breakwater with cur t a in wa l l " . La B o i l l e
Blanche , N o 5 , 1 97 2 .

24. A l l sop N W H , Franco L & Hawkes P J . "Waves

run-up on steep s l ope s : a l i tera ture review" .
Report SR 1 , Hydraul ics R e s earch , Wal l ing ford ,
March 1 98 5 .

25. A l l sop N W H , Hawkes P J , Jackson F A & Franco L .

"Wave run-up on steep s l ope s : model t e s t under
random wave s " . Report SR 2 , Hydrau l i c s Research ,
Wal l in g ford , Augus t 1 9� 5 .

26. Tautenhaim E , Koh l ase S & Parten scky H W . "Wave

run-up at sea dikes und er ob ique wave approach" .
Proc 1 8 th Coa s t a l Fng Conf , Cape Town, 1 9 8 2 .

27. A l l sop N W H . " Sea wa l l s : a l i t erature rev iew11 •

Report EX 1490 , Hydraul i c s Rese arch , Wal l ing ford ,
Sep tember 1 98 6 .

28. Jensen 0 J . ''Breakwater s upers tructures'' . Proc

Conf Coa s t a l Struc tures R3 , ASCE , Ar l in � t o n ,
March 1 98 3 .

29. Gravesen H & Sorensen T . '' S t ab i l i ty o f rubble

mound breakwa ters11 • 24th l'IANC Congres s ,
Leningrad , 1 9 7 7 .

30. Jensen 0 J . " A monograph o n rubble mound

breakwa ters11 • Danish Hyd r a u l i c Ins t i tute ,
Horsholm, November 1 984 .

31. Gunbak A R & Gokce T . "Wave s creen s tab i l i ty o f

rubbl e mound hreakwa t er s " . Athens Conf 1984 .

32. Lundgren H . "Wave shock forc e s : An an a l y s i s o f

de formations and forc es in the wave and in the
found a t ion" . Proc Symp R e s e arch on wave ac t ion ,
Del ft , 196 9 .

33. Goda Y . "Random seas and d e s ign o f mar i t ime

s truc t ure s . " Univer s i t y o f Tokyo Pre s s , 'l'okyo
1985.

34. Jensen 0 J & Juhl J . "Wave over topp i ng on

breakwa ters and sea d ike s " , l'roc 2nd COPEDE C ,
Be i j ing , Ch ina , Sep tember 1 9 8 7 .

35. Bradbury A P & A l lsop N W H . ''Hyd r a u l ic e f fe c t s

o f breakwa ter crown wal l s " . Paper to breakwa ters
8 8 , ICE, E a s tbourne , May 1 9 8 8 . [ A l s o ava i l able
a s HH Conference Paper No 1 3 ] .

39
36. Fukud a N , Uno T & Irie 1 . " F i e l d observat ions o f
wave overtopping o f wave absorbing revetmen t s .
Coa s t a l Engineering in Japan , Vo l 1 7 , 1 9 7 4 .

37. Goda Y . ''Expected r a t e o f irregular wave

over topping o f sea wal l s" . Coa s t a l Eng ineering
in Japan, Vo l 1 4 , 1 9 7 1 .

40
TABLES.
TABLE 1 Tes t cond i t ions

TEST H (m) T (s) h ( m)

si p
(Water depth
at toe of t e s t
s e c t i on)

1 0 . 09 1 . 20 0 . 50
2 0 . 16 1 . 20 0 . 50
3 0.12 1 . 40 0. 50
4 0. 16 1 . 40 0 . 50
5 0 . 09 1 . 60 0 . 50
6 0 . 12 1 . 60 0 . 50
7 0.16 1 . 60 0 . 50
8 0 . 20 1 . 60 0 . 50
9 0. 16 1 . 80 0 . 50
10 0 . 20 1 . 80 0 . 50
11 0. 16 2 . 00 0 . 50
12 0 . 16 1 . 20 0 . 40
13 0. 1 5 1 . 40 0 . 40
14 0. 13 1 . 60 0 . 40
15 0 . 16 1 . 60 0 . 40
16 0. 19 1 . 60 0 . 40
17 0 . 16 1 . 80 0 . 40
18 0. 18 1 . 80 0 . 40
19 0.16 2 . 00 0 . 40
TABLE 2 Tes t sect ion construction

TEST SLOPE SLOPE WALL R F A G

c c c c
SECTION TYPE CREST CREST

( Co t cx=2 ) L E VEL LEVEL (m) (m) (m) (m)

l SMOOTH 0. 555 0. 70 0. 20 0 . 145 0. 055 o. 15

2 SMOOTH o . 700 0 . 70 0 . 20 0 . 000 0 . 200 o. 00
3 SMOOTH 0 . 555 0 . 70 0 . 20 0. 145 0.055 o . oo
4 ARMOURED 0 . 555 0 . 70 0 . 20 0 . 145 0 . 055 0. 15
5 ARMOURED 0 . 55 5 0. 76 0.26 0 . 20 5 0 .055 0.15
6 ARMOURED 0. 555 0 . 64 0 . 14 0 . 085 0 . 05 5 0. 15
7 ARMOURED 0 . 55 5 0.67 0. 17 0. 115 0 . 05 5 0.15

8 ARMOURED 0. 555 0 . 64 0 . 14 0 . 040 0 . lOO 0.15

9 ARMOURED 0 . 555 0 . 64 0 . 14 0 . 085 0 . 055 0. 1 5

10 ARMOURED 0 . 555 0 . 64 0 . 14 0 . 000 0 . 140 0. 1 5

ll ARMOURED 0 . 555 0 . 64 0 . 14 0 . 0 85 0.055 0 . 30
12 ARMOURED 0 . 55 5 0 . 68 0 . 18 0 . 000 0 . 180 0.15
13 ARMOURED 0 . 55 5 0 . 64 0 . 14 0 . 08 5 0 . 05 5 0.15

ALL LEVELS ARE RELATIVE TO THE TOE OF THE TEST SECT ION (m)
TABLE 3 Summary of empirical coefficients for various crown wal l

configurations

Test Sec t ion A B Corr e l a t ion Coe f f i e ien t

R2

1 5.0 X 1 0- 7 -3. 098 0 . 93

2 3.4 X 1 0- 6 -2 . 03 3 0 . 81
3 1.4 X 1 0- s -1 . 848 0 . 70
4 6.7 X 1 0- 9 -3 . 4 5 7 0. 81
5 3.6 X 10-9 -4 . 36 8 0 . 93
6 5.3 X 1 0- 9 -3 . 5 1 4 0 . 84
7 1.8 X 10-9 - 3 . 600 0 . 96

8 1.6 X 1 0- 9 -3 . 1 8 2 0 . 84

9 1.3 X 1 0- s -2. 585 0 . 67

10 3. 7 X 1 0- 1 0 - 2 . 920 0 . 73
11 1.0 X 1 0- 9 - 2 . 82 3 0.61

12 1.3 X 1 0- 9 -3 . 8 1 7 0 . 80

13 5.9 X 1 o- 1 o - 3 . 1 54 0 . 71
TA5LE 4 Resulte of comparative •ave loading

SECTION IMPACT RATIO IR/IR 8 f ORDER OF 5% EXCEEDENCE FH 5 /FH 5 ORDER OF HAXIHUM FILTERED F /F ORDER OF
Bf Hmax Rmax
NUMBER IR • T. /T INCREASING ZILTERED PEAK INCREASING PEAK FORCE INCREASING
1mp m
SEVERITY FORCE (N /m run) SEVERITY F m (N/m run) SEVERITY
H ax
FH 5

Teat A, wave cond i t ions : H8 • 0 . 2m , T • 1 . 6s , d • 0 . 5m

m

4f l. 52 o. 71 5 442 3.13 6 75B 3 . 59 6

Bf I . 9B 1 . 00 4 141 1 . 00 3 211 1 . 00 3

!Of 2 . 29 1.16 2 117 0 . 83 2 145 0.69 2

!If 2 . 05 1 . 04 3 169 1 . 20 4 260 1.23 4

12f 4 . 78 2.41 I 42 0.30 I so 0.24

1 4f l .41 o . 71 6 222 l . 57 5 375 I, 7B 5

te s t B , wave cond i tions : U 8 • 0 . 2m , Tm • 1 .8s, d • O . Sm

4f 2 . 06 ! . 04 4 553 3.14 6 B49 3 . 52 6

Bf ! . 9B 1 . 00 5 176 1 . 00 3 241 1 . 00 3

!Of 2. 2 7 1.15 2 lOB 0 . 61 2 149 0 , 62 2

llf 2.07 1.05 3 232 1 . 32 4 313 1 . 30 4

1 2f 3 . 72 ! . BB 1 48 0.27 I 65 0,27

14f 1 . 54 0.7 B 6 247 1 . 40 5 434 1 , 80 5

F I G U R ES.
 -+-1
Be
�------ ' '
t--

...... ......
� E �...... _,"'::"' '
'
'
)- -·< ,. -' ,... ,...
/ '
'
" "
/ /
"""
'
/..J- - - - � .,.,
_,
'
/L.
'
_ _ _ _ _

'

Overall geometry A, B c E, t • . t u , o: 1 , o: , , h
Armour o.so· o .as• o.,s. w, '6 • . nv

Crown wall g e o m e t r y A c B c fc R,, G J, K

R e a r armour D,so· D,ss· D,,s

F ig 1 Geometry of breakwater cross sections

 5 6 7 8 9

Fig 2 O v e r t o p p i n g o f b r e a k w a t e r [ r o w n w a lls - a f ter G r�v e s e n e t al

 :

1 -

1
'
=o
� �
...
"
�
"

£
1 !

H--- , , � :�i
,1

/I
LJ_

r-
�
"

T
'
'"
�

H
•

l {I 1 'i
-
• �

�
�

--;,
!
�
'

l I!

F ig 3 D e e p r a n d o m w a v e flume
 T e s t s e c tion 1 - smooth slope
1SOmm 1700 g 1 66 5 l evel
o f top o f
f o r c e table
(section 14 fl
d = 15002

d = 14002

555mm
Impermeable
membrane

1000 'VL._ ______________ L___________

T e s t s e c t io n 2 - smooth slope

'V 1 555

Impermeable
membrane
1000
'V�-----

T e s t s e ct i o n 3 - s m o o th plain s l o p e
1700
'\7
2

1 �

Impermeable
membrane
1000
'\7�------

'V Levels relative t o floor o f flume

In all t e s t section figures

All dimensions mm model

Fig 4 . T e s t s e c t i o n s 1, 2, 3 a n d 1 4 ( s m o o t h s lo p e s )
 C o m p a r i s o n o f t e s t s e c ti o n s 4 - 7
Variations o f crown wall crest level

=
T e s"--
t = ='-'-'5'---"-
section ="----- ..---- ...,
sz1760
I I
I I
I I
I I
I I
_,_
T-"'
e>J
1
s t__._
s.,
e ,_,_ 1 o'L'
c t'-" n_4 sz
,__,L1!!7-"
0!L
O L
1 ---
�
_

1
g 16 6 5 Is e c t ion 6 f I
L e v e l o f top o f f o r c e table
rest s o c tion 7
=:_:_=='--'--:LCC�--
_
y
�1670 t-- -- - -,1
T e s t section 6 & 9sz1640 : I

��t: ����":�
_.,---- - r-

r-
_
·

_ ____ 1 5 0m
m

R o c k armour ...-- - ----'iZJ�i.

/
....\// �---?---

Level o f top o f f o r c e table

T e s t s e c ti o n 11 w i d e r o c k b e r m
(section . 1111

Si' 1 6 4 0
v 16 6 5

Crown wall section

\
Rock a r mou
�
...-- ...- ------ - - - - - - - ::z 155� - - ­

..-­
/
/

Fig 5 Test sections 4-7 and 11

 Test section 8 Si' T o p o f f o r c e table

I 164o
(section B f l

Test section 10
Is e et la n 10 fl
Top of force table

----t-Si'"----'-"16"'-6"-5
1640

Rock

1555

T e s t s e c t io n 12

T o p of force table
!section 12fl
1665

R o c k armour 1640

\ /
..... .......\... 1555

Fig 6 C o m p a r i s o n o f a r m o u r c o n f i g u r a t i o n at c r e s t
 T e s t s e c t i o n 13 - r e c u r v e d w a ll

\1 1640

Rock armour

./
./
J.. 1 5 0mm -----1�

_ _ _ _ _ _ _ _ _ _ _
1555
.;L------���------�

Fig 7 T e s t s e c t io n 1 3 - recurved wall

 Proof ring
reference
number

[ ance between est section base

channel and base
board Isomet ric projection

Resin bonded
foil strain
1mm thick aluminium gauges 70mm diameter 1.5 m m
channel section t h ickness p r o o f rings
height = 110mm

Rock armourlng -
placement details 6mm thick aluminium
vary ortlng plate

ks

Impermeable w ooden
test section base

E n d elevation

r t:__
S u pp o.:.:
0 0 0

__ ...-� Channel sectio n

plate
0 0 0
(top flange rem oved)

S e c t i o n a l plan on A - A

Fig 8 F o r c e t a b le
 --- --- ,
TEST SECTION 3
- -
1 8 -,-- -- - --
- - -- --- - - - -- -
I
17
l I
l I
0 Me asured
16

� :: I
--- Predict e d this s t u d y

Predicted I a f t e r Owenl

� 13
�
1 2 -l
!
I
-�
o

11 J
I
10 � 0
0
'

�
--
9 o_ _
__

0 0 0 --
-
--
8� 0 --
-
-

�
-
0 0 - .a- -
-
..,
0
7
- l!t'
-
-
I --

6 +-·r - -,-----T--.--,-- -r-- .--.---

-, --,-
----,- ,--- -,
----,-- .- --,- -,--- 1 ----, --i
0.05 0.07 0.09 0. 1 1 0. 1 3 0. 1 5 0.17 0.1 9 0.21 0.23 0.25

Dimensionless Freeboard R *

Fig 9 T e s t s e c t i o n 3 R * v s - L n 0. *

TEST SECTION 4
18 , . l
- ______ _ __ ,__ ____ ------
-

17 0 Me as ur e d

16 -
0
P r e di c t e d this s t u d y
0
15 -

* 0 0
14
d 0 0

13
c
-' --
' 0 0
.,
0
0
1 2 -r
=
�
ro
0
.c
u
� -
Ci 11 '

0
�
10 -
� 0
c
0
·v; 9 0
.,
-,
c

.§
Cl 8 -

7 _j

6 -+ -i
-
- - I -T--l--T-- r·--- T·-· "T -T T - l l l
,
-- - 1 - - -- r
,

0.05 0.07 0.09 0. 1 1 0. 1 3 0. 1 5 0. 1 7 0. 1 9 0.21 0.23 0.25

Dimensionless Freeboard R *

Fig 10 T e s t s e c t i o n 4 R* v s - L n 0. *
 TEST SECTION 8
l
18 -:·

0
17 -·

16 �
0

* 15 � 0
d

0
"
--' 14 �
I

13
"'
"' �
�
m

0
.c

.!:2
u

0
12
0
�

OD
0
1 1 -

0
0
10 �

0
o Measured

0
9 ..,
--- P r e d i c t e d t h i s s t u d y

0 0
8 ...,_

7 -
i
'

s -+ 1 , --r � 1 � T -
r
i
-r r -, 1 . 1 •. ·1 ----j
- - --- -- -- ----- -"-"' ·-··- -- --· - - - -�

0.05 0.07 0.09 0. 1 1 0. 1 3 0. 1 5 0.17 0. 1 9 0.21 0.23 0.25

Oimenslonless F r e e b o a r d R *

Fig 11 T e s t s e c t io n 8 R * v s - L n 0. *

TEST SECTION 9
18 �

'
17 �
0

16 �

0
*
15 �

0
"
--'
14
I

"'
"' 13 �

�
m
.c

.!:2
u 12 �

0
� ...,
� 11
.!!
"
-� 10 -
"
�

"' o Measured
E 0
0
9
i5
�

-i 0
--- P r e d i c t e d t h i s s t u dy

0
8

7 .J!

r---�----T- --�- - - --1-- - -T -- - 1 '

6 -t ··- T -T r ---r··· T --�--T T- .,
-

0.05 0.07 0.09 0. 1 1 0. 1 3 0.15 0.17 0. 1 9 0.21 0.23 0.25

Oimensionless F r e e b o a r d R *

Fig 12 T e s t s e c t i o n 9 R * v s - L n 0. *
 TEST SECTION 3
17 ' ---------,

i
16
o M e as u r e d
15 -i
--- P r e d i c t e d this s t u d y
1 4 -:

1 3 ...

d
--'
c 12 ..,
I

11 I
�
"'
�

!
"'
.c

-�
u
10 '
c:J
0 0
9 -

8 �
0 0
§
7

'
6 - 0

5 - i---�--- - 1 ---- ·r ·
+ · --
r
--·--
--r
· .- -···· ··
1
·
1
·
• r · ·· ----r �
0.2 0.6 0.8 1 1 .2 1 .4

D i m e n s lonless Freeboard F'

Fig 13 T e s t s e c t i o n 3 F ' v s 0.

17 T
nEST SECTION 4
-- ---- - -l
1 6 -- I
'
I
0
15 -·

14 .. 0

13 ..,
0 0 0
d
--'
c
I 12 0 0 0
� 0 0
=
�
"' 11 ..,
.c
u
"'
c 10 _,

9 _,

0
0
8 . 0 0 Measured

7 --l -----'-- P r e d i c t e d t h i s s t u d y

6 --1
'
+
-i -- - - - �T - -·- ----1------�,----- -·-·· · · ------·-- ----- - r- -
5 I 1 r r-- ·· i
0.2 0.4 0.6 0.8 1 1 .4

Dlmenslonless Freeboard F'

Fig 14 T e s t s e c t i o n 4 F ' v s 0.
 :-· -
TEST SECTION 8
17

0
0
16 -i

0
'

1 5 -;

14

0
'

!
'

d 13

0
�

"

12 l
--'
I

"
en
�

11
"'
.c

0 0
.�
u '

c::!

0
10 '

0
0
9 -!

o
'
8 �

7 � 0
M e a su r e d

mJ
I
--'-- P r e d i c t e d t h is s t u d y

6 �

i'
5 + T
-· -·
'

0.2 0.4 0.6 0.8 1 .2 1 .4

Olmenslonless Freeboard F'

Fig 15 T e s t s e c t i o n 8 F' vs Q

TEST SECTION 9
'
0
1 7 -,-------- --�-- ---- ------------- ----

J
i
16

0
15 -1
14�
i
0
d 13 �
"

0
--'
I
QJ
12
i
0
I �
0
,

a
I
I

0
10

0
9 J 0
I
0
o
8 -!
0
7 ....
! M e asured

0
0 Pre d ic l e d t h is s t u d y

J
----'--

6
i

s -4- - -- r -- - r - ·-·--r� - - r --�-- ---

1 r-- -,---,--- ---,--- --,--- ---r --��
0.2 0.4 0.6 0.8 1 1 .2 1 .4

Olmenslonless Freeboard F '

F i g 16 T e s t s e ction 9 F ' vs Q
 TEST SEC110 N 1
1 8 -y-- ·---- ----· - - -- - ------------------------- - - - -------------

'

Ij o
17
* M e a s u re d
d
i
16
c ---'- P r e d i c t e d t h i s s t u d y
...!.
-J
15
'"
-
�
l
'" 14 -i
E
"'
"' 1 3 _ j'
�

0..

'"

0
"'
0
"'
I
� 12
.c

� 1 1 _j
u

c::J I
-

'
�
1 0 _j
�

�
.2
c I
I

I
� 9 .....

0
'"
c

0
...
J
0
c::J 8

0
I

7 ....'.

6 -+ - - · -- 1 ---- r- -- · -r · - --�- - r··---- i----- - r----,- - -- - -�----·-· r---- r·-r----T-- -··r · - · 1

0.4 0.8 1 .2 1 .6 2 2.4 2.8 3.2

Olmensionless F r e e b o a r d p a r a m e t e r (-Ln f * l

F i g 17 T e s t s e c t i o n 1 - L n f * vs - L n Q. *

TEST SECTION 2
1 8 T. l

1 7 --;
0 Measured
*
d i 6 -:
--- P r e d i , t e d t h i s s t u d y

�
0
--!
0
�
12

�
"'
.c

-� 11
c::J

1 0 --:

9 -;

6 -j- - -
i -- j " _T ___l . - --1 - - --T--
· :- ·r - r·· - - 1- -r
-r r
1.6
-

0.4 0.8 1 .2 2 2.4 2.8 3.2

Oimensionless F r e e b o a r d p a r a m et e r 1-Ln f * l

Fig 18 T e s t s e c t i o n 2 - L n f * v s - L n O. *
 TEST SECTlON 3
1 8 -,------- ------ ·-·--·-··- ·--·--1
1

�
I 0
17

�
Measured

* 16

��I
d --- P r e d i c t e d t h i s s t u d y

15

4
1
]"'
�
13

� !
QJ 12 �

� 11 J
� :j
0
1 0

-� l
�
.§
aJ
! i1

c : L-,.-.---,-----,---,--o-,---,---,--.---,-o--,---r--r --,----,---J
0.4 0.8 1 .2 1.6 2 2.4 2.8 3.2 3.6

Dlmensionless freeboard p a rameter (-Ln f * l

Fig 1 9 T e s t s e c t i o n 3 - L n F * vs - L n O. *
TEST SECTION 4
18 ··- -----�-------- ------ -- -- -··-- · -·- - - --�------ ·- - - -- - - - - - - - ·---,

*
17 r· o
1 6 -j
M e asure d
d '
c I
_,
..!.
---'- P r e d i c t e d t h i s s t u d y
15 _J
�
� I
.,
� I

e
14 �
"'
0
"'
� !
c. 1 3 -J 0
J
.,
"'
I

"' 12
�

.<=
u 0
.!::!
Cl 1 1
Ii
"
"' i
� 10
c l 0
.2
"'
" 9 -< 0
.,
e
i5 8 -
1
7 �

6 �- r···-

0.8
- -�------�---

1 .2
l------,- ----1-·----·-r- -·---,--- :--·-- - r -- - - 1 - -·

1 .6 2 2.4 2.8
T
3.2
--- r--
J
3.6
0.4
Oimenslonless freeboard p a r ameter (-Ln f*l

Fig 20 T e s t s e c t i o n 4 - L n F * v s - L n O. *
 TEST SECTION 5

lI "
18

17 _,

* o Measured
d 16 -1
� -----'- P r e di c t e d t h is s t u dy
� I I:t!
..!. O
15 __j O
�
"'
�
"' 0
E
14 '
"' 0

i'
�
"' 0
c. 13 -·

"'
"'
� 12 -·
'
I
"'

-1
.c 0
.!<!
u '
Cl
11

I
VI
VI
"' 10 '
c
0
·v; 9 -i
�
'
I
"'
.§
c:l 8 -j
7 _j
'

6 J._' r·
""
1
- -
r
- -- T � - --
1
- �------ ·r-· - .-
, - -- ·r - .. · r --- T
·--- - -
r
-- ---
·r
-
"1 ..,

0.4 0.8 1 .2 1 .6 2 2.4 2.8 3.2 3.6

Oimenslonless freeboard p a r a m e ter 1-Ln f*)

Fig 2 1 T e s t s e c t i o n 5 - L n F * v s - L n O. *

TEST SECTION 6
18 ---. -· ·· - -- ·- -- · -· ---- - - -- - - ---- - . - - - - - -- · ·- ... - - -- ---- ----- ----- -,

17 __j
o Measured
*
d 16 -i
� ---'- P r e d i c t e d t h is s t u d y
�
..!. 15 - 0 0
�
"'
14
�
"' -
E 0
"'
�
"' 13 -
c.
0
"'
"'
�
"'
12 ...j 0
.c 0
.!<!
u
11 -; o ct:J
Cl

VI
VI
..!:!
10 --1
'
<=
0 '
0 0
·v; 9 ...,

.§
<=
"'

c:l 8 ..., 0
I
i'
7 ..;

-1
'
.. - -- --�- - - r-----T · ·-r ··r--·-: - --, - --- r -�
+ I
-- - -- - --
6 �
· --
r
�--
.,.
-- ---- -
1
- -----

0.4 0.8 1 .2 1 .6 2 2.4 2.8 3.2 3.6

O i m enslonless F r e e b o a r d p a ra m e t e r 1-Ln F * l

F ig 2 2 T e s t s e c t io n 6 -L n F * v s - L n O. *
 TEST SECTION 7
1 8 -,--- - - - - -- ----- ----�-- - --------- -----,

1
'

:: j
o
* Measured
d
--- P r e dict e d this s t u d y

I
:U
15
I

r:;
0
0

Qj
0 0 0

0
a 1 1 -�
I 0
�
0
10

9 --.
i 0

I
0 0
8

J !
6h- 0.4
·r--T-----,---- --,-,---...,---,--,-r--·�--,- ·-------,-r -

0.8 1 .2 1.6 2 2.4 2.8 3.2

·-,---�
3.6

Dimensionless F r e e b o a r d p a r ameter 1-Ln F * l

F i g 23 T e s t s e c t i o n 7 - L n F * v s - L n O. *

TEST SECTION 8
---- --- - -- - - - - - ---,

17
!
-i 0 M e a sure d
*
d
16 _j
I P r e di c t e d t h i s s t u d y
I

�
c
15 JI
...!.
.../

I 0
14
�
<U
- 0
<U
e 0
� 13
�
�
Q. i
<U 1 2 -l
I
0
"'
0
i
�
�
0 0

::g j�
.c

.!:!
�

0
c:J
�
0
�
<U 0 0
0
"
0
·v;
0

c I
<U
a -j
I
e
0 ' 0

7
lI I
+---, - '1 - - 1---- ··r-- -T---�
6 - -- �--T- --r-·-··,-- · --T-·--T ---- -·T·-----i

0.4 0.8 1 .2 1 .6 2 2.4 2.8 3.2 3.6

Oimensionless F r e e b o a r d p a ra m e t e r 1-Ln F * l

F ig 2 4 T e s t s e c t i o n 8 - L n F * v s - L n O. *
 TEST SECTION 9
1 8 ·---; · ---- -.,

17 -
o Measured

*
d 16 -, ---'-- P r e d i c t e d t h i s s t u d y
�
�
-'- 15 --

�
<U
� 14
<U 0
e D
"'
�
D
"' 13
a.
'
<U '
"' 12 0
�
"'
-=
u
"' 11 __;
D
Ci D
"' D
"' 10 ' D
<U 0 0
"

·v; 9
0
_,
"
<U
0
.§ 8 -' 0
"'
D
7 -

6 _,' _ l- T . - l ·- 1 · - -r-----.- -r- · ·- r-· 1- - · r·--·-· r · -- 1

1 T . -j
0.4 0.8 1 .2 1 .6 2 2.4 2.8 3.2 3.6

Oimensionless F r e e b o a r d p a r a m e t e r 1-Ln F * l

Fig 25 T e s t s e c t i o n 9 - L n f * v s - L n O. *

TEST SEC110N 1 0
18 ,___
•.
-·-- - · -- ·

17 o
-·

*
M e as u r e d

d 16
i --- P r e dicted this s t u d y
i
�
�
-'- 1 5 _,
�
D
14
<U
<U
I
� -

E
"'
� __j
13
a.
"'

-.j
<U
"'
� 12 0
"' D
-= D
Ci
u
"' 11 -,

i
[JJ
D
"'
"' 10
�
0
0
.5!
"
0
"' 9 -

<U
"

E 0
i5 8 '
0

7 -1
I
'

- -j
r--- ·-r-·· -r ·-- -r- · - - ,- -· - T- -, -- --,- ·--r - - -- --, ,.--- -. - - -.·-
6 -L
' - .,
·· · -
,

0.4 0.8 1.2 1 .6 2 2.4 2.8 3.2 3.6

Oimensionless Freeboard p a r a m e t e r 1-Ln F * l

Fig 26 T e s t s e c t i o n 1 0 - L n F * v s - L n O. *
 TEST SECTION 1 1
18 :

17 _j 0 Measured
*
d 1 6 _; --'- P r e d i c t e d t h i s s t u d y

15 �

0 0

-t 0
i
12
o o
11 � 0 0
I 0
1 0 -1

0
i
9 --,
I
0
'
8 -,
I
I
7 -c

6 + - , - -- 1 - r -1 - , T - �
0.4 0.8 1 .2 1.6 2 2.8 3.2 3.6

Olmenslonless F r e e b o a r d p a r a m e t e r 1-Ln F * l

Fig 27 T e s t s e c t i o n 1 1 - L n F * vs - L n O. *

TEST SECTION 1 2
1 8 ,-- --- - -- -- -- - -- - - - ------ -- - --- -------- -- ---

1
17 �
* 0 Measured
d

---'- P r e d i c t e d t h i s s t u d y
1 5 __:
�

-
"'
0 0
"' 0
e 0

j
"'
�
"'
c. 13
"'

1 2 . ._;
"'
�
.c: 0
a
u
0
�
i
11
"' 0 0
"'
"' 1 0 -"
c:
0
·;;; 9 ..:
<=

e
"'
0 0
0 8 -�
'

, J
I
r

-+· .. ·-- ·-r , ,- - 1 - -,-- , -- -r- - ·r-- · -1 · r ----- '1 ----,

T 1'
-
6
- · - - - -- - -- -

0.4 0.8 1 .2 1 .6 2 2.4 2.8 3.2 3.6

Oimensionless F r e e b o a r d p a r am e t e r 1-Ln F * l

Fig 28 T e s t s e c t i o n 1 2 - L n F * v s - L n O. *
 -
TEST SECnON 1 3
18 :- -

1 7 -j

* 16 -.

d
c D
--' 15 �

-'-
D
�
�
14
�
�
D
E
"'
�
13
"'
c.

� 12 �

= D
�
"' D
11
'
""

-�
u
D D
0
D
V> 10 -
D
V> D
�
<=
9 �

·;;;
0

.§
<=
�
8
0
7

.. "-- . r· -- ·r·· .. . · r ·· . - ---1

6 ·+ r I- I !
-
r r
-
1 T 1
0.4 0.8 1 .2 1.6 2 2.4 2.8 3.2 3.6

Oimensionless F r e e b o a r d p a r a meter 1-Ln f*l

Fig 29 T e s t s e c t i o n 1 3 - L n f * v s - L n O. *

-8,00£+00 �------�-----
1
-9.00[+00 -i

- 1 .00E+01 j I
I
'0
I
w
- 1 . 1 0E+01
l
'0 I
u

-=
- 1 .2DE+01
-iI
w

*
�

d
!
c
- 1 .30E+01
-ji
I
�

- 1 .40E+01
.J�
Restricted d a t a s e t
- 1 .50E+01
� Ln U * Predicted L n 0. * observed
lfull data s e t )
=
I

j I
//
/
- 1 . 60E+01 D

- 1 .70[+01
l
-1---- - - - - - · ,- ·- ·· -
T
-··- -- ·
r
---- -,-· --- -
---,- --- .. -
.,-
·-- - ---- r· -·- -
\
--1
- 1 .60E+01 - 1 .40E+01 - 1 .20E+01 - 1 .00E+01 -B.OOE+OO

LN 0* (measured)

Fig 30 Comparison of measured and predi[ted dis[harges

-n

PROOF R I NG 1 PROOF R I NG 2
20 20 ,------,
=

.------.----,
UJ
·-.----,-
-r--,-----.- ,

m 15 15 � �
3
�
X

10 10
Q)

-o
i
5 5 �
ro
z: z:

0
� UJ UJ

0 I � NU
.,.. it •• 1
<->

0
<->

1t\111
1 1 tij;���-tttw
ill'\'
·
� er er

l
0 0
0

I"
..... .....
...,
'"'

-5 L-�---L--��
ro

_!__

220 120 220

-.­

TIME ( s i TIME ( s i
Q)

n>
a-

0
P R O OF R I N G 3 PROOF R ING 4
20 20 --,
c

I
-.-
-o I I
c

15 15
--.-

Vl

10 10
::::>
Q)

Vl � �

5 5
:z :z

UJ UJ
<->

0
u

0
er er
0 0
..... .....

L_ _L ._J I I I I I I I

220 220
TIME ( s I T I ME ( s J
 lal RAW TOTAL FORCE DATA
9 0 1 - --,- ..,- ,-- ,- - -, .. -, ...r

l
- - - - - -- - .
,
-
,

- 70 �I �
�
c

r
"'
....
50
E
'

� �
:z:

30
�

'

----- - � EVENT THRESHOLD

UJ

1 0 �--
<...)
"'
0
Ll- ____ __ _ __ _ _

-10 l . ...L . .. L. --L _l_ _ _ _ L_ _ L_ _L _ ___L__ _ _t_ __ _j

735 74 0 7 45
TIME (s J

l b ) F I LTERED T O T A L F O R C E DATA
90 r - -
.
- --
-,-
-- .
---
.- - -,-- --. - - - -
,- -
-,
- - - -.--
-
-l
i '

-
70
� �
�
c

�
"'
<-
50
E
' i

�
�
:z:

30
�

. -----J EVENT THRESHOLD

UJ

f=-- - --- - -
w
""
0
10
I
Ll- -----·- -·-

-1 0 _ __L _ .. _L _ _ _!_ _ _ _L _ _L _ _L_ ....L .

_ _L __ __L__
__ _ j
7 35 740 7 45
TIME Csl

Fig 32 E x a m p le o f d e r i v e d t o t a l f o r c e t im e s e r ie s
 70

50

30
uJ
(..)

10
"'
0
u... �-�---�--;EVENT THRESHOLD
- 1 0 L---L--_i___.JL __l _i_
1 _
I � I I j
1 65 . 5 166 . 0 166 .5
_

Tl ME ! sl

!bl F I L T E R E D T O T A L F O R C E D A T A

�
� � EVENT THRESHOLD
__L_ __L_

1 65 . 5 166 . 0 166 . 5
TIME !sl

Fig 33 E x a m p le o f m u ltiple t h r e s h o l d c r o s s i n g w i th r a w d a t a
 -- ------�--�
,---
�
�

E
11)
0
-
11
Ul 0 - ;:::
_c
Ul 0::
� w
:;:
lD m
::::>
�

11 z
E z
f--
0
E i=
('j - 0

1- 0
� w
('j 0 (f)
0 f--
�

11 (f)
Ul w
I f--

�
f--
(f)
w
f--

1-
-
0 OCl

I I I I I I 'i'
11) 11) � 11) r<) 11) ('j 11)
..t r<) ('j

W_L/dW!J_ OI.L'v'� .L8'v'd V'J I

Fig 3 4 Influence of s ection geometry o n impact ratio

 i
I
I �
0
��

E
lfl
0
11 �

"' 0
_c
"'
�

0:::
w
r--
::::;:
r-- [D

::l
�

11 z
E z
I-
0
E r=
lD u
�
�
w
N 0 (/)
0 �
0 I-

"'
11 (/)
w
I I-

[D
I-
(/)
w
I-
�
0 00

I
I
'

lfl lfl lfl lfl N

�
Lf!
�
�

"<t

-t I') N �

W .L/dWLJ_ 0 1.1\fd .lJ\fd VII I

Fig 35 Influence of s e ction g e ometry on impact ratio

 0 1- 0J �

E
I!)
0
11 -

0 f--
0::
.c
(/J �

w
�

'<t
(/J

(0 £D
:::;:
�
::::>
11 z
E z
f--
0
E f--
0J u
-
w
0J (/)
0 f-- �
�

0 f--
11 (/)
(/J w
f--
I

<i.
f--
(/)
w
f--
-

0 f-- CO

I I I I I
0 0 0 0 0 0 0
0 0 0 0 0 0
lO I!) "'<t 10 0J

( u nJ W/N) :3 :J � O..::l X 9 %<;; 'v'.L'v'O >N:3 d 0:3�:3rll..::l

fig 3 6 I n f lu e n c e o f s e ct i o n g e o m e t r y o n f i l t e r e d 5 % e x c e ed e n c e
force
 0

E
Li)
0
11 -
fJ) 0
..c 1- �

fJ) n::
I' w
I' Ill
::2'
:::>
11 z
E
1- z
0
E 1-
lD 0
�
-
w
('J 0 VJ
0 r-
0 1-
�

11 VJ
fJ) w
I 1-

Ill
1-
VJ
w
1-

0 r- CO

I I I I I
0 0 0 0 0 0 0
"<!-
0 0 0 0 0 0
lD Li) 1'1 ('J

(un.J W/N) 30CJO..::I X a % £ V.LVO >N3d 0 3Cl3.LII..::I

Fig 3 7 I n f lu e n c e o f s e c t i o n g e o m e t ry on f i lt e r e d 5 % e x c e e d e n c e
force
 0

E
t{)
0
-
11
m 0 - �

.r:.
m 0::
w
..,. rn
lO
2
�
:J
11 z
E z
1-
0
E 1-
0
N - w
- 0 (()
�
N 0
1-
�
0
(()
11 w
m 1-
I
;j_
1-
(()
w
1-
-
0 - 00

�-----,-,-
-,-l£r--,-l----.-l--,l---.l---,l.---l.-- ---+- ..,.
0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0
en 00 � lO t{) '<t � N �

(un� W /N) 3 8 � 0..:l >IV'3d 0 3 � 31.l l..:l

Fig 3 8 I n f l u e n c e o f s e c t i o n g e o m e t r y o n f i lt e r e d m a x i m u m
force
 0

E
t[)
-
0
11
Ul 0 �

.c �

Ul lY
1'- w
:;;:
1'- m
�
::J
11 z
E z
1-
0

-
E 1-
lD u
w
N 0 0 VJ
�
0 1-
11 VJ
Ul w
I 1-

m
1-
VJ
w
1-
-
0 00

0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0
en 00 1'- lD t[) "<!- I") N

( u nJ W / N) 3 8 t:I O..:l >l'i3 d 0 3 tl 3.Ll l..:l

Fig 39 I n f lu e n c e o f s e c t i o n g e o m e t r y o n f i lt e r e d m a x im u m
force
=
--------
..,
SECTI O N 8f: Const H s = 0 . 1 6 m (nom) hs=0.5m

=
J
..,..
240

230

220
I
' �
c
:::>
_..,

f1)
210
:::>
n
f1)
"
c 200
:J
0 I..
_.., 1 90

'-...
E
3
-;

D
1 80
z I
0
:::> 1 70
'-'

u
w

::;:
_..,
1 60
0
a:::
f1)

l1..
Cl.,
..., 1 50
"'

�
3
lS
1 40

0.. 1 D
Ql
1 30
3 0
X

c: w
3 a:::
1 20

w 1 10
0
_..,

�
iL
...,
1 00
0
n
"'

90
I
0
n

:::>
80
VI
....
Ql 70

I
0
--r-
:::>
....
60
::c
VI
1 .3 1 .5 1 .7 1 .9

I NSHORE M EAN ZERO-CROSS I N G PERIOD Tm(s)

 SECTION Sf: Const H s =0 . 1 6 m (nom) h s = 0 . 5 m
:-n

0
= 6.4
�
� 6.2 J 0

6
:::>
�

c
5.8
r1>
:::>
n 5.6
r1>

0 5.4
�
E
"
-1 1-
5.2
3 a.

:::>
CJ
E 5 --i 0
i=
3 0 4 .8

�n::
LJ

0
"'
n
4 .6
_,..
I
u
1-
-,
4.4
<(
"'

D..
_,..

2
4.2
-
n 4

3.3 6
0
::J
Vl
_,..
3.8
"'
::J
_,..

3
:::r:
Vl .4
I

-: L1 .3
-r-
1 .5 1 .7
�
1 .9

I NSHORE M EAN ZERO - CROSSING PERIOD Tm(s)

 .,
u:::J 220 r -
SECTION 8f: Const Tm = 1 . 6 s(nom) hs=0.5m

+- 210 0
r-..J '
200

::J 1 90
�
1 80
rn
c:

.......
::J c 1 70
rtl :J
n

L 1 60
0

"
�
E 1 50
z
:::c
1 40
V)
..__,
0
w 0
0
1 30 1
0:::
::J

�
0
1 20
...,.
rn u.. 1 10
...,
rtl ::.:: 1 00
l50..
Cl..

3
90
"'
0
80
3 w
X

0:::
3 w
c:
70
60
�
iL
0
�
50
4o
...,

0
rtl
1
n

30
0
n
20
::J
"'
...,. 10
"'
::J 0
...,.

3
--I 0.08 0.1 0. 1 2 0. 1 4 0. 1 6 0. 1 8 0.2 0.22

I NS H O R E S I G N I FICANT WAVE H EIGHT Hs(m)

-,.

<0 SECTI O N 8f: C on st s = 0.04(nom) h s = 0 . 5 m

+- 260
+-

240 D
::J
-n

c
rt>
220
::J
n
,......
c
rt>
200
0 :J
-n L

3
-I 1 80
"
E
0 z

w
::J .._,
1 60

-..- u
-n

0:: 1 40
D
rt>

rt>
-, 0
lL
"'-
::.:: 1 20

6(]_
3
Q)
X
1 00
0
3
3 w
c

w
0:: 80
-n
0

�
[;:
-,

D
n
rt> 60

40
0
n

::J
-..-
VI
Q)
20
-..-
::J

VI 0 1
�----.- -,--- I r---· �-- -- -r-- -- I

1 .2 1 .4 1 .6 1 .8

I NSHORE M EAN ZERO - CROSS I N G PERIOD Tm(s)

 CO

m
�

'-"
E w
Ll) E
1--
0 0
11 0
m
_c 0 et:
�
w
()_
E
0 (.')
c z
'-"
--;j- (/)
0 (/)
0 0
et:
11
I
VI
u
VI
_,_,
0
c et:
0 w
u N
z
L5
.._
CO
::::;;
0
z
0 w
et:
w
t; 0
(/)
:r:
(/)
Ol
z
c
::<: 8

�t
-- a:

W_i/d WLL O l.lV� .l8Vd N I

Fig 45 I n f lu e n c e o f Tm o n i m p a c t r a t i o - c o n s t a n t s
 0 \ 0
'<1-
N

\ 0
0
0
N

\
\

\'
0
\
\
ID -o
�
Q)

\ :::J
L..

E m
ll1 0
Q)
c)
,-...
11 E
0
m
.r:: c
0 :::J
.;_: N L..
ro

"-
�
E
z
0 z
f- ..___,
0
0
w X
0
(f)
E
I
0 lL
ro

0
0
'<1-

0
0 0 0 0 0 0 0 0 0 0 0 0
ID '<I" 0 aJ ID '<1- N 0 ro ID N
N N N

Fig 4 6 C o m p a ri s o n b e t w e e n m e a s u r e d f o r c e s a n d t h e s e p r e d i c t e d
 =
.,

SECTI O N 8f: hs=0.5m Ac =0. 1 m hf=0 . 1 1 m

..,_
_,
0 .0 6

'-- 3:
ro 0
:::>
'"

VI QJ
VI 0.05
c:
:::>
,

ro
...,

CL
0
0
......

J e n s e n's r e c o m m e n d e d
n
...,

ro 0.04 line ( s e e F i g 4 9) -------..J

VI
0
"0
ro c.
...,

�
--'
ro
VI
0
:::>
.....
B e s t f i t line t o
__,.. ..c
ro
0.03 d a t a f o r s e c t io n 8 f
...9-
=
CL
0
.......
:::>
X
"' 0
::r E
__,..

ro :c
u._ 0.02
0
......

...,

3
"0

0
...,

"0 0.0 1 0
0
VI
"'
CL

o-
'<

0.8 1 1 .2 1 .4 1 .6 1 .8 2 2.2

Hs / A c
 " SECTI O N 8f: hs =0 . 5 m A c = 0 . 1 m h f = 0 . 1 1 m
u:J 260
+­ 0
CO /
240
/
Predicted /
measured /
_, " -
0 0 220
1.3 x
� 3
"'
,.,
QJ
-o "/
Vi
..., 200 /
c

::;• ::J
"' 0 / --..... p red id e d measured
=
1 80 /
0
'- er
rt> rt>
/ /
/
/
;:, .... 1 60 / /
"'
rt>
::J
�
rt>
rt>
/ / o /
/
/

.,- ::J "0

QJ
1 40 / /
cr 3 /
/
....
u

rt> rt>
0
/
/
"0
·� QJ 1 20 /
..9- �Predicted
QJ
-.- V> �
c /
::;: m
_, ...,
X
1 00
/
/ o
/
/ 0.1 m easured
x

/
0...
E
"'
0
:c /
u.. /
-- _,

/
::J 0
ro ..., 80 /
n
/
/
rt>
·� /
60 / /
/
QJ
::J
/
Cl.
,.0
40 / /
/
'0
..., /
rt> /
0...
20 / /
0-/
,1- /
n

ro
0...
0

0 40 80 1 20 1 60 200 240

FHm ax (m easured )
 Yh LP
0 .1 :2

0.11

/
0 . 10

o.o�

0.06
./
0.07 �
0.00
/.
0.0 �
� V:
/
�u
/
0.04
4
�
0.03
/
•

0 . 0 :2
n£i
IV

0.01 �/" Hs

0
0
/
0.� 1.0 1 .5 2.0 :2.5
6h

L EG E N D '

WAT ER LEVEL P E A K WAVE PER I O D

(m> Tp ( s )

14 HI 16

• 5.3 • • •

• 2.3 .. 4 [J

• 0. 3 0 6 0

ft.,O 2 1.0
..£
h ·16.7

35

QUARRY STONES 6•9t

Fig 4 9 F o r c e d a t a p r e s e n t e d b y J e n s e n (R e f 28)
 I

[1 1 ! t J

Fig so S p a t ia l dist ribution o f m a ximum w a v e p r e s s u r e o n C r o ,w n wall - a f t e r

Jensen
PLATE.
Plate 1 Force table showing proof ring details
APPE N D I X.
APPENDIX A

Analys i s of crown wal l force data

A least squares regre s s ion a n a l y s i s was appl ied to the

resul ts for s e c t ion 8f to appro x imate the mea s ur e d
ma x imum wave f o r c e s b y t h e equa t ion :

= (Al)

Where A , B and C are d imens ional c o n s t an t s . The forc e s

pred icted from the der ived rel a t io n sh i p are p l o t t e d
against the respec t iv e measured forces in Figure 4 6 .
Clear l y , the r e l a t io n s h i p f i t s we l l , espec i a l l y for
the l a rger forc e s . The mean magn i t ude o f d i s c repancy
between measured and pre d i c t e d forces for sec t io n 8 f
was l l . SN /m run . For s e c t ion 8 f at water depth
h = O . Sm , the der iv e d coe f f i c ients for equa t ion Al
5
are :

A 1 0 6 1 N /m2
B = 3 1 8 N / s /m
C -546 N / m

I t should be n o t e d that t h e c h a r a c t er i s t i c l in e a r

�
r e l a t ionships between forc e , F , and wave height , H 5 ,
and between F and wave p e r i o d T , shown in F i g u r e s
m m
40 and 4 2 a r e only v a l id a t m o d e l sca l e . The
l inear i t y w i l l become d i s to r t e d when the d a t a i s
sca led t o prototype t e rms .

It has not been p o s s i b l e to d e r ive a s a t i s fac tory

d imen s ionl e s s rel a t io n sh i p from the data . Howev e r ,
the model r e s u l t s c o u l d be a p p l ied to a proto type
problem of s im i l a r geome t r ic configura t io n using
Froude model s c a l ing l aws . Hence , if d is a
p
character i s t ic pro totype d imen s ion such a s c r own wa l l
height or water depth , and d i s the correspond ing
m
d imension from te s t s e c t ion 8 £ , then we can d e f i n e
A = d /d .
p m

The prototype max imum hor i zo n t a l force 1n Newtons p e r

metre run may then be e s t imated from :

F (A2)
Hmax (Proto type )

Where H and T are p r o t o t ype wave par ame ters and A , B

s m
and C are the c o e f f i c i e n t v a l u e s tabul a ted abov e .
Where pre d i c ted f o r c e s are l e s s than zero , inc iden t
waves would no t b e exp e c t e d to reach the crown wal l .
The above procedure may t en t a t iv e l y be extrapola ted t o
other armour s l ope c o n f i gu r a t i o n s for wh i c h for c e
measurements were m ad e , b y mul t ip l y ing the re s u l t i n g
force from equ a t i o n A2 by the appropr i a te value o f
a rmour coe f f ic ie n t F
ijma x / FHma x S f from T a b l e 4 . It
should be borne i n m1nd t h a t t h e se factors a r e each
only s t r i c t l y v a l i d fo r a s in g l e random wave
cond i t ion .

Further work i s r e q u i r ed to s ub s tan t i a te the formu l a e

proposed above and t o ex amine ful l y t h e ir range o f
validity, At pre s en t , only a s ingle wa ter d e pth h a s
been inv e s t i g a t ed . Fo r s a ti s fac tory use o f the
proposed f o rmu l a , it i s n ec e s s ary to inc lude a t e rm

r igorously the s c a l e fac tor X in e q u a t ion A 2 .

for s t ructural freeboard , and a l s o to d e f ine more
Th i s
h a s no t been po s s ib l e w i t h i n the scope o f the pre s e n t
s tudy .