Coastal Engineering 87 (2014) 112121

Contents lists available at ScienceDirect

Coastal Engineering
journal homepage: www.elsevier.com/locate/coastaleng

Upgrade of coastal defence structures against increased loadings caused

by climate change: A rst methodological approach
Hans F. Burcharth a, Thomas Lykke Andersen a,, Javier L. Lara b
a
b

Aalborg University, Denmark

Environmental Hydraulics Institute "IH Cantabria", Universidad de Cantabria, C/Isabel Torres n15, Parque Cientico y Tecnologico de Cantabria, 39011 Santander, Spain

a r t i c l e

i n f o

Article history:
Received 30 June 2013
Received in revised form 23 November 2013
Accepted 11 December 2013
Available online 7 January 2014
Keywords:
Sea level rise
Revetments
Upgrading concepts

a b s t r a c t
The paper presents a design exercise of upgrading a typical rock armoured revetment by modifying the structure
prole and adding structure elements. Several concepts of upgrading are examined. A sea level rise corresponding to the mean of the IPCC 2007 predictions is used together with a slight increase in long-term wind/wave
conditions as predicted for the North Sea by the Danish Coastal Authority. Both conditions of non-acceptable
and acceptable increase in structure crest level are considered. Moreover, a scenario for steepening of the foreshore due to morphological changes caused by increased wave impacts is included. Only desk study tools are
used for the upgrade designs. A simple comparative cost optimization analysis of the various upgrading solutions
is presented, and conclusions are given for the preferred upgrading concept valid for the case study structure. A
short discussion of the uncertainties related to upgrading design is included. The importance of physical model
tests of the structures is underlined due to insufcient desk study tools for rubble mound upgrade design.
2013 Elsevier B.V. All rights reserved.

1. Introduction
Climate change might cause sea level rise and increase in the intensity of storms. Both phenomena will increase the risk of ooding
of low lying areas, accelerate erosion of exposed soft beaches, and
cause damage to existing coastal protection structures. This makes
it necessary to upgrade the structures so they comply with the
original design performance criteria. Upgrading can be done by
modifying the structure prole and/or adding structure elements.
The 2007 report of the Intergovernmental Panel on Climate Change
(IPCC) presents six scenarios with estimated sea level rises in the
range 0.180.59 m, i.e. a mean value of 0.35 m, by the end of the 21st
century.
Figures for increase in storm intensity are not given by IPCC but are
dealt with regionally. As an example a 2% and a 5% increase in signicant
wave height are predicted for the North Sea for years 2050 and 2100
respectively, STOWASUS (2001). This is due to foreseen higher average
wind velocities. However, related to storm wind velocities a 10%
increase, which will add approximately 0.3 m to the wind generated
set-up on a at coast, is expected. On sandy coasts the higher water

Corresponding author at: Aalborg University, Dept. of Civil Eng., Sohngaardsholmsvej

57, Dk-9000 Aalborg, Denmark. Tel.: +45 99408486.
E-mail address: tla@civil.aau.dk (T. Lykke Andersen).
0378-3839/$ see front matter 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.coastaleng.2013.12.006

levels and the larger waves generally cause steepening of the coastal
prole and subsequent accelerated retreat of the coastline.
The paper presents a design exercise of the upgrade of one of the
most commonly used types of coastal protection structures, namely revetments with a sloping seaward front armoured with randomly placed
quarry rocks. The structures are typically used in more shallow waters
with depth limited design waves. The seabed can be both erodible and
resistant.
The paper presents several concepts of upgrading for conditions of
non-acceptable and acceptable increase in structure crest level. Specic
upgrade design of a shallow water conventional revetment is performed
assuming a sea level rise corresponding to the mean of the predictions
by IPCC. The performance of the upgraded structure is assumed unchanged compared to the existing structure.
Desk study tools for the design of conventional revetment and
breakwater structures are readily available but are not covering all
performance aspects related to upgrading. An example is the effects of
adding an extra armour layer. In the present design exercises are
therefore used modications of existing formulae. This introduces
extra uncertainty for which reason it is stressed that performance of
physical model tests is a necessity for the nal evaluation of the proposed upgradings. Numerical models are not applied in the present
paper because the available models, e.g. for armour stability, have not
yet been sufciently calibrated to cover the presented concepts of
upgrading.
A simple comparative cost optimization analysis of the various
upgrading solutions is presented. Finally safety aspects related to

H.F. Burcharth et al. / Coastal Engineering 87 (2014) 112121

113

Fig. 1. Concepts of upgrading in which an increase in crest level is acceptable.

climate change motivated upgrading are discussed including the potential of application of Levels 1, 2 and 3 design procedures.

strongly interlocking complex types of armour units. In fact, complete

replacement of the armour units might be a necessity.

2. Concepts of upgrading

3. Procedure and steps in upgrading-design of rubble

mound structures

Increasing the crest level of the structure is an obvious measure for

counteracting the increase in wave run-up and overtopping. However,
in many places with promenades and restaurants this is not acceptable
as the sea view will be blocked or reduced.
Fig. 1 illustrates concepts of upgrading in which an increase in crest
level is acceptable. Strengthening of the main armour and reduction in
overtopping are obtained by placing an extra layer of armour units on
the front slope and on the crest. The related increase in pore volume
improves the armour stability and reduces run-up and overtopping.
Concepts in which a concrete superstructure is added or an existing
superstructure is heightened and strengthened are also shown.
Fig. 2 illustrates concepts with no change in crest levels. Only armour
units on the front are added in order to form a atter slope or to form a
berm. A atter slope and a berm reduce the overtopping and increase
armour stability. Solutions with a separate breakwater, a front reservoir
or an articial reef, all of which have the effect of reducing the wave
impact on the existing structure, are also shown.
It should be noted that in order to ensure good connection, the
adding of armour layers should preferably be done with the same type
and size of armour units as in the existing structure. Even so, placement
which ensures good interlocking can be very difcult to obtain for

The procedure in upgrading design of a structure can be divided in

the following steps:
1. Examination of the existing structure with respect to degradation of
the structure elements and possible continuous use as parts of the
upgraded structure.
2. Denition of the service lifetime of the upgraded structure.
3. Denition of geometrical and esthetical restrictions/limitations for
the upgraded structure, as for example increase in crest level not
allowed, concrete armour units not allowed, etc.
4. Denition of the performance criteria for the upgraded structure related to Serviceability Limit State (SLS) and Ultimate Limit State
(ULS). Repairable Limit State (RLS) might be included as well
a. Overtopping discharges and related exceedance probabilities
b. Stability of structure elements. Types of damage, damage levels
and related exceedance probabilities
I Toe, front, crest and rear side (if relevant) armour displacements
II Crown wall breakage, sliding and geotechnical slip failures
III Geotechnical overall stability and settlements

Fig. 2. Concepts of upgrading in which an increase in crest level is not acceptable.

114

H.F. Burcharth et al. / Coastal Engineering 87 (2014) 112121

Fig. 3. Cross section of existing structure with a buried toe.

5. Denition of the climate change scenario over the structure service

lifetime in terms of rise in water level and increase in storm intensity offshore. Construction of the related long-term wave statistics
(at some distance from the coast in deeper water).
6. Transformation of the waves to the near shore zone.
7. Investigation of morphological changes in terms of changes in
coastal proles.
8. Establishment of the combined long-term statistics of wave heights,
wave periods and water levels in front of the structure. Subsequent
extraction of the conditions corresponding to various return periods
and related encounter probabilities within the structure service lifetime including the design exceedance probabilities dened in 4a
and 4b.
9. Identication of the weaknesses in the performance of the existing
structure with respect to stability and overtopping if exposed to the
new scenario sea states identied in 8. Subsequent selection of potential alternative concepts of upgrading among the solutions given
in Figs. 1 and 2.
10. Design the upgraded structure for each alternative by use of formulae, neural network and computational models.
11. Estimation of the upgrading costs for each alternative structure.
12. Denition of a repair policy in terms of repair actions corresponding
to SLS and ULS (and maybe RLS). Denition of unit costs of repairs.
13. Desk study simulation of the structural and hydraulic performances,
and identication of the occurrences of repair thresholds by using
a synthetic lifetime wave-water level exposure time series which
represents the long-term wave climate in front of the structure.
14. Calculation of the lifetime cost (construction plus repair costs) of
each alternative as basis for the selection of the preferred one.
Checking and optimizing of the preferred structure in physical
model tests.

example, is a conventional high berm. This difference inuences the

upgrading.
4.1. Existing structure with buried toe
4.1.1. Design criteria and performance of existing structure and performance criteria for upgraded structures
Fig. 3 shows the sections of an existing structure with a buried toe
situated on a at prole sandy coast. The armour consists of two layers
of randomly placed quarried rock with mass density 2.65 t/m3. The tidal
range is small, and for this reason the storm surge, consisting of wave
and wind generated set-up and atmospheric low pressure effect, dominates the high water levels. The slope of the foreshore seabed is 1:100.
The structure is designed for 50 year service lifetime using a deterministic design approach and readily available formula for structure performances. For the estimation of overtopping the CLASH
EU-project Neural Network (Van Gent et al., 2007) is used.
The design performance criteria are as follows:
Armour stability corresponding to initiation of armour damage
(S = 23 in the Van der Meer, 1988 formula, KD = 2 in the Hudson
formula) for 50 years return period sea state, and repairable damage
for the 100200 years return period sea state (S b 45 in the Van
der Meer formula, KD b 3 in the Hudson formula). The number of
waves in the sea states is set to Nz = 2000.
Toe stability corresponding for the 200 year return period sea
state to a small acceptable damage (Nod b 1.5) for the most critical water levels creating wave breaking directly onto the toe. For
the estimation of the stability of the buried toe the diagram Fig. 2
in Baart et al. (2010) extrapolated to conditions of a buried toe is
used. The toe stability formulae by Gerding (1993), Van der Meer
et al. (1995), Ebbens (2009) and Muttray (2013) are for high toe
berms consisting of two layers of stones, and therefore not
applicable.
Admissible average overtopping discharges are q b 0.1 l/sm for the
1-year return period sea state, q b 5 l/sm for the 10 years return
period sea state, q b 10 l/sm for the 50 years sea state, q b 25 l/sm
for the 100 years return period sea state, and q b 40 l/sm for the

4. Example application
A simplied desk study of the upgrading of two rock armoured rubble mound revetments is presented. The only difference between the
two cross sections is the toe structure which, in one example, is buried
and therefore has the crest in level with the seabed and, in the other

Table 1
Design hydraulic conditions and related performance for existing revetment.
Return period (years)

10

50

100

200

Hso in deeper water (m)

Tp (s)
WL incl. storm surge (m) relative to MWL
Water depth in front of structure (m)
H1/3 in front of structure (m)
Armour damage parameter S (Van der Meer, 1988)
Armour damage parameter KD (Hudson formula)
Toe damage, Nod (Baart et al., 2010.Fig. 2, slope 1:50)
Average overtopping discharge (l/sm)

2.7
710
+1.0
4.0
2.45
0.4
0.9
0.0006
0.04

3.1
811
+1.3
4.3
2.73
1.2
1.2
0.001
0.3

3.5
912
+1.8
4.8
3.01
2.0
1.6
0.002
2

3.8
913
+2.1
5.1
3.27
3.0
2.1
0.004
7

4.0
1014
+2.3
5.3
3.44
3.9
2.4
0.005
17

4.1
1315
+2.4
5.4
3.53
4.4
2.6
0.006
29

H.F. Burcharth et al. / Coastal Engineering 87 (2014) 112121

115

Table 2
Design hydraulic conditions for the upgraded structure.
Return period (years)

10

50

100

200

Hso in deeper water (m)

Tp (s)
WL incl. storm surge (m) relative to MWL
Water depth in front of structure (m)
H1/3 in front of structure (m), sea bed slope 1:100
H1/3 in front of structure (m), sea bed slope 1: 30

2.84
711
+1.25
4.25
2.67
2.93

3.26
812
+1.60
4.60
2.93
3.23

3.68
913
+2.15
5.15
3.28
3.68

3.99
914
+2.50
5.50
3.47
3.83

4.20
1015
+2.75
5.75
3.70
4.12

4.31
1316
+2.87
5.87
3.83
4.21

200 years sea state. The actual structure performs somewhat better,
cf. the overtopping values given in Table 1. However, the admissible
discharges are used for the upgraded structures. We will for simplicity only consider the average discharge in this paper, but updated
tools for maximum overtopping volumes were developed in the
Theseus project, cf. Nrgaard et al. (2014).
The design hydraulic conditions and the related structure performance for the existing structure are given in Table 1. At the time of
the design the effects of climate change were not foreseen and are
therefore not included in wave heights and water levels.
The S-values correspond to the most critical Iribarren number for
which the armour stability is minimum, i.e. the transition between
plunging and surging waves. The average overtopping discharges are
determined by the EU-project CLASH Neural Network using the upper
values of the range of wave periods.

4.1.2. Environmental conditions for upgrading

The service lifetime of the upgraded structure is assumed to be
50 years from the day of upgrading. The service lifetime could also be
100 year, but in most cases surroundings and related requests for performance conditions are changed long time before.
On the basis of the IPCC predictions a conservative estimate of
0.20 m sea level rise over the next 50 years has been assumed. Over
the same period a 5% increase in deep water signicant wave height
and wave peak period, and 0.05 m, 0.10 m, 0.15 m, 0.20 m, 0.25 m
and 0.27 m increase in storm surge level at the coastline for the 10,
50, 100 and 200 years return periods respectively have been assumed.
Thus, over the next 50 years the total increases in design water levels
in front of the structure for the 1, 5, 10, 50, 100 and 200 years return
period sea states are 0.25 m, 0.30 m, 0.35 m, 0.40 m, 0.45 m and
0.47 m, respectively. More precise estimates of climate changes
might be found by considering a specic site, cf. Weisse et al.
(2014in this issue).
Two scenarios for change in coastal prole are considered. One in
which no signicant change in near-structure sea bed prole occurs,
i.e. the 1:100 foreshore seabed slope is practically maintained. Another
in which a steepening of the prole to slope 1:30 due to increased impacts is assumed. This rather drastic change in seabed slope is included

in order to explore the sensitivity of upgrading to morphological

changes.
The structure performance criteria for the upgraded structure are assumed to be the same as given in 4.1 for the existing structure.
The design hydraulic conditions for the upgraded structure are given
in Table 2.
By comparison with Table 1 it can be seen that the increase in incident wave height is signicantly larger for the steepened seabed prole
of slope 1:30 than for the unchanged seabed prole of slope 1:100. This
underlines the importance of making a thorough analysis of expected
morphological changes. Both cases of seabed slopes are presented in
the following analyses.
4.1.3. Upgrading of revetment with buried toe by adding armour layer and
increasing wave wall height
The rst upgrading concept studied is simply adding an extra armour layer on the front slope and extending the toe as shown in Fig. 4.
The effect of the extra armour layer on armour stability is taken
into account by increasing the notional permeability coefcient in
the Van der Meer rock armour 1988 formula from P = 0.4, valid for
two-layer rock armour on a thick lter layer to P = 0.6, valid for a
homogeneous structure. As this value might be too high, also a
value of P = 0.55, corresponding to two-layer rock armour placed
directly on coarse core material, is investigated. The correct value
might actually be in between 0.55 and 0.6. Only model tests can verify this estimate.
The effect of the extra armour layer on overtopping is taken into
account by changing the roughness factor in the NN from
f = 0.55, valid for double layer rock armour, to f = 0.4, valid for
Icelandic berm breakwaters and double layers of cubes, Accropodes
and Dolosse. This is an approximation in lack of more specic
information.
It turned out that the upgrading shown in Fig. 4 both satised the
stability and the overtopping performance conditions, but only if the
sea bed slope remains 1:100. If, due to morphological changes, the sea
bed slope steepens to 1:30, then the armour stability might be just
acceptable, but the overtopping will be too large. By increasing the
crest level of the wave wall to 1.00 m above the armour crest level as
shown in Fig. 5, it is possible to satisfy the criteria for admissible
overtopping. Alternatively an extra armour layer can be placed on the

Fig. 4. Upgrading by adding an extra armour layer. Foreshore seabed slope 1:00.

116

H.F. Burcharth et al. / Coastal Engineering 87 (2014) 112121

Fig. 5. Upgrading by adding an extra armour layer and a wave wall extending 1.00 m over the armour crest level. Foreshore seabed slope 1:30.

Fig. 6. Upgrading by adding an extra armour layer on the front slope and on the crest plus increase in existing wall height by 1.00 m to crest level + 7.50 m. Foreshore seabed slope 1:30.

crest as shown in Fig. 6, but still the wave wall has to be raised by
1.00 m. The overtopping and the wave forces on the wall will be smaller
than for the upgrade shown in Fig. 5.
The performances of the upgraded structures are given in Table 3.
From Table 3 it is seen that the upgraded structures shown in Figs. 4,
5 and 6 rather closely satisfy the performance criteria dened in Table 1
for the existing structure. The exception is the armour damage in case of
a foreshore slope of 1:30 and application of P = 0.55. Only model tests
can verify the reduction in armour stability.
If an increase of the structure crest level is not acceptable, then the
most efcient solutions would probably be to add a high berm or to
place a low crested submerged breakwater on the foreshore cf. Fig. 2,
b) and c).
The 0.1% exceedance values of the 200 years return period wave induced forces and tilting moments on the walls, shown in Figs. 5 and 6,
are estimated by the upgraded Pedersen formula (Nrgaard et al.,
2014). The formula predicts the total horizontal force on wall front, FH,
the tilting moment around base front corner, MFH, and the pressure at
the base front, pb. The formula is valid only for Iribarren numbers

Table 3
Performance of upgraded structures with buried toe.
Return period (years)

Armour damage parameter S for foreshore

slope 1:100
P = 0.60
0.5
P = 0.55
0.6

10

50

100

200

0.8 1.4
1.0 1.7

1.9
2.3

2.6
3.1

3.1
3.7

3.1
3.7
0.005

4.4
5.3
0.008

5.0
6.0
0.010

Armour damage parameter S for foreshore

slope 1:30
P = 0.60
0.8
1.3 2.6
P = 0.55
1.0
1.6 3.0
0.001
0.004
Toe damage Nod (Baart et al., 2010,
Fig. 2),
slope 1:50
Slope 1:20
0.027
0.12
Average overtopping discharge (l/sm)
Fig. 4. Foreshore slope 1:100
Fig. 5. Foreshore slope 1:30
Fig. 6. Foreshore slope 1:30

0.16

0.26

0.30

below 4.5. For larger values the limitation Ru, 0.1%/H1/3 b 2.6 is used in
the calculations. The wave induced loadings are given in Table 4.
For the design of the wave wall superstructures in the upgraded
structures a target safety factor of 1.2 against sliding and overturning
for the 200 years return period sea state is assumed. In the stability calculations the following parameter values are used:
Friction coefcient for concrete slab against quarried subsoil, = 0.6
Mass density of concrete, c = 2300 kg/m3
Effective weight of soil material, = 11,000 N/m3
Passive earth pressure coefcient for quarried materials, rough wall
(Brinch Hansen), KY = 10
The 0.1% exceedance values of FH and pb do not occur simultaneously. In the stability calculations pb is reduced by 20% compared to the
Table 4 values.
The dimensions of the upgraded wall shown in Figs. 5 and 6 correspond to Tp = 13 s and to a safety coefcient of 1.2 against sliding
and overturning. Walls designed for Tp = 16 s will have approximately
25% larger volumes.

4.1.4. Upgrading of revetment with buried toe by adding a high berm

The increases in crest levels, shown in Figs. 5 and 6, can be avoided
by adding a berm on the front slope. The optimum dimensions of a
berm which satises the restrictions on overtopping given in Table 1
have been identied from simulations by means of the EU CLASH
Overtopping Neural Network.

Table 4
Estimated 0.1% wave loadings on wave walls corresponding to 200 years return period.
Figure

Tp (s)

FH (kN/m)

MFH (kNm/m)

pb (kN/m2)

13
16
13
16
13
16

167
205
119
147
135
167

263
324
170
209
218
269

73.6
73.6
59.5
59.5
59.5
59.6

6
0.07
0.08
0.05

2.2
3.2
2.8

6.9
8.7
7.9

19
25
23

34
39
36

11

H.F. Burcharth et al. / Coastal Engineering 87 (2014) 112121

117

Fig. 7. Upgrading by adding an extra armour layer and a berm in order to avoid increase in structure crest level. Foreshore seabed slope 1:30.

It is assumed that the berm has to be made of the same size of stones
as used in the armour layers in order to obtain simplicity in construction. Two layers are applied for the berm.
The cross section is shown in Fig. 7.
The overtopping performance of the structure is given in Table 5.
It should be noted that according to Van Gent et al. (2013) it is possible to reduce the size of the armour on the slope above the berm. However, if doing so, the armour crest width of 7.5 m will be slightly reduced
and overtopping slightly increased. A related optimization has not been
performed. Neither, for reasons of missing readily available formulae,
has the optimization of the berm stones with respect to stability been
studied.
4.1.5. Upgrading of revetment by placing a submerged breakwater on
the foreshore
Fig. 2 c) shows the concept of using a detached breakwater to reduce the impact on the revetment by triggering wave breaking over
the breakwater. The 200 years return period sea state, which for the
1:100 slope foreshore is given by the incident signicant wave
height Hs,i = 3.83 m, the wave period T m 1,0 = 14.5 s, and the
SWL = + 2.87 m, is used for the design of the breakwater. If gaps
in the breakwater are not rather closely spaced, the wave breaking
over the breakwater causes a pile-up of the water, which is taken
as 0.1 Hs,i = 0.38 m, estimated from Fig. 13.31 in Burcharth et al.
(2007). This gives a SWL = + 3.25 m and a water depth of 6.25 in
front of the revetment. The energy on longer frequencies in the
wave spectrum of the incident waves are somewhat reduced by
the wave transformation over the breakwater, while relatively more energy will be present at higher frequencies. However,Tm 1,0 = 14.5 s is
on the safe side assumed maintained for the transmitted waves as
overtopping is almost entirely dependent on the lower frequencies.
The transmitted signicant wave height corresponding to the admissible overtopping of 40 l/sm, cf. 4.1.1 is determined by use of the EU
CLASH Overtopping NN to be Hs,t = 3.00 m. The related wave transmission coefcient is Ct = 3.00/3.83 = 0.78.
The needed freeboard of the detached breakwater is determined
by use of the wave transmission formula by Briganti et al. (2003) to
be Rc = 2.10 m, which corresponds to a crest level of + 2.87
2.10 = + 0.77 m + 0.80 m.
The size of the rock main armour is determined by the design formula for low-crested breakwaters in shallow water by Kramer and
Burcharth (2003) to be M50 = 5.5 t. For construction reasons the
thickness of the armour layer must correspond to either two or
three layers of stones. Fig. 8 shows both a solution with two layers
placed on a high toe, in which case the needed armour layer thickness corresponds to M50 = 7.5 t, i.e. a conservative size, and a solution with two layers, in which the needed armour layer thickness

Table 5
Overtopping performance for structure upgraded by adding an armour layer and a berm.
Return period (years)

10

50

100

200

Average overtopping discharge (l/sm)

0.08

2.9

7.4

20

33

corresponds to M50 = 5.5 t. The rst solution makes easy construction directly on the seabed possible. The other solution demands
excavation for the buried lter layer and toe, but is a more stable
solution in case of further seabed erosion.
The distance between the breakwater and the revetment would be
approximately 20 m.
4.2. Existing revetment with a high toe
Fig. 9 shows an existing revetment with a toe berm extending 1.2 m
over the seabed. The increase in wave impact makes it necessary to
strengthen the toe in the upgraded structure. Upgrading by adding an
extra armour layer on the front slope and on the toe berm is an obvious
solution, but for structures in shallow water the increase in toe berm
level increases overtopping and reduces the toe armour stability as
will be demonstrated in the following. If instead the toe berm is lowered
by digging the toe into the sea bed, then smaller toe stones would be
sufcient and overtopping will be reduced. However, the construction
will be much more difcult and expensive.
In general, upgrading by adding height to a toe berm is not a very
efcient solution for structures in shallow water as the higher toe demands larger stones, and the overtopping will increase.
The design hydraulic conditions and the performance of the structure are as given in Table 1 for the structure with the buried berm
because the relative small toe berm height of 1.2 m does not change
armour stability and overtopping performances signicantly. Only the
stability of the 0.30.7 t rock toe is less. Actually the shoulder of the
toe berm is rounded in the case of foreshore slope 1:100, and is eroded/attened in the case of foreshore slope 1:30, but the toe berm is
still supporting the main armour.
4.2.1. Upgrading of revetment with high toe by adding armour layer
The performance criteria for the upgraded structure given in 4.1
and the related environmental design condition given in 4.1.1 are
maintained.
As the rst obvious concept for upgrading the addition of an armour
layer on the front slope and on the toe berm is chosen. As no applicable
formula for the inuence of a relatively high berm on the main armour
stability is available, it is assumed that the stability given in Table 3 for
the structure with a buried toe is valid. For the toe berm it would be
preferable to use only one layer of stones in order to keep the height
of the berm as low as possible. However, the stability of such a toe
would have to be determined in model tests because generic formulae
and design diagrams for the stability exist only for toe berms consisting
of two or more layers of stones. In the present desk study two-layer
berms are therefore used.
Calculations by use of the Van der Meer (1998) and the Muttray
(2013) formulae as well as data given in Fig. 2 in Baart et al. (2010)
show that berm stones of 24 t with M50 = 3.0 t, Dn50 = 1.04 m
and berm height 2.08 m will be stable in case of foreshore seabed
slope 1:100, but for foreshore seabed slope 1:30 it is necessary to
use 46 t stones with M50 = 5.0 t, Dn50 = 1.24 m and berm height
2.5 m.

118

H.F. Burcharth et al. / Coastal Engineering 87 (2014) 112121

Fig. 8. a (upper). Fig. 8b (lower). Protection of the revetment by detached breakwater. Foreshore seabed slope 1:00.

Fig. 9. Existing revetment with a high toe berm. Seabed slope 1:100.

Fig. 10. Two solutions for upgrading: One (a) by adding an armour layer on the front slope, the other (b) by adding an armour layer both on the front slope and on the crest.

Overtopping calculations show for the 1:100 slope foreshore that it

will be necessary to add an armour layer on the front slope. In addition,
an armour layer could also be placed on the crest. Fig. 10 shows these
solutions. In the case of a 1:30 foreshore slope it is necessary to place
one layer of armour stones on the crest and increase the wall height to
the level of the armour layer in order to comply with the performance
criteria. Fig. 11 shows this solution.

The performances of the upgraded structures shown in Figs. 10

and 11 are given in Table 6. Regarding armour damage see Table 3.
The toe damage gures given in Table 6 indicate very small damages,
even for the 200 years return period sea state. The prediction by the
various formulae of the toe damage differs somewhat. This is to be
expected for small damages. However, it should be noted that the
model tests on which the formulae are based are performed with

Fig. 11. Upgrading by adding an armour layer on the front slope and on the crest combined with a wave wall raised to armour crest level.

H.F. Burcharth et al. / Coastal Engineering 87 (2014) 112121

119

Table 6
Performance of upgraded structures with high toe berms.
Return period (years)
Toe damage. Fig. 10. Foreshore slope 1:100
Nod (Fit to slope 1:50 data in Fig. 2 in Baart et al., 2010)
Nod (Muttray, 2013)
Toe damage. Fig. 11. Foreshore slope 1:30
Nod (Van der Meer, 1998)
Nod (Fit to slope 1:20 data in Fig. 2 in Baart et al., 2010)
Nod (Muttray, 2013)
Average overtopping discharge (l/sm)
Foreshore slope 1:100, Fig. 10 a (left)
Fig. 10 b (right)
Foreshore slope 1:30, Fig. 11

10

50

100

200

0.03
0.62

0.06
0.92

0.07
1.07

0.06
1.19

0.06
1.28

0.13
0.10
0.56

0.29
0.46
0.85

0.35
0.19
1.0

0.39
0.24
1.1

0.43
0.22
1.2

0.04
0.03
0.03

1.3
1.5
1.8

4.2
5.0
5.2

11
15
16

21
27
26

Fig. 12. Upgrading of structure with high berm by adding an armour layer and a berm in order to avoid increase in structure crest level. Foreshore seabed slope 1:30.

various toe structure geometries some of which deviate somewhat from

the actual toe geometries. This underlines the importance of performing
model tests. The overtopping discharges comply with the target performances dened in 4.1.1.
The solutions for upgrading, shown in Figs. 10 and 11, involve increase
of the structure crest level. If an increase of the structure crest level is not
acceptable, then one of the concepts shown in Fig. 2 must be applied.
The wave wall stability calculations are explained in 4.1.3. The wave
loadings are given in Table 4.
Because the 0.1% exceedance values of FH and pb do not occur simultaneously, pb in the stability calculations is reduced by 20% compared to
the values given in Table 4.
The dimensions of the upgraded wall shown in Fig. 11 correspond to
Tp = 13 s and a safety coefcient of 1.2 against sliding and overturning.
A wall designed for Tp = 16 s will have a volume which is approximately 25% larger.
4.2.2. Upgrading of revetment with high toe by adding a high berm to avoid
increase in structure crest level
Similar to the upgrading shown in 4.2.1, simulations with various
berm geometries are performed in order to identify the optimum
shape of the berm. The result is shown in Fig. 12.
The overtopping performance is given in Table 7. It is seen that it just
complies with the admissible overtopping as specied in 4.1.1.
By comparison with Table 4 a slight increase in overtopping caused
by the toe can be seen.
4.2.3. Upgrading of revetment with high toe by placing a submerged breakwater on the foreshore
If the foreshore seabed slope remains 1:100, the detached breakwater will be as shown in Fig. 8. However, if erosion steepens the foreshore
slope to 1:30 before the detached breakwater is built, a much stronger
breakwater is needed. The same design procedure as explained in
4.1.3 for the foreshore slope of 1:100 is used.
The design sea state corresponding to the 200 year return period is
given by Hs,I = 4.21 m, Tm 1,0 = 14.5 s, WL = + 2.87 m. The pileup is 0.42 m giving WL = + 3.29 m and water depth 6.29 m in front
of the revetment. The transmitted signicant wave height, which complies with the admissible overtopping of 40 l/sm, is Hs,t = 2.90 m,

corresponding to a wave transmission coefcient of Ct = Hs,I/Hs,I =

0.69. The needed freeboard of the breakwater is Rc = 1.21 m, and
the crest level will be +2.87 m1.21 m = +1.66 m.
The size of the breakwater rock armour is determined to be M50 =
13.0 t, Dn50 = 1.70 m, which is signicantly larger than the armour
size for the structure on the foreshore seabed slope 1:100.
The distance between the breakwater and the revetment would be
approximately 20 m.
Fig. 13 shows the cross section of the breakwater.
5. Cost optimization
The construction costs are estimated as volumes multiplied by the
in-place bulk volume unit prices for the different parts of the structure.
The unit prices for materials handled by land-based equipment are
listed in Table 8. The unit prices for materials in the detached breakwaters are estimated to be 25% higher than the prices given in the table.
The unit prices are based on information from a large international
consulting engineering company and are related to North-European
sites to which rock material has to be transported over long distances,
e.g. in Denmark where rock materials are transported on barges from
Sweden or Norway. The unit prices for rock materials will be considerably lower for sites with nearby quarries. The in-place unit costs are
always site specic and can therefore vary considerably. The volumes
of the structure parts are given in Table 9.
The following indications, limited to the investigated cases, can be
deducted from Table 10:
- Upgrading by installing a detached breakwater in front of the revetment seems signicantly more expensive than the other investigated upgrading concepts.
- Increase in crest level of the revetment is not necessary and not economically feasible as solutions consisting of adding an armour layer

Table 7
Overtopping performance for structure upgraded by adding an armour layer and a berm.
Return period (years)

10

50

100

200

Average overtopping discharge (l/sm)

0.09

4.0

10

26

41

120

H.F. Burcharth et al. / Coastal Engineering 87 (2014) 112121

Fig. 13. Detached breakwater for protection of revetment. Foreshore seabed slope 1:30.

Table 8
In-place bulk volume unit construction costs for materials handled by land-based equipment.
Main armour (EUR/m3)

Toe/berm armour (EUR/m3)

Filter layer 0.30.7 t

Shingle layer

Concrete wall

5.5 t

7.5 t

13 t

3.0 t

5.0 t

7.5 t

(EUR/m2)

(EUR/m2)

(EUR/m3)

60

65

70

55

60

65

50

30

350

Table 9
Volumes of structure parts.
Figure

Main armour (m3/m)

5.5 t

4
5
6
7
8a
8b
10a
10b
11
12
13

7.5 t

Toe/berm (m3/m)
13 t

3.0 t

5.0 t

7.5 t

23.98
23.98
32.92
48.98
25.76
40.66
20.95
29.02
30.09
42.45

7.35
7.35
10.00
64.52

Filter layer

Shingle layer

Concrete wall

(m2/m)

(m2/m)

(m3/m)

2.52
2.52
2.52
7.52
23.20
23.28
6.02
6.02
6.62
7.71
28.05

4.69
4.69
4.69
9.68
26.20
26.07
4.32
4.32
4.84
7.72
30.77

11.45
6.63

7.33

The estimated construction costs are given in Table 10 as function of foreshore seabed slope, type of toe and allowable increase in structure crest level.

on the front slope and in case of steepening of the foreshore also a

front berm will satisfy the target structure responses.
- Steepening of the foreshore, e.g. from slope 1:100 to 1:30, increases
the cost of upgrading signicantly.
- The costs seem to be somewhat less for revetments with a buried toe
compared to revetments with a high toe.
6. Discussion of safety aspects
The example structure upgrading presented above is based on
estimated 200 years return period environmental parameters. The
structure lifetime is set to 50 years. The reliability of the upgraded

structure would then correspond to an exceedance probability of

22% on the structure response (encounter probability) if the environmental parameters and the applied desk study design tools are
correct. This, however, is not the case as considerable uncertainties
are involved at all stages in such studies. For such conditions it is
of importance to perform a reliability analysis in which the uncertainties related to all parameters and design tools are taken into
account. A Level II (safety index) or a Level III (e.g. Monte Carlo simulations) method could be used on the condition that realistic
uncertainties and bias related to all elements in the analysis can be
given. This is unfortunately not the case for the present study. Besides the huge uncertainty related to the effects of climate change and

Table 10
Construction costs of upgrading in EUR/m of structure.
Foreshore slope

Toe type

1:100

Buried

Concept of upgrading and construction costs in EUR/m

No increase in crest level

High

1:30

Buried
High

Increase in crest level

Front armour, Fig. 4

Detached breakwat. Fig. 8a
Detached breakwat. Fig. 8b
Detached breakwat. Fig. 8a
Detached breakwat. Fig. 8b
Front armour + berm, Fig. 10a
Front armour + berm, Fig. 7

1.825
4.525
5.483
4.525
5.483
2.197
3.850

Front armour + berm, Fig. 12

Detached breakwater, Fig. 13

3.376
8.553

Increase in crest level not necessary

Front + crest armour + berm, Fig. 10b

2.721

Front armour + wall, Fig. 5

Front + crest armour + wall, Fig. 6
Front + crest armour + berm + wall, Fig. 11

5.833
4.727
5.548

H.F. Burcharth et al. / Coastal Engineering 87 (2014) 112121

Table 11
Preliminary classication of inuence of parameter uncertainties on upgrading of
revetment on soft seabed in shallow water.
Parameter

Inuence on upgrading
Very large

Environmental changes
Water level
Wave height (wave period)
Wind velocity
Subsequent morphological changes
Structure responses, by formulae etc.
Armour stability
Toe stability
Detached breakwater stability
Overtopping
Wave wall stability

Large

Moderate

Small

x
x
x
x

x
x

x
x
x
x

x
x

related morphological changes, the problem is also that structure

response formulae for many of the upgrading interventions do not exist,
as is explained in the paper. Also the use of a Neural Network and diagrams makes the set-up of an analysis complicated. Moreover, interactions between the various possible failure modes are quantitatively
unknown. The use of the PIANC safety factor system (Levels I and II) for
breakwaters (Burcharth and Sorensen, 2000) is not a possibility because
the system does not cover all the upgrading interventions.
Because of this situation rather conservative performance criteria
are used in the present upgrading study.
From the desk study tools it can be detected that the sensitivity of the
various structure and hydraulic responses to variations in water depth
and waves vary. Table 11 presents a preliminary classication of the inuence of parameter uncertainties on upgrading of a revetment on soft
seabed in shallow water. From this the most needed areas for studying
uncertainties related to desk studies can be seen.
The situation underlines the importance of performing physical model
tests for optimization of the structures and verication of the structure
performances.
7. Conclusions
Various concepts of upgrading of a typical rubble mound revetment
situated in shallow water on movable seabed material have been studied and related costs estimated. A scenario for climate change and a
structure service lifetime of 50 years have been anticipated. The method
of design is a conventional desk study using readily available formulae
and design tools, in several cases necessarily modied by simple judgement to be applied outside the valid ranges because of lack of covering
formulae. A probabilistic design procedure has not been practically possible given the available design tools.
The analyses have shown that if the foreshore remains with a gentle
slope of say 1:100, then the most economical concept of upgrading
seems to be adding an extra armour layer on the front slope. However,
if the foreshore is steepened due to erosion, it will be necessary to add
also a front berm. The costs of upgrading will be signicantly larger in
case of steepening of the foreshore due to the larger increase in wave
height. This underlines the importance of being able to predict the
future morphological changes on the coast. The study shows rather
small differences in upgrading costs for revetments with buried and
high toes.

121

Increase in the crest level of the revetment by adding an extra armour layer on the crest and/or installing a wave wall seems unnecessary
and not economically feasible compared to the solution described above.
Upgrading by installing a detached breakwater in front of the revetment seems signicantly more expensive than other upgrading concepts.
The overall conclusion of this specic study is that the anticipated
climate change does not impose a big problem as the structure can easily be upgraded at moderate costs.
The study reveals a lack of desk study design tools related to structure upgrading e.g. for the prediction of the stability of slopes upgraded
by adding extra armour layers, and for the prediction of the related
overtopping. At present it is absolutely necessary to perform physical
model tests along with the design work and not solely after as verication model tests.

Acknowledgement
The support of the European Commission through FP7. 20091, Contract 244104-THESEUS (Innovative technologies for safer European
coasts in a changing climate) is gratefully acknowledged as is also the
assistance of Jorgen Q. H. Nrgaard in calculating wave forces on wave
walls.

References
Baart, S., Ebbens, R.E., Nammuni-Krohn, J., Verhagen, H.J., 2010. Toe rock stability for rubble mound breakwaters. Proc. 32 International Conference on Coastal Engineering,
Shanghai, China.
Briganti, R., Van der Meer, J.W., Buccino, M., Calabrese, M., 2003. Wave transmission
behind low crested structures. Proc. Coastal Structures, ASCE, Portland, Oregon,
US.
Burcharth, H.F., Sorensen, J.D., 2000. The PIANC safety factor system for breakwaters. In:
Losada (Ed.), Proc. Coastal Structures '99, Santander, Spain. Balkema, Rotterdam,
pp. 11251144.
Burcharth, H.F., Hawkins, S.J., Zanuttigh, B., Lamberti, A. (Eds.), 2007. Environmental Design Guidelines for Low Crested Coastal Structures. Elsevier.
Ebbens, R.E., 2009. Toe structure of rubble mound breakwaters. (MSc. Thesis) Delft University of Technology, The Netherlands.
Gerding, E., 1993. Toe stability of rubble mound breakwaters. (MSc. Thesis) Delft University of Technology, The Netherlands.
Kramer, M., Burcharth, H.F., 2003. Stability of low-crested breakwaters in shallow
water short crested waves. Proc. Coastal Structures, ASCE, Portland, Oregon,
USA, pp. 137149.
Muttray, M., 2013. A pragmatic approach to rock toe stability. Coast. Eng. 82, 5663.
Nrgaard, J.H., Lykke Andersen, T., Burcharth, H.F., 2014. Distribution of individual
overtopping wave volumes in deep and shallow water wave conditions. Coast. Eng.
83, 1523.
STOWASUS, 2001. Regional storm, wave and surge scenarios for the 2100 century.
Denmark's Climate Center Report 0103. DMI.
Van der Meer, J.W., 1988. Rock slopes and gravel beaches under wave attack. Delft
Hydraulics Communication. . , No. 396.
Van der Meer, J.W., 1998. Geometrical design of coastal structures. In: Pilarczyk, K.W.
(Ed.), Dikes and revetments. Balkema, Rotterdam.
Van der Meer, J.W., d'Angremond, K., Gerding, E., 1995. Toe structure stability of
rubble mound breakwaters. Proc. Advances in Coastal Structures and Breakwaters Conference, Institution of Civil Engineers. Thomas Telford Publishing,
London, UK, pp. 308321.
Van Gent, M.R.A., Van den Boogaard, H.F.P., Pozueta, B., Medina, J.R., 2007. Neural network
modelling of wave overtopping at coastal structures. Coast. Eng. 54, 586593.
Van Gent, M.R.A., Smith, G.M., Van der Werf, I.M., 2013. Stability of rubble mound breakwaters with a berm: the upper slope. Proc. 33 International Conference on Coastal
Engineering, Santander, Spain.
Weisse, R., Bellaore, D., Menndez, M., Mndez, F., Nicholls, R.J., Umgiesser, G.,
Willems, P., 2014. Changing extreme sea levels along European coasts. Coast.
Eng. 87, 414 (in this issue).