See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/281750966

Damage risks and cost-benefit analysis of

climate adaptation strategies of RC structures
in Australia

Article · August 2015

CITATIONS READS

0 81

2 authors, including:

Mark G. Stewart
University of Newcastle
238 PUBLICATIONS 3,852 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Climate change effects and adaptation of civil infrastructure and buildings View project

All content following this page was uploaded by Mark G. Stewart on 15 September 2015.

The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document
and are linked to publications on ResearchGate, letting you access and read them immediately.
 CONCRETE INSTITUTE
CONCRETE IN AUSTRALIA

of AUSTRALIA

Volume 41 Issue 3 September 2015 $9.95 Inc. GST

www.concreteinstitute.com.au

CONCRETE 2015 DETAILS INSIDE

■ FULL TECHNICAL PROGRAM
■ RILEM WEEK
■ KEYNOTE PRESENTERS
■ EXHIBITORS AND SPONSORS
■ PRECAST CONCRETE WORKSHOP

SPECIAL FEATURE
■ DETERIORATION OF
REINFORCED CONCRETE
VOL 41, NO 3

CONCRETE 2015
SEPTEMBER 2015

Research into Practice

01 - Cover.indd 4 1/08/15 12:53 PM

 FEATURE: DETERIORATION OF REINFORCED CONCRETE

Damage risks and cost-benefit analysis of climate

adaptation strategies of RC structures in Australia
Lizhengli Peng and Mark G. Stewart
Centre for Infrastructure Performance and Reliability
School of Engineering, University of Newcastle

An increase of CO2 concentration in the atmosphere and changes in temperature and humidity may cause an acceleration of
deterioration processes of reinforced concrete (RC) structures as a result of climate change. This paper will describe cost-benefit
analysis of climate adaptation strategies for new RC structures in Australia. Spatial time-dependent reliability analysis is used
to predict likelihood and extent of carbonation-induced corrosion damage and corresponding damage costs. The cost-benefit
analysis includes the estimation of costs of adaptation strategies and corrosion damage, as well as cost effectiveness of adaptation
strategies for RC buildings in three Australian cities (Sydney, Canberra and Brisbane) under two climate emission scenarios.
Adaptation strategies include increasing concrete cover and upgrading concrete strength. The results show that upgrading concrete
strength for beams and small slabs may be more cost effective than the current durability design provisions. Break-even analysis
is also applied to estimate the maximum adaptation cost to make an adaptation strategy cost effective.

1.0 INTRODUCTION initiation and a shortened service life of up to 15 years for RC

Environment and exposure conditions are key variables that structures in France.
influence concrete deterioration. A changing climate has the In some cases, the adverse impact of climate change may
potential to change these variables in time and space. Increases necessitate the need for climate adaptation measures to
in atmospheric CO2 concentrations and temperature, and ameliorate damage risks. This requires a risk-based decision
reductions in relative humidity (RH), will increase the rate of support framework to optimise climate adaptation strategies (eg
carbonation-induced corrosion. Stewart and Deng10). Bastidas-Arteaga and Stewart11, 12
The impact of climate change on carbonation-induced assessed the costs and benefits of two climate adaptation
corrosion of reinforced concrete (RC) structures has been measures to reduce the impact of chloride-induced corrosion
the subject of recent research. Some of this research has damage on new and existing RC structures in France. In this
paper, two climate adaptation measures in order to reduce the
been deterministic1, 2. However, there is clear recognition
impact of carbonation-induced corrosion damage for new RC
that material, dimensional, deterioration and environmental
buildings are considered: (i) increase the concrete cover; and
parameters are stochastic in time and space. This is particularly
(ii) upgrade the concrete strength. The cost-effectiveness of an
apt for climate change projections for CO2 concentration,
adaptation strategy is measured in terms of Net Present Value
temperature, and RH. Preliminary stochastic modelling by
(NPV) which is equal to benefit minus cost. Sheltered cast
Stewart et al3, 4 showed the impact of a changing climate on
in-situ slabs and beams in RC buildings in three Australian
damage risks for carbonation and chloride-induced corrosion
cities (Sydney, Canberra and Brisbane) will be used as an
for RC structures in Australia. Peng and Stewart5 improved
deterioration modelling to more accurately characterise damage
risks, this time for RC structures in China. Peng and Stewart6
then developed spatial time-dependent reliability methods
to include not only the uncertainty of climate predictions,
deterioration processes and predictive models, but also the
spatial variability in concrete properties and dimensions. The
results show that a changing climate can cause the extent
of damage to increase by up to 6% for RC infrastructure in
Kunming, China. The time of the first repair can occur as
early as 35 years after construction. De Larrard et al7 coupled
a finite element model with a reliability approach to estimate
carbonation-induced corrosion of RC structures in France
under climate change. Climate change effects on durability of
chloride induced corrosion studied by Bastidas-Arteaga Figure 1: Projected global annual average CO2 concentration for RCP 8.5,
et al8, 9 found a 5–15% increase in the probability of corrosion RCP 4.5 and Year 2015 emission scenarios.

36 Concrete in Australia Vol 41 No 3

36-47 - Peng.indd 36 1/08/15 1:28 PM

 application of these cost-benefit analyses. Break-even analysis
is also applied to estimate the maximum adaptation cost
to make an adaptation strategy cost-effective. This paper
provides a methodology that could be extended to assess other
adaptation strategies or deterioration processes for RC structures
worldwide.
2.0 SPATIAL-TIME DEPENDENT RELIABILITY
ANALYSIS OF DAMAGE EXTENT AND RISKS
2.1 Scenario-based climate projection
Future climates may be influenced by changes in technology,
energy, population, economy, land use and agriculture, etc,
which makes projections difficult. The IPCC projects future
climate by defining carbon emission scenarios. The IPCC Fifth
Assessment Report (AR5) uses Representative Concentration
Pathways (RCPs)13. RCP 8.5 and RCP 4.5 emission scenarios
Figure 2: Projected annual average temperatures and relative humidity for the six GCM projections for RCP 8.5, RCP 4.5 and Year 2015 emission scenarios.
36-47 - Peng.indd 37
are included in this study, representing high and medium
emission scenarios, respectively. An emission scenario based
on the year 2015 CO2 level of 400 ppm is also used as a
reference for other emission scenarios. Figure 1 presents the
IPCC projection of the annual average CO2 concentrations
from 2015 related to RCP 8.5 and RCP 4.5 CO2 emission
scenarios.
Most Australian cities are coastal cities in a temperate climate
zone. Therefore, two coastal cities, Sydney and Brisbane,
are selected as case studies, as well as the largest inland city,
Canberra. Various Atmosphere-Ocean General Circulation
Models (AOGCMs) have been developed and these depend on
physical principles at the continental scale to project spatially
dependent temperature and RH changes in the future under
emission scenarios. Multiple models are used to take into
account the uncertainties of models in any impact assessment.
Six GCMs are used to project temperature and RH change
from 2015 to 21005, 6. Figures 2 and 3 present the projections
Concrete in Australia Vol 41 No 3 37
1/08/15 1:28 PM
 FEATURE: DETERIORATION OF REINFORCED CONCRETE
Figure 3: fT (t) and fRH(t) value of three cities under RCP 4.5 emission scenario.
of changes in temperature and relative humidity in three cities.
The average trend of temperature increases in the three cities
following the global warming projection, while the mean trend
of relative humidity decreases. All the projections based on the
six climate models and two emission scenarios will be used in
the simulation of concrete deterioration under climate change.
2.2 Deterioration and spatial time-dependent
modelling
Corrosion-induced concrete cover cracking and damage can be
found along the steel rebar. When the concrete crack reaches a
limit crack width, the concrete structure is severely damaged.
The various stages of carbonation-induced corrosion process
can be described as: (i) Ti – time to corrosion initiation; (ii)
t1st – time from corrosion initiation to crack initiation (visible
crack widths of 0.05 mm); (iii) tsev – time for a crack to
propagate from crack initiation to a limit crack width. The
time to severe corrosion damage is thus Tsev= Ti +t1st +tsev.
2.2.1 Corrosion initiation
The preliminary models of carbonation depth developed by
Duracrete14, Yoon et al15 and Stewart et al3 are improved by
four aspects: (i) CO2 levels, temperature and RH are time-
dependent variables; (ii) a correction factor ksite is introduced
to take into account increased CO2 levels in non-remote
environments; (iii) time-dependent change in the diffusion
coefficient due to changes in temperature; and (iv) time-
dependent change in the diffusion coefficient due to changes
in RH. The carbonation penetration process is simulated as
a steady state diffusion process, and the carbonation depth at
time t is5:
2 DCO2 (t ) t
t0nm
xc (t ) = ∫ fT (t ) f RH (t )ksiteCCO 2 (t )dt
a 2015
t − 2014
t ≥ 2015
DCO2 (t ) = D1 (t − 2014)
− nd
M CO2 −3.38 w/ c
a = 0.75CeCaO αH αH ≈ 1 − e
M CaO
38 Concrete in Australia Vol 41 No 3
36-47 - Peng.indd 38
where t is defined in calendar years starting from 2015,
assuming the buildings go into service in the year 2015;
CCO2(t) is the time-dependent increase in atmospheric CO2
concentration (10-3 kg/m3); ksite is a factor to account for
increased CO2 levels in non-remote environments; fT(t) is
time-dependent change in diffusion coefficient due to changes
in temperature; fRH(t) is time-dependent change in diffusion
coefficient due to changes in RH; the age factor nm for
microclimatic conditions related to the frequency of wetting
and drying cycles and equals to zero for sheltered outdoor, and
nm = 0.12 for unsheltered outdoor exposures. The functions
for ksite, fT(t) and fRH(t) are detailed in Peng and Stewart5, 6.
Generally, fT(t) will increase when temperature rises, however,
fRH(t) decreases when RH becomes higher. The fT(t) and fRH(t)
value of three cities under RCP 4.5 are shown in Figure 3.
It can be found that temperature increase caused by climate
change will have an adverse impact on carbonation induced
corrosion. The site correction parameter ksite of suburban
areas will be applied to CO2 concentrations projections in
Canberra and outer suburbs of Sydney and Brisbane (see
Table 3). DCO2(t) is time-dependent CO2 diffusion coefficient
in concrete; D1 is CO2 diffusion coefficient at t = 2015; nd
is the age factor for the CO2 diffusion coefficient; t0 is one
year; a is binding capacity relies on the degree of hydration of
the cement; Ce is cement content in concrete (kg/m3); CaO is
CaO content in cement (0.65); αH is a degree of hydration for
ordinary Portland cement (OPC) after more than 400 days16;
MCaO is molar mass of CaO and equal to 56 g/mol; and MCO2
is molar mass of CO2 equal to 44 g/mol.
2.2.2 Corrosion propagation
When carbonation depth reaches the reinforcing bar, corrosion
initiates (Ti) where Ti is the year when the carbonation depth
exceeds the concrete cover. The corrosion rate is variable and
(1) highly dependent on exposure conditions and atmospheric
situations. Because the corrosion reactions consume oxygen
(2) and water, the reactions will be slowed down when RH is
either too high or too low. A very high humidity in concrete
may reduce oxygen diffusion through the concrete while
(3) water shortage in concrete also declines the corrosion rate.
The corrosion rate will become negligible when RH is less
1/08/15 1:28 PM
 Table 1: Durability design specifications26 and deterioration models for building structures.
Class w/c ratio Ce a (kg/m3)
A2 0.56 320
Note – a: from AS 5100.527. b: from Yoon et al.15.
Table 2: Random field parameters 6.
Parameter Mean
Concrete cover Tables 3 & 4
Concrete strength fc(28) Tables 3 & 4
Diffusion coefficient D1 Table 4
Binding capacity a Eq.(3)
than about 50%17, 18. The peak RH for corrosion is 70-80%,
and then the corrosion rate decreases as RH increases18. An
empirical model that includes both RH and temperature
effects on the corrosion rate is proposed by Breysse et al19.
The corrosion rate at a reference state is assumed to be
log-normally distributed with statistical parameters shown
in Table 3 for moderate humidity sheltered exposure, and
the temperature and RH effects on the corrosion rate are
modelled as Breysse et al19 model. For more details see Peng
and Stewart6.
2.2.3 Crack initiation and propagation
The time to crack initiation is estimated by a model proposed
by El Maaddawy and Souki20. Note that the accuracy of the
time to corrosion damage is dominated by the accuracy of
time to corrosion initiation and the time to crack propagation,
and so damage risk predictions are relatively insensitive to the
crack initiation model21. The definition of how ‘excessive’ the
crack width depends on individual conditions and asset owner
policies; in this paper, concrete structures will be regarded as
‘severely cracked’ when crack width reaches a crack width of
1.0 mm22. The crack initiation and propagation models used
in the simulation analysis are influenced mainly by concrete
strength, concrete cover and corrosion rate.
2.3 Spatial time-dependent reliability model
The inherent variability of RC structures due to material
properties and quality control in construction can result in
corrosion damage being spatially variable. Moreover, modelling
material and dimensional parameters as homogeneous can lead
to non-conservative predictions of failure for RC structures in
corrosive environments23. Also, a spatial analysis can provide
more detailed information on the extent of damage at a given
point in time to be estimated, allowing detailed analysis of
maintenance strategies and damage cost estimation. As such,
incorporating spatial variability into predictive models is
important, particularly for deterioration processes. The spatial
variability of concrete cover, concrete strength, diffusion
coefficient and binding capacity are included in the analysis6.
Random field theory is applied to model the spatial variability
of these parameters that a surface is discretised into a number
36-47 - Peng.indd 39
Mean D1b (× 10-4 cm2/s) Dbar (mm) Cover (mm) F’c (MPa)
2.22 12 30 25
COV Scale of fluctuation θ (m) Distribution
Table 3 2 Truncated normal
Table 3 1 Truncated normal
σ=0.15 2 Lognormal
0.3 2 Lognormal
of identical elements of size Δ and the random variables
within the random field are statistically correlated based on the
corresponding correlation function24. The midpoint method is
used that an element is represented by the value at the centroid
of that element, and this value is assumed to be constant within
the element. The validation and calibration of correlation
functions demand large amounts of spatial data that is hard to
achieve. Therefore, engineering judgement or expert experience
is required to match a correlation function to a particular
random field. The scale of fluctuation (θ) defines the distance
over which correlation persists in a random field. Once the
stochastic random field is defined, Monte Carlo simulation
methods are used to generate random variables for each
element, and the spatially variable parameters of each element
are then defined using the correlation function. This allows the
proportion of the concrete surface to experience severe corrosion
damage at time t (dcrack(t)) to be calculated.
2.4 Durability design specifications and statistical
parameters for RC structures
The exposure classifications for RC structures in the three
Australian cities are near-coastal (B1: 1-50 km from coast)
and coastal excluding tidal and splash zones (B2: <1 km
from coast) for the most parts of Sydney and Brisbane, and
A2 (>50 km from coast) for Canberra and outer suburbs of
Sydney and Brisbane. Design cover for B1 is 40 mm, which
is higher than 30 mm cover for A2. Note that RC structures
close to the shore (B1 and B2) have higher concrete cover
and are not susceptible to carbonation induced corrosion.
Therefore, only RC structures of exposure A2 are studied here,
as shown in Table 1.The geometry of RC building slabs is
assumed to be 6 m × 6 m two-dimensional random field, with
element size Δ=0.5 m. A RC beam with size of 0.5 m × 0.5 m
× 6 m can be discretised into a one-dimensional random field.
For review details about discretisation methods, the correlation
function, scale of fluctuation and element sizes, etc, see
Vanmarcke24 and Peng 25. The spatial and spatially dependent
random variables, standard random variables and deterministic
variables, as well as the model errors used in the spatial time-
dependent reliability analysis, are summarised in Tables 2 and
3. The results for RC slabs can be applied to floors, panels
Concrete in Australia Vol 41 No 3 39
1/08/15 1:28 PM
 FEATURE: DETERIORATION OF REINFORCED CONCRETE

Table 3: Statistical parameters for corrosion parameters, material properties and dimensions.

Parameters Mean COV Distribution Reference

Concrete cover Cnome +6 mm σ=11.5 mm Truncated normala 28

Compressive strength fc(28):

25 MPa 1.05F’cf 0.156 Truncated normalb 29

32 MPa 1.06F’cf 0.152 Truncated normalb 29

40 MPa 1.07F’c f
0.151 Truncated normal b 29

Tensile strength ft 0.53(F’c)0.5 0.13 Normal 30

Elastic modulus Ec 4600(F’c)0.5 0.12 Normal 30

Age factor nd Table 4 0.12 Normal 3

Model error ME(rcrack) 1.04 0.09 Normal 22

Thickness of pore zone δ0 15 μm 0.1 Normal 3

Correction factor ksite: 5

Urban area 1.14 0.08 Truncated normalc 5

Suburban area 1.07 0.06 Truncated normalc 5

Rural area 1.05 0.04 Truncated normal c 5

Corrosion rate icorr(ref) 0.172 μA/cm2 0.5 Lognormald 14

Notes – a: truncated at 8 mm. b: truncated at 0 MPa. c: truncated at 1.0. d:1 μA/cm2 = 0.0116 mm/year. e: Cnom is the nominal or
design cover. f: F’c is the nominal design concrete compressive strength.

and walls, etc. Note that RC structures are assumed to use
OPC concrete, and standard formwork and compaction.
3.0 MAINTENANCE STRATEGIES
A patch repair maintenance strategy will be used for
comparison purposes to study the impact of climate
adaptation strategies. Patch repairs are the most common
type of repair for corrosion damaged RC structures31, 32. It
is a corrective repair strategy in which repair takes place
after severe concrete cracking, but the loss of the cross-
sectional area of rebars is not significant. The patch repair
process involves the mechanical removal of the damaged
concrete (typically to approximately 25 mm beyond the
rebar), the cleaning and treatment of the corroded steel, and
reinstatement of the concrete cover with a suitable material.
The use of patch repairs in a maintenance strategy can be
modelled very effectively using a spatial time-dependent
reliability analysis as both the likelihood and extent of damage
can be predicted33. Therefore, the patch repair maintenance
technique defined herein will be integrated into the cost-
benefit analysis of an adaptation strategy. A maintenance
strategy includes: (i) inspection interval; (ii) damage limit
state; (iii) maintenance technique.
The inspection interval Δt is usually based on the controlling
authority’s policy or the prior experience of the asset owner/
operator. A longer inspection interval will save on inspection
costs but can mean that when detected, the damage to the
structure has exceeded the pre-defined repair threshold and
hence results in a more costly repair area. Further, at longer
inspection intervals, there is a risk of the corrosion damage
reducing the structural safety of the RC structure, and this must
be carefully considered when defining the inspection interval.
For an RC building, such as a residential apartment which
can be easily inspected, any corrosion damage can be noticed
quickly, therefore the routine inspection interval Δt = 1 year34.
The damage limit state, Xrepair (also referred to as the repair
threshold), is usually over a range of practical values from
0.5% to 5% for patch repairs, and 12% and 20% for complete
rehabilitative overlay35. In this study, the minimum repair area is
0.5 m × 0.5 m for RC slabs and beams. Therefore, Xrepair is 0.7%
and 8.3% for RC slabs and beams, respectively. Patch repair is
assumed to be carried out immediately if the extent of corrosion
damage has been discovered to exceed the repair threshold Xrepair
at the time of ith inspection iΔt. In other words, all observed
corrosion damaged areas will be repaired immediately after
inspection.
It is assumed that (i) severe corrosion induced cracking is
always detected when the structure is visually inspected; (ii)
only the damaged area will be repaired; the remaining RC
surface area will continue to deteriorate; (iii) repair will not
improve durability performance of the repaired structures; ie
cover and concrete quality are the same as the original design
specification; and (iv) damage will not re-occur for the repaired
area during the remaining service life of the structure. This last
assumption is slightly non-conservative for estimating corrosion
damage losses; however, Peng25 shows less than 0.2% probability
that damage may reoccur once repaired.
4.0 ADAPTATION STRATEGIES
A wide range of methods can be used to enhance the
durability of RC structures, and these can be applied to
reduce the adverse effects of a changing climate. The design

40 Concrete in Australia Vol 41 No 3

36-47 - Peng.indd 40 1/08/15 1:28 PM

 options generally include a selection of cover, cement and
concrete mix, surface coating barriers, extraction, and cathodic
protection36.
4.1 Increase in concrete cover thickness
An increase in cover thickness can increase the time of
carbonation ingress to reach concrete reinforcement and in
turn delay carbonation-induced corrosion. According to Peng
and Stewart5, carbonation-induced corrosion damage is most
sensitive to concrete cover thickness. It is, therefore, one of
the more obvious adaptation options in the design of concrete
infrastructure. Moreover, increased concrete cover can provide
thermal insulation, which protects reinforcement from fire.
However, increasing cover thickness may result in more costly
forms, concrete, reinforcement, finishing and labour, as well as
increasing the structure’s deadweight.
4.2 Increase strength grade
Another option may be the use of concrete with lower water/
cement ratio or the increase in cement content. Rather
than giving details on the concrete mix or water/cement
ratio, selection of a higher strength grade of concrete is one
approach to reducing the diffusion coefficient, in addition
to enhancing its mechanical properties. However, using high
strength concrete may cause difficulties in curing and increases
in cement content that may lead to more CO2 emissions,
since producing cement is one of the biggest sources of CO2
emissions37.
4.3 Other adaptation measures
Any measure to reduce the diffusion coefficient of CO2 or
inhibit the corrosion rate will increase the adaptive capacity of
concrete structures against a changing climate. There are other
adaptation measures that may be suitable to ameliorate the
effects of a changing climate. These include special concretes,
corrosion inhibitors, concrete surface coatings, stainless steel,
galvanised reinforcement, concrete-polymer composites and
cathodic protection, etc. Their cost and effectiveness vary and is
beyond the scope of the present paper. For more details about
the relative merits of other adaptation measures see Stewart et
al36 and Wang et al38.
4.4 Adaptation strategy C1, C2, S1 and S2
Increase in concrete cover and increase in concrete strength
grade are used as adaptation strategies for RC structures.
Business as usual (BAU) and four adaptation strategies are
defined (see Table 4):
• Adaptation strategy C1: increase concrete cover by 5 mm;
• Adaptation strategy C2: increase concrete cover by 10 mm;
• Adaptation strategy S1: increase concrete strength by one
grade (25 MPa → 32 MPa);
• Adaptation strategy S2: increase concrete strength by two
grades (25 MPa → 40 MPa).
5.0 COST-BENEFIT ANALYSIS
5.1 Decision rules
Stewart35 proposed that the total life cycle cost (LCC) can be
described as:
LCC (T)
CD + CC + CQA + CIN (T) + E damage (T) (4)
where CD is the design cost, CC is the construction cost
(materials and labour), CQA is the cost of quality assurance/
control, CIN(T) is the cost of inspections, and Edamage(T) is the
expected cost of repair or rehabilitation of corrosion-induced
damage during service life T.
The expected cost of damage can be described as a present
value39:
T / Δt Cdamage
E damage (T) = ∑ ΔPf ,i (5)
i =1 (1 + r)iΔt
where Δt is the inspection interval, ΔPf,i is the probability that
the extent of damage exceeds the repair threshold between the
(i-1)th and ith inspections, Cdamage is the cost of damage and r
is the discount rate. Cdamage will consist of the maintenance
costs and user delay costs for the corresponding maintenance
strategy. At the time of the first repair, ΔPf, i can be expressed in
terms of the repair threshold (Xrepair) and the extent of damage
(dcrack) as:
(
ΔPf ,i = Pr d crack (t ) ≥ X repair d crack (t − Δt ) < X repair (6) )
The probabilities used in the estimation of the expected
damage costs are calculated using Monte Carlo Simulation
techniques.
The aim of the current study is to compare the effect of
adaptation strategies on the life cycle cost of RC structures. As
such, initial costs of design, construction, quality assurance and
inspection costs will be assumed to be equal for all cases analysed.
The economic performance is then determined primarily by the
expected damage costs which are affected by adaptation strategies
and the extra costs to conduct adaptation strategies. The ‘benefit’
of an adaptation measure at time t, B(t), is the reduction in
damage costs related to the adaptation strategy plus any co-
benefit, and the ‘cost’ is the extra costs to take the adaptation
strategy at time t. The Net Present Value (NPV) is:

Table 4: Durability design requirements of Australian RC buildings for BAU and four adaptation strategies.

Cover (mm) F’c (MPa) w/c Ce (kg/m3) Mean D1 Mean nd

BAU 30 25 0.56 320 2.22 0.240
Adaptation C1 35 25 0.56 320 2.22 0.240
Adaptation C2 40 25 0.56 320 2.22 0.240
Adaptation S1 30 32 0.50 320 1.24 0.235
Adaptation S2 30 40 0.46 370 0.65 0.218

Concrete in Australia Vol 41 No 3 41

36-47 - Peng.indd 41 1/08/15 1:28 PM

 FEATURE: DETERIORATION OF REINFORCED CONCRETE
Table 5: Costs of four adaptation strategies and damage for RC structural elements in Australia.
Costs Structural element D (mm)
Slabs – small 100
Cadapt ($/m2) Slabs – large 250
Beams 500
Cdamage ($/m )2
Slabs – small 100
Cadapt / Cdamage Slabs – large 250
Beams 500
T T
NPV = ∑ B (t ) − ∑ Cadapt (t )
t =0 t =0
B(t) = Edamage− BAU (t) − Edamage− adaptation (t) + Δ B
where Edamage-BAU(t) and Edamage-adaptation(t) are the expected
damage costs for BAU and each adaptation measure,
respectively; and Cadapt(t) is cost of adaptation strategies at
time t. The co-benefit of adaptation (ΔB) may include thermal
insulation provided by increased concrete cover or improved
mechanical properties due to upgraded concrete strength, etc.
For instance, flexural capacity of an RC beam will increase if
concrete strength is upgraded from 25 MPa to 32 MPa. These
indicate that concrete or reinforcement bars usage can be
reduced to meet the capacity requirement if concrete strength
is upgraded. However, co-benefits of adaptation can be
negative values, such as increased structure’s deadweight due to
a higher concrete cover, or increased CO2 emissions because of
higher consumption of cement in higher strength concrete.
An adaptation measure that results in a NPV larger than zero
is a cost-effective adaptation measure. NPV is time-dependent
because costs and benefits are time-dependent. Thus, an
adaptation measure may not be profitable in the short-term due
to the high adaptation cost, but be cost-effective in the longer-
term if the benefits increase over time.
5.2 Costs of adaptation (Cadapt)
Adaptation strategies, such as increasing concrete cover and
upgrading concrete strength, are conducted at the design
stage. Therefore, additional construction costs will be
generated due to these adaptation strategies. Further, all these
costs are country, site and structure specific. Therefore, in this
section, the estimation of costs of adaptation strategies (Cadapt)
is investigated based on the average level of local construction
costs in Australia in terms of price per unit area or volume.
Note that all the costs are expressed in terms of 2014
Australian dollars unless noted otherwise.
5.2.1 Cadapt for increasing design cover (C1, C2)
The baseline case for Australia construction cost per unit
volume of concrete, including forms, concrete, reinforcement,
finishing and labour, is approximately $173-258/m2 and
42 Concrete in Australia Vol 41 No 3
36-47 - Peng.indd 42
C1: + 5 mm C2: + 10 mm S1: + 1 grade S2: + 2 grades
8.7 17.3 0.5 1.1
5.2 10.3 1.3 2.8
7.9 15.9 2.5 5.5
1000
0.0087 0.0173 0.0005 0.0011
0.0052 0.0103 0.0013 0.0028
0.0079 0.0159 0.0025 0.0055
$1395-1585/m3 for RC slabs and RC beams, respectively40. It
(7) is assumed that the additional cost of forms, reinforcement,
concrete and labour to increase design cover is proportional to
the additional volume of concrete needed. Since all cost units
(8) are $/m2 of surface area, but Ccv is given as per unit volume,
then Cadapt should be converted to cost per surface area
exposed, and so is corrected for structural member dimensions
such as slab depth or width of a beam. Table 5 presents the
adaptation costs for a 5 and 10 mm increase in additional
cover.
5.2.2 Cadapt for increasing strength grade (S1, S2)
The cost of normal weight ready mix concrete using Ordinary
Portland Cement including aggregates, sand, cement and
water (but excluding additives and treatments) delivered on-
site increases from $208/m3 to $213/m3 for 25 MPa upgrade
to 32 MPa concrete40. Similarly, for the adaptation strategies
S1 and S2, all costs are converted to cost per surface area
exposed to deterioration. Table 5 presents the additional costs
of increasing strength grades Cadapt ($/m2).
5.3 Costs of damage (Cdamage)
The cost of damage (Cdamage) includes direct costs of patch
repair and indirect costs caused by user delays, etc. The cost
of repairs for RC structures will be based on the method
of repair and the extent of repaired area. The Crepair of patch
repair method in France12 is estimated to be $360/m2, and
$670/m2 in the US41. Therefore, Crepair is assumed to be
approximately $500/m2 for Australian RC buildings. The
total cost of a repair can be calculated based on the unit cost,
the area to be repaired and any fixed costs associated with
a particular repair technique. Corrosion damage is assumed
to occur on one (exposed) face of the slab and beam. Repair
costs can vary with geographic regions and current economic
circumstances. Further, site-specific access costs and asset
owner restrictions must be considered. For example, a repair
on an internal wall would likely involve significantly more
associated costs than an external wall. Moreover, user delay
costs including all costs associated with user disruption due to
the RC structure operating at a reduced level of service can be
considerable23. In this study, the user delay costs are assumed
to be the same as the patch repair costs12. Therefore, Cdamage is
equal to $1000/m2.
1/08/15 1:28 PM
 6.0 APPLICATIONS: AUSTRALIAN RC BUILDINGS
Sheltered cast in-situ RC structural members are generally
subject to more severe carbonation-induced corrosion damage
compared with precast or unsheltered members5, 6. Hence,
sheltered cast in-situ RC structural members constructed in
2015 are studied herein. The durability design requirements
for sheltered RC slabs and beams in three Australian cities
are presented in Table 1. The spatial and spatially dependent
random variables, standard random variables and deterministic
variables, as well as the model errors used in the spatial time-
dependent reliability analysis, are summarised in Tables 2
and 3. The durability design requirements for BAU and four
adaptation strategies are described in Table 4. Cost data for
Australian cases are presented in Table 5, the discount rate is
4%, and all costs are expressed in 2014 Australian dollars.
To demonstrate the spatially-distributed carbonation
induced corrosion progress on a two-dimensional surface, the
carbonation depth, time to corrosion initiation, time to first
Run 1
2070:
2080:
2090:
2100:
Corrosion not initiated. Corrosion initiated.
Figure 4. Simulation of spatially distributed corrosion process showing three typical Monte Carlo realisations for
sheltered cast in-situ RC slab in Sydney under RCP 8.5 emission scenario.
36-47 - Peng.indd 43
cracking, and time to severe corrosion damage of each element
are calculated. Figure 4 shows three typical Monte Carlo
realisations for 6m × 6m RC slabs. The area where corrosion
initiated, cracking initiated and severe corrosion damaged is
demonstrated in every ten years after 2070. Each realisation
shows a quite different corrosion progression in time and space.
It also shows the clustering of damage.
For the spatial time-dependent reliability analysis, the
proportion of surface corrosion damage was calculated at each
time step for 10,000 simulation runs. Figure 5 shows the effects
of adaptation strategies on the mean extent of corrosion damage
of RC buildings in Sydney, Canberra and Brisbane under RCP
8.5 and RCP 4.5 emission scenarios. These four adaptation
strategies can bring about a noticeable reduction of the mean
extent of corrosion damage. Adaptation strategies C1 (cover +5
mm) and C2 (cover +10 mm) are more effective in decreasing
the mean extent of corrosion damage than adaptation strategies
S1 (strength +1 grade) and S2 (strength +2 grades). RC
Run 2 Run 3
Crack initiated. Severe corrosion damage.
Concrete in Australia Vol 41 No 3 43
1/08/15 1:28 PM
 FEATURE: DETERIORATION OF REINFORCED CONCRETE
Figure 5: Mean extent of surface corrosion damage of BAU and four
adaptation strategies for RC buildings in Sydney, Canberra and Brisbane
under RCP 8.5 and RCP 4.5 emission scenarios.
buildings in Brisbane have the highest mean extent of corrosion
damage, so they are most in need of adaptation strategies. The
results of mean extent of corrosion damage under RCP 8.5 are
more severe than those under RCP 4.5.
Figure 6 shows the mean expected damage cost Edamage of
BAU and four adaptation strategies for RC buildings in Sydney
and Brisbane under the RCP 8.5 emission scenario. Compared
to Figure 5, the expected damage costs are found to be closely
Figure 6: Expected damage costs ($/m2) of BAU and four adaptation strategies for RC buildings in Sydney and Brisbane under RCP 8.5 emission scenario.
44 Concrete in Australia Vol 41 No 3
36-47 - Peng.indd 44
related to the mean extent of corrosion damage. On the other
hand, even though adaptation strategies can significantly reduce
the expected damage costs by 50–90% compared to those for
BAU, the reduced damage costs $0.5/m2–$1.4/m2 are lower
than most of the adaptation costs (see Table 5), which indicates
that these adaptation strategies are unlikely to be cost effective.
6.1 Net present value and break-even analysis
The mean NPV assuming no co-benefit of four adaptation
strategies for RC buildings in Sydney, Canberra and Brisbane
for all emission scenarios are given in Table 6. As can be seen
in Table 6, the mean NPVs of strategies S1 and S2 for small
slabs in Brisbane are positive values, while other cases show
net losses. The overall trend indicates that the mean NPV
is highly dependent on the location. The mean NPVs for
RC buildings in Brisbane are slightly higher than those for
Sydney and Canberra. These results are expected because the
predictions of the likelihood and extent of corrosion damage
for BAU in Sydney and Canberra are very low, and then
the reduced damage costs due to adaptation strategies will
be accordingly low. Therefore, the higher the likelihood and
extent of corrosion damage, the more chances for adaptation
strategies to be cost effective. In other words, it is highly likely
that climate adaptation strategies will not be cost effective
1/08/15 1:28 PM
 Table 6: Mean NPV of four adaptation strategies for RC slabs and beams assuming no co-benefit.

slab 100 mm slab 250 mm beam

Year Year Year
RCP 8.5 RCP 4.5 RCP 8.5 RCP 4.5 RCP 8.5 RCP 4.5
2015 2015 2015
C1 -8.0 -8.1 -8.2 -4.5 -4.6 -4.7 -7.2 -7.3 -7.4
C2 -16.3 -16.4 -16.6 -9.3 -9.4 -9.6 -14.9 -15.0 -15.2
Sydney
S1 0.0 0.0 -0.1 -0.8 -0.8 -0.9 -2.0 -2.0 -2.1
S2 -0.2 -0.3 -0.4 -1.9 -2.0 -2.1 -4.6 -4.7 -4.8
C1 -8.2 -8.3 -8.4 -4.7 -4.8 -4.9 -7.4 -7.5 -7.6
C2 -16.7 -16.7 -16.8 -9.7 -9.7 -9.8 -15.3 -15.3 -15.4
Canberra
S1 -0.2 -0.2 -0.2 -1.0 -1.0 -1.0 -2.2 -2.2 -2.2
S2 -0.5 -0.5 -0.6 -2.2 -2.2 -2.3 -4.9 -4.9 -5.0
C1 -7.7 -7.8 -7.9 -4.2 -4.3 -4.4 -6.9 -7.0 -7.1
C2 -15.9 -16.0 -16.3 -8.9 -9.0 -9.3 -14.5 -14.6 -14.9
Brisbane
S1 0.2 0.1 0.1 -0.6 -0.7 -0.7 -1.8 -1.9 -1.9
S2 0.1 0.1 -0.1 -1.6 -1.6 -1.8 -4.3 -4.3 -4.5

Table 7: Mean break-even value for Cadapt/Cdamage for RC structures.

Sydney Canberra Brisbane

RCP 8.5 RCP 4.5 Year 2015 RCP 8.5 RCP 4.5 Year 2015 RCP 8.5 RCP 4.5 Year 2015
C1 0.0007 0.0006 0.0005 0.0005 0.0004 0.0003 0.0010 0.0009 0.0008
C2 0.0010 0.0009 0.0007 0.0006 0.0006 0.0005 0.0014 0.0013 0.0010
S1 0.0005 0.0005 0.0004 0.0003 0.0003 0.0003 0.0007 0.0006 0.0006
S2 0.0009 0.0008 0.0007 0.0006 0.0006 0.0005 0.0012 0.0012 0.0010

for those RC structural components having a small extent of
corrosion damage. On the other hand, even though increased
concrete cover (C1 and C2) reduces extent of damage and
damage cost more, Cadapt values are too high to make it cost
effective.
The likelihood and extent of corrosion damage, cost of
damage, and adaptation costs are subject to considerable
uncertainty and are country, site and structure specific. For
this reason, calculations of corrosion risks, costs and benefits
will be imprecise. Hence, a break-even analysis may be useful
to estimate the maximum adaptation costs required to make
the benefit of an adaptation strategy equal to its cost. In other
words, if adaptation costs exceed the predicted break-even value,
then adaptation is not cost-effective. In this case, adaptation
costs are normalised by the ratio Cadapt/Cdamage to reflect
uncertainty in Cadapt and Cdamage. Table 7 shows the mean break-
even value for Cadapt/Cdamage for RC structures under all emission
scenarios.
6.2 Consideration of co-benefit
The co-benefit (ΔB) for adaptation measures S1 and S2 can
be considerable. An increase in concrete grade to 32 MPa
(S1) can reduce the effective depth of RC beams by 0.5% to
2%, or the amount of tensile steel by 1% to 4% whilst still
maintaining flexure and shear capacities. Adaptation strategy
S2 reduces the effective depth of RC beams by 1% to 3%,
or the amount of tensile steel by 2% to 6%. For RC beams,
Table 5 shows that a 1% change of depth (5 mm) costs
$7.90/m2. Hence, the minimum co-benefit of reduced RC
beam depth is $4/m2 and $8/m2 for S1 and S2, respectively.
To be sure, there may be design outcomes that do not change
if concrete grade is increased; however, if only 50% of designs
result in a more efficient use of materials due to a higher
concrete grade then co-benefit will still easily exceed $2/m2
and $4/m2 for S1 and S2, respectively. If a lower bound co-
benefit of ΔB=$2/m2 for S1 is added to the mean NPV results
in Table 6 then mean NPV for RC beams in Brisbane would
equal +$0.2/m2 and +$0.1/m2 under RCP 8.5 and RCP 4.5
emission scenarios, respectively. Hence, it is highly likely that
adaptation strategy S1 is cost-effective for Brisbane. However,
adding a co-benefit of ΔB=$4/m2 for S2 to the mean NPV
results in Table 6 would still result in a net loss for all cities.
For RC slabs, an increase in concrete grade to 32 MPa
(S1) can reduce the effective depth or tensile steel by 0.3%
to 1.1%. Adaptation strategy S2 reduces the effective depth
or tensile steel of RC slabs by 0.5% to 1.9%. For RC slabs,
Table 5 shows that a 1% change of depth costs $1.74/m2 and
$3.12/m2 for small and large slabs, respectively. Hence, the
minimum co-benefits for small and large RC slabs designed
to S1 are $0.50/m2 and $0.95/m2, respectively. The minimum
co-benefits for small and large RC slabs designed to S2 are

Concrete in Australia Vol 41 No 3 45

36-47 - Peng.indd 45 1/08/15 1:28 PM

 FEATURE: DETERIORATION OF REINFORCED CONCRETE
$0.90/m2 and $1.60/m2, respectively. If only 50% of small
RC slab designs result in a more efficient use of materials due
to a higher concrete grade then co-benefit will easily exceed
$0.25/m2 and $0.5/m2 for S1 and S2, respectively. In this case,
adding these co-benefits to mean NPV results in Table 6 shows
a positive NPV for small RC slabs in Brisbane, Canberra, and
Sydney for adaptation strategy S1. Hence, it is highly likely
that adaptation strategy S1 is cost-effective for small RC slabs
in Brisbane, Canberra and Sydney. Similarly, it is highly likely
that adaptation strategy S2 is cost-effective for small RC slabs
in Brisbane and Sydney, but is less likely to be cost-effective for
Canberra, where mean NPV=0. If only 50% of large RC slab
designs result in a more efficient use of materials due to a higher
concrete grade then co-benefit will easily exceed $0.45/m2
and $0.8/m2 for S1 and S2, respectively. In this case, mean
NPV shows a net loss when these co-benefits are added to the
mean NVP results in Table 6. It follows that it is unlikely that
adaptation strategies S1 and S2 are cost-effective for large RC
slabs in Brisbane, Canberra and Sydney.
A precise calculation of co-benefit will depend on the
structural configuration under consideration. Nonetheless,
there is likely to be a strong co-benefit in terms of increased
flexural and shear capacity, as well as increased member stiffness,
resulting from a design with higher concrete grade. If this is
factored into the cost-benefit decision process, then adaptation
strategies S1 and S2 may well be cost-effective. This is an area
for further work before revisions to existing RC design practices
(such as specified in AS3600) can be recommended.
7.0 CONCLUSIONS
This paper has focused on climate adaptation strategies for
RC structures in three Australian cities subject to carbonation-
induced corrosion damage under two different climate
scenarios. Four adaptation strategies are included which
consist of increasing the concrete cover by 5 or 10 mm,
and increasing the concrete strength by one or two grades.
The spatial time-dependent reliability analysis is used to
estimate the likelihood and extent of corrosion damage and
corresponding damage costs for RC slabs and beams. The
net present value of each adaptation strategy is calculated.
According to current information, except for small slabs
in Brisbane, ‘business as usual’ for RC buildings’ members
in three Australian cities is likely to be more cost-effective
than these four adaptation strategies, even under a changing
climate. However, if co-benefit is considered, then increasing
concrete strength may be cost effective for small slabs in three
cities and RC beams in Brisbane.
ACKNOWLEDGEMENT
The authors acknowledge the financial support provided by
the Australian Research Council. The authors also would like
to express their appreciation to Dr Xiaoming Wang and Dr
Stacey Dravitzki of CSIRO for providing climate data for
Australian cities.
46 Concrete in Australia Vol 41 No 3
36-47 - Peng.indd 46
REFERENCES
1. Talukdar, S., Banthia, N., Grace, J.R. and Cohen, S.,
“Carbonation in concrete infrastructure in the context
of global climate change: parts 1 and 2”, Cement and
Concrete Composites, Vol. 34, No. 8, 2012, pp. 924-935.
2. Saha, M. and Eckelman, M.J., “Urban scale mapping
of concrete degradation from projected climate change”,
Urban Climate, Vol. 9, No. 0, 2014, pp. 101-114.
3. Stewart, M.G., Wang, X.M. and Nguyen, M.N.,
“Climate change impact and risks of concrete
infrastructure deterioration”, Engineering Structures, Vol.
33, No. 4, 2011, pp. 1326–1337.
4. Wang, X.M., Stewart, M.G. and Nguyen, M.N.,
“Implications of climate change to the durability of
concrete structures – An investigation by simulation”,
Concrete in Australia, Vol. 36, No. 4, 2010, pp. 27-35.
5. Peng, L. and Stewart, M.G., “Climate change and
corrosion damage risks for reinforced concrete
infrastructure in China”, Structure and Infrastructure
Engineering, 2015, (in press).
6. Peng, L. and Stewart, M.G., “Spatial time-dependent
reliability analysis of corrosion damage to RC structures
with climate change”, Magazine of Concrete Research, Vol.
66, No. 1, 2014, pp. 1-16.
7. de Larrard, T., Bastidas-Arteaga, E., Duprat, F. and
Schoefs, F., “Effects of climate variations and global
warming on the durability of RC structures subjected to
carbonation”, Civil Engineering and Environmental Systems,
Vol. 31, No. 2, 2014, pp. 153-164.
8. Bastidas-Arteaga, E., Chateauneuf, A., Sanchez-Silva,
M., Bressolette, P., & Schoefs, F. “Influence of weather
and global warming in chloride ingress into concrete: A
stochastic approach”, Structural Safety, Vol. 32, No. 4,
2010, pp. 238-249.
9. Bastidas-Arteaga, E., Schoefs, F., Stewart, M.G. and
Wang, X., “Influence of global warming on durability
of corroding RC structures: A probabilistic approach”,
Engineering Structures, Vol. 51, No. 0, 2013,
pp. 259-266.
10. Stewart, M.G. and Deng, X., “Climate Impact Risks and
Climate Adaptation Engineering for Built Infrastructure”,
ASCE-ASME Journal of Risk and Uncertainty in
Engineering Systems, Part A: Civil Engineering, 1(1),
2015, 04014001.
11. Bastidas-Arteaga, E. and Stewart, M.G., “Damage risks
and economic assessment of climate adaptation strategies
for design of new concrete structures subject to chloride-
induced corrosion”, Structural Safety, Vol. 52, Part A,
2015, pp. 40-53.
12. Bastidas-Arteaga, E. and Stewart, M.G., “Economic
Assessment of Climate Adaptation Strategies for Existing
RC Structures Subjected to Chloride-Induced Corrosion”,
Structure and Infrastructure Engineering, 2015, (in press).
13. IPCC. Climate Change 2013: The Physical Science Basis.
Contribution of Working Group I to the Fifth Assessment
Report of the Intergovernmental Panel on Climate Change.
Cambridge, United Kingdom and New York, NY, USA:
Cambridge University Press; 2013.
1/08/15 1:28 PM
 14. DuraCrete. Modelling of degradation, DuraCrete –
Probabilistic performance based durability design of concrete
structures, EU – Brite EuRam III. Contract BRPR-
CT95-0132. Project 1998; BE95-1347/R4-5, 174 p.
15. Yoon, I.S., Copuroglu, O. and Park, K.B., “Effect
of global climatic change on carbonation progress of
concrete”, Atmospheric Environment, Vol. 41, No. 34,
2007, pp. 7274-7285.
16. de Larrard F. Concrete Mixture Proportioning: A Scientific
Approach. Taylor & Francis; 1999.
17. Gonzalez, J.A., Algaba, J. and Andrade, C., “Corrosion of
reinforcing bars in carbonated concrete”, British Corrosion
Journal, Vol. 15, No. 3, 1980, pp. 135-139.
18. Neville, A., “Chloride Attack of Reinforced-Concrete – an
Overview”, Materials and Structures, Vol. 28, No. 176,
1995, pp. 63-70.
19. Breysse, D., Chaplain, M., Marache, A. and Rodney, E.,
“Simulation of synthetic climate at local scale as a mean
to assess the impact of climate change on infrastructures”,
Civil Engineering and Environmental Systems, Vol. 31,
No. 2, 2014, pp. 165-178.
20. El Maaddawy, T. and Soudki, K., “A Model for Prediction
of Time from Corrosion Initiation to Corrosion
Cracking”, Cement & Concrete Composites, Vol. 29, No. 3,
2007, pp. 168-175.
21. Stewart, M.G. and Mullard, J.A., “Spatial time-dependent
reliability analysis of corrosion damage and the timing of
first repair for RC structures”, Engineering Structures, Vol.
29, No. 7, 2007, pp. 1457-1464.
22. Mullard, J. A. and Stewart, M. G. “Corrosion-induced
cover cracking: new test data and predictive models”, ACI
Structural Journal, Vol. 108, No. 1, 2011, pp. 71-79.
23. Mullard, J.A., Spatial Time-Dependent Reliability Analysis
and Optimisation of Maintenance Strategies for RC
Structures Subject to Corrosion Damage. PhD Thesis, The
University of Newcastle, Australia; 2010.
24. Vanmarcke, E., Random Fields: Analysis and Synthesis.
Cambridge, Massachusetts, MIT Press; 1983.
25. Peng, L., Spatial time-dependent reliability analysis of
carbonation induced corrosion damage to RC structures
under a changing climate and cost-benefit analysis of climate
adaptation strategies. PhD Thesis, The University of
Newcastle, Australia; 2015.
26. AS3600-2009, Concrete Structures. Standards Australia,
Sydney; 2009.
27. AS5100.5, Bridge Design Part 5: Concrete. Sydney:
Australian standard; 2004.
28. McGee, R., “Modelling of Durability Performance of
Tasmanian Bridges”, ICASP8 Applications of Statistics
and Probability in Civil Engineering. 1999. R.E. Melchers
and M.G. Stewart (Eds.). 1: 297-306.
29. Foster, S., Stewart, M. G., Loo, M., Ahammed, M. and
Sirivivatnanon, V., “Calibration of Reinforced Concrete
Structures Standard AS3600 Part I: Statistical Analysis of
Material Properties and Model Error”, Concrete 2013,
16-18 October 2013, Gold Coast, Paper no. 24.
30. Mirza, S.A., Hatzinikolas, M. and MacGregor, J.G.,
“Statistical descriptions of strength of concrete”, Journal
36-47 - Peng.indd 47
View publication stats
of the Structural Division, Vol. 105, No. ST6, 1979, pp.
1021-1037.
31. BRE. Residual Life Models for Concrete Repair – Assessment
of the Concrete Repair Process, Building Research
Establishment; 2003.
32. Canisius, T. and Waleed, N., “Concrete patch repairs
under propped and unpropped implementation”,
Proceedings of the ICE-Structures and Buildings, Vol. 157,
No. 2, 2004, pp. 149-156.
33. Mullard, J.A. and Stewart, M.G., “Life-cycle cost
assessment of maintenance strategies for RC structures in
chloride environments”, Journal of Bridge Engineering, Vol.
17, No. 2, 2012, pp. 353–362.
34. Sommer, A., Nowak, A. and Thoft-Christensen, P.,
“Probability-Based Bridge Inspection Strategy”, Journal of
Structural Engineering, Vol. 119, No. 12, 1993,
pp. 3520-3536.
35. Stewart, M.G., “Spatial variability of damage and
expected maintenance costs for deteriorating RC
structures”, Structure and Infrastructure Engineering, Vol.
2, No. 2, 2006, pp. 79-90.
36. Stewart, M.G., Wang, X.M. and Nguyen, M.N. “Climate
change adaptation for corrosion control of concrete
infrastructure”, Structural Safety, Vol. 35, 2012,
pp. 29-39.
37. Worrell, E., Price, L., Martin, N., Hendriks, C. and
Meida, L.O., “Carbon dioxide emissions from the global
cement industry”, Annual Review of Energy and the
Environment, Vol. 26, No. 1, 2001, pp. 303-329.
38. Wang, X., Nguyen, M., Stewart, M.G., Syme and M.,
Leitch, A., Analysis of Climate Change Impacts on the
Deterioration of Concrete Infrastructure – Part 2: Modelling
and Simulation of Deterioration and Adaptation Options.
Published by CSIRO, Canberra. ISBN 9780643103665;
2010.
39. Val, D. and Stewart, M.G., “Life-cycle cost analysis of
reinforced concrete structures in marine environments”,
Structural Safety, Vol. 25, No. 4, 2003, pp. 343-362.
40. Rawlinsons. Construction Cost Guide. Rawlinsons
publishing, Perth; 2014.
41. RSMeans Engineering Department, RSMeans Facilities
Maintenance and Repair. 2012.
Concrete in Australia Vol 41 No 3 47
1/08/15 1:28 PM