Construction and Building Materials 40 (2013) 775–782

Contents lists available at SciVerse ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat

Carbonation in concrete infrastructure in the context of global climate change:

Development of a service lifespan model
S. Talukdar ⇑, N. Banthia
Department of Civil Engineering, University of British Columbia, Vancouver, Canada

h i g h l i g h t s

" A service life model is developed to evaluate the effects of climate change on concrete infrastructure.
" Climate change will affect the progression of carbonation induced corrosion.
" Exposure conditions have a significant effect on corrosion propagation times in concrete.
" Structures being constructed in 2020–2030 may have to begin taking into account such degradation during design.
" In areas where carbonation induced corrosion is a concern structures may show a service life reduction of 15–20 years.

a r t i c l e i n f o a b s t r a c t

Article history: There is nearly unanimous consensus amongst scientists that increasing greenhouse gas emissions,
Received 23 May 2012 including CO2 generated by human activity, are affecting the Earth’s climate. Increasing atmospheric
Received in revised form 1 October 2012 CO2 emissions will likely increase the rates of carbonation in reinforced concrete structures.
Accepted 9 November 2012
In this paper, the serviceable life, from construction through to cracking due to carbonation induced
Available online 25 December 2012
corrosion of concrete infrastructure is considered in various cities throughout the world. It was concluded
that global climate change will affect the progression and will result in much higher ultimate carbonation
Keywords:
depths in the long term.
Concrete
Carbonation
Ó 2012 Elsevier Ltd. All rights reserved.
Degradation
Modelling
Climate change
Corrosion

1. Introduction model is then applied to a number of major urban centers located

throughout the world.
Increased generation of greenhouse gases including CO2 are
slowly changing the Earths climate. For many key parameters,
the climate system is already moving beyond the patterns of natu- 2. Model formulation
ral variability within which our society and economy have devel-
oped and thrived. Until recently, one overlooked area of research When developing a model to determine the time to end of serv-
is the impact of climate change on our concrete infrastructure. icable life for a specimen due to corrosion, the time frame can be
Increasing atmospheric CO2 emissions will increase the rates of split into two parts: The time to initiation of corrosion (To), and
carbonation in reinforced concrete structures [1]. Talukdar et al. the propogation time (Tcr) (Fig. 1).
[2] developed a numerical model which was able to accurately Carbonation induced microcell corrosion is considered to have
forecast carbonation depths in concrete structures. In this paper, initiated once the carbonation front has reached within 5 mm of
the model developed in Talukdar et al. [2] was extended to assess the embedded rebar [4]. Until now, the authors have focused
the servicable life span of an element, including formation of exclusively on the time to initiation of carbonation induced corro-
cracks due to the initiation of carbonation induced corrosion. The sion in uncracked concrete [1,2]. Here the focus is on the propaga-
tion phase of the model, and the effects climate change would have
on reducing the duration of the propagation phase of the process.
⇑ Corresponding author. Tel.: +1 604 822 6039; fax: +1 604 822 6901. A number of models have been developed to predict the time
E-mail address: [email protected] (S. Talukdar). from the onset of corrosion propagation to cracking and spalling

0950-0618/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.conbuildmat.2012.11.026
776 S. Talukdar, N. Banthia / Construction and Building Materials 40 (2013) 775–782

density of steel (g/mm3), d is the thickness of the porous zone

(10 lm), c is the ratio of the mass of steel to rust (0.622), Wrust is
the mass of corrosion product formed per unit length of reinforcing
bar (g/mm).
A relation between the expansion of the rebar and the critical
pressure to cause cracking is given by Eq. (2) [5]:
2
!
d d 2cfct
Dd ¼ 1þ# ð2Þ
Eeff 4cðc þ dÞ d

where Eeff is the effective modulus of elasticity of concrete (MPa), m

is the Poissons Ratio of concrete (0.18), c is the concrete cover (mm),
fct is the tensile strength of concrete (MPa).
Expressions for the modulus of elasticity of concrete (E) and the
tensile strength of concrete (fct) can be found in ACI 318 [7] and the
formula for the effective modulus of elasticity (Eeff) can be found in
Chernin and Val [5].
Faradays Law is used to calculate the rate of rust production [5]:

Fig. 1. Corrosion initiation and propogation diagram [3]. dW rust IðtÞM

¼ ð3Þ
dt zF

of the concrete cover. Chernin and Val [5] have thoroughly summa- where t is time (s), I(t) is the corrosion current density with respect
rized the development of all existing empirical and analytical mod- to time (A/mm2), M is the molar mass of iron (55.85 g/mol), z is
els for prediction of corrosion induced cover cracking. In particular, the valence number of the reaction (2), F is Faradays Constant
one recently developed model by El Maaddawy and Soudki [6] has (96458 C/mol).
been widely utilized. Essentially, the model uses an analytical A number of studies have looked at the effect of temperature on
thick-walled uniform cylinder which assumes that the generation the corrosion of steel in concrete [8–11]. They all have concluded
of a volume of corrosion products around the corroding steel that the temperature dependency of steel corrosion in concrete
causes an expansion in the diameter of the steel. The expansion ex- follows the Arrhenius Theory.
erts a uniform pressure on the concrete surrounding the rebar, and 1 1Þ
aðTðtÞ T
the concrete will crack when the average tensile stress is equal to IðtÞ ¼ Io o
ð4Þ
the tensile strength of the concrete. They also assume the presence
of a porous zone (approximately 10 lm thick) around the rebar Io is the current flow density at the reference temperature (A/mm2),
which must first be filled by corrosion products before they can T(t) is the temperature at a given time (K), To is the reference tem-
start exerting pressure on the surrounding concrete. A schematic perature (K), a is the activation energy constant (K).
for crack propogation is given in Fig. 2. Parameters such as activation energy, and the reference current
The rigorous mathematical derivation for this model can be flow are generally not constant and may depend on interactions
found in El Maaddawy and Soudki [6]. A simplified description is between concrete resistivity, saturation level, and the cover depth.
as follows: Resistivity and saturation levels will vary according to concrete
The expansion of the diameter of the embedded corroding rebar cover, and will also change as corrosion cracks propagate. An anal-
is given by Eq. (1) [5]: ysis of previous literature shows the relations between depth,
  resistivity, saturation, and current for a corrosion microcell are
2W rust 1 c wide ranging and difficult to relate [12].
Dd ¼   2d ð1Þ
pd qrust qsteel It was identified that the corrosion current flowing through re-
bar is influenced mainly by temperature, concrete saturation, con-
where Dd is the increase in bar diameter (mm), d is the initial bar
crete resistivity, and cover depth. For concrete in service, these
diameter (mm), qrust is the density of rust (g/mm3), qsteel is the
values will tend to change with time, and therefore, generally,
the corrosion current in the rebar will also tend to vary with time.
In order to obtain a reference current flow for Eq. (4), a constant va-
lue for concrete resistivity and cover were assumed during the
propagation phase and inserted into El Maaddawy and Soudki’s
corrosion model [6], to describe the time to cover cracking. Review
of literature indicates no clear concensus exists on what the refer-
ence corrosion current density should be. The value is highly
dependent on cover and concrete resistivity. Even within a partic-
ular geographical area, the corrosion rate within a structure can
vary heavily depending on its exposure conditions, which in turn
affect the microclimate around the embedded rebar. The European
Standard EN 206-1 [13] tries to consider this by assigning a refer-
ence corrosion current as per the exposure class, with a particular
exposure class corresponding to a set of local environmental and
exposure conditions a given part of the concrete structure is ex-
posed to. Pour-Ghaz et al. [8] found that for concrete with resistiv-
ity of 1250 Xm and 60 mm cover depth, w/c = 0.5, the reference
corrosion density (Io) at 20 C was approximately 0.1 lA/cm2. The
Fig. 2. Crack propogation schematic [6]. reference corrosion current density is that which is considered to
 S. Talukdar, N. Banthia / Construction and Building Materials 40 (2013) 775–782 777

correspond to a low level of corrosion in rebar [14], or a XC1-dry mean global temperature increases are obtained using data
exposure class in EN 206-1. reported in the 3rd Intergovernmental Panel on Climate Change
For simplicity, we have considered 0.1 lA/cm2 as the value for (IPCC) Report [18]. Based on this data, using the method originally
the reference corrosion density in Eq. (4). The authors recognize proposed by Bastidas-Arteaga et al. [19] and described in Talukdar
that the value is speculative and considering that concrete cover et al. [1], a simplified climate model was developed which allowed
depths are often <60 mm, would somewhat overestimate propoga- for development of equations for emissions, temperature, and
tion times. In fact, concrete cover depths can often be as low as humidity over the next 100 years for each city in question.
20 mm, and resistivity values are greatly affected by cover mois- Talukdar et al. [2] developed a new carbonation model to pre-
ture contents. Stewart and Ping [15] have used higher values of dict the depth of carbonation in non-pozzolanic, unloaded concrete
Io in an empirical model (0.1–0.431 lA/cm2 at 20 C). These values specimens, taking into account for the first time, time-varying con-
are reported by DuraCrete [16] and based on the different concrete centrations of CO2, temperature and humidity. Essentially, this
exposure classes in EN 206-1 [13]. A justification of using 0.1 lA/ numerical model simultaneously solves equations for the diffusion
cm2 as the value for the reference corrosion density, and an analy- and reaction of CO2 and Ca(OH)2 in concrete based on Ficks Second
sis of what may happen when exposure classes are varied for a city Law. This model was used to determine the time to initiation of
is given in Section 3.2. The value of the activation energy constant carbonation induced corrosion.
(a used in Eq. (4) is 7500 K for a corrosion microcell as per Otsuki et Depending on national code requirements, cover requirements
al. [10]. will vary. Generally, codes require covers ranging from 20 mm to
Therefore, the model can predict the change in the current flow 75 mm depending on the grade of concrete, and the exposure class.
density for a time dependent temperature in Eq. (4). Equations For example, as per ACI 318 [7], concrete slabs, joists and walls ex-
describing the fluctuations in temperature over time for a given ur- posed to earth or weather require a minimum cover of 25 mm. For
ban area can be derived as per Talukdar et al. [1]. Eq. (4) can be a normal strength concrete for a similar exposure class, this is a
substituted into Eq. (3) and then integrated to find the mass of rust typical value of concrete cover used throughout the world. There-
produced over a certain time. This time-dependent expression for fore, in this study, rebar in the reinforced concrete is assumed to
rust formation may then be substituted into Eq. (1), and finally have 25 mm of cover, meaning that carbonation induced corrosion
Eq. (2) may equated with Eq. (1) to solve for the time taken for en- would begin when the carbonation front had reached 20 mm of
ough rust to be generated to generate a pressure which cracks the depth. The cities selected were:
cover concrete. The formulation is represented graphically in Fig. 3.
 Mumbai, India
 London, UK
3. Model results  New York City, USA
 Sydney, Australia
3.1. Structures built in the year 2000  Toronto, Canada
 Vancouver, Canada
To begin with, six cities were selected, and carbonation progress
for a typical non-pozzolanic concrete building, having w/c of 0.5, a Each city is from a different region in the world, and all are con-
compressive strength of 40 MPa and an air content of 6% were sidered to be major urban centers.
modeled. Assuming the building goes into service in the year 2000, the
Based on the Special Report on Emissions Scenarios (SRES) [17], carbonation depth is shown for each city for the three scenarios
three emissions scenarios are considered, the worst case scenario in Figs. 4–9.
(A1FI), the best case scenario (B1), and the control scenario (CO2 From the figures, four initial observations are made. First, we
levels held constant at initial levels). Corresponding forecasts for note an increase in the ultimate carbonation depths of between

Fig. 3. Time to cracking formulation.

778 S. Talukdar, N. Banthia / Construction and Building Materials 40 (2013) 775–782

Fig. 4. Mumbai carbonation depths, constructed in 2000. Fig. 8. Toronto carbonation depths, constructed in 2000.

Fig. 5. London carbonation depths, constructed in 2000. Fig. 9. Vancouver carbonation depths, constructed in 2000.

Table 1
Increase in ultimate carbonation depths, control vs A1FI.

London 27%
NYC 32%
Mumbai 27%
Sydney 31%
Toronto 44%
Vancouver 45%

Fig. 6. NYC carbonation depths, constructed in 2000. climate change does not begin to seriously affect the rate of car-
bonation until about 30 years into the future (in these simulations,
the year 2030). It is only after 30 years that we start to see the car-
bonation depth curves for the A1FI and B1 scenarios begin to seri-
ously divert from the control curves for all six of the cities. Finally,
we notice that over 100 years, for all cities, there is clearly a notice-
able effect of climate change on carbonation depths.
However, it is also important to consider what the actual effects
of climate change will be on the service life of a structure. The
times to initiation and cracking due to carbonation induced corro-
sion for each of the cities are given in Table 2.
Note that for all the combinations considered, the time to end of
service life is in excess of 55 years. Considering that the structures
in question would likely be subjected to deterioration by means
Fig. 7. Sydney carbonation depths, constructed in 2000.
other than carbonation during the same time, the reduction in ini-
tiation time caused by climate change would be one of many fac-
tors affecting service life. Secondly, we also note that there is a
27% and 45% when comparing the control carbonation depth minimal reduction in propagation time (difference between time
against the A1FI carbonation depth for the cities in question (Table to initiation and time to failure) between the control cases, and
1). Second, carbonation is a highly site dependent phenomenon. the climate change cases, for each respective city.
Coastal cities (Mumbai, NYC, Sydney) which exhibit moderate Therefore, in the short term, it is reasonable to conclude that cli-
humidity, and higher temperatures are more prone to corrosion mate change will not significantly affect the durability of our con-
than cooler, wetter cities (London, Toronto, Vancouver). Although crete infrastructure. However, it was noted earlier that the effects
% increases for Mumbai, NYC and Sydney were lower than that of of climate change become significant in fact after approximately
London, Toronto or Vancouver, ultimate carbonation depths of 30 years. Therefore, in Section 3.3, we consider the next ‘genera-
Mumbai, NYC and Sydney were significantly higher than that of tion’ of concrete structures which may be built, using the year
London, Toronto, and Vancouver (Table 2). Third, we notice that 2030 as the time of construction.
 S. Talukdar, N. Banthia / Construction and Building Materials 40 (2013) 775–782 779

Table 2 Table 4
Time to corrosion initiation – constructed in year 2000. Propagation times, Sydney, year 2000 construction, A1FI.

Time to initiation Time to cracking Ultimate carbonation Exposure class Time to Time to Propagation % Reduction in
(years) (years) depth (mm) initiation cracking time propagation
(years) (years) (years) time
London
Control >100 >100 11 XC1-Dry or permanently 54.5 60.1 5.6 –
B1 >100 >100 13 wet
A1FI >100 >100 14 XC2-Wet, rarely dry 54.5 56.2 1.7 69.4%
XC3-Moderate humidity 54.5 57.8 3.3 40.9%
NYC
XC4-Cyclic wet–dry 54.5 55.9 1.4 74.7%
Control 70.6 80.1 25
B1 58.9 66.4 30
A1FI 57.1 63.4 33
Mumbai
Control 62.8 65.8 26
B1 56.6 59.2 30
A1FI 55.3 57.4 33
Sydney
Control 65.0 72.1 26
B1 55.3 61.3 31
A1FI 54.5 60.1 34
Toronto
Control >100 >100 16
B1 >100 >100 18
A1FI 88.2 94.4 23
Vancouver
Control >100 >100 11
B1 >100 >100 14
A1FI >100 >100 16 Fig. 10. Control and A1FI scenarios for year 2030.

3.2. Effect of varying Io

Stewart and Ping [15] has recommended the following mean

values for Io based on different exposure classes for concrete de-
fined in EN 206-1 [13] and the DuraCrete Report [16] (Table 3).
Each exposure class would correspond to a different microcli-
mate for the rebar embedded within the concrete, and therefore,
each exposure class has a difference reference corrosion density.
For example, under dry conditions, corrosion density is far lower
as an insufficient amount of moisture is available for the current
to flow. Even within a city, it is entirely possible that concrete cast
could be subjected to a different exposure class from site to site. All
Fig. 11. Mumbai carbonation depths, constructed in 2030.
the simulations presented in this paper assume a XC1-Dry indoor
exposure class, with indoor humidity being equal to outdoor
humidity, right from initiation, through to propagation and failure.
This would in fact provide conservative estimates by overestimat-
ing values for the time to cracking. Both for carbonation progress,
and corrosion propagation, a cyclic wet–dry condition would accel-
erate the processes. However, as exposure is highly slight specific,
and it is currently impossible to predict moisture conditions to that
level of accuracy over the next 100 years, rather than comparing
different exposure classes in different cities, to maintain unifor-
mity, the dry exposure class was assumed for all simulations.
However, to try and quantify just how much propagation times
may have been underestimated in the simulations, a single set of
simulations were run for a concrete structure built in the city of
Sydney, Australia in the Year 2000, where the difference in propa- Fig. 12. London carbonation depths, constructed in 2030.

gation times was noted for the worst case scenario (A1FI) for each
different exposure class. The results are presented in Table 4: Clearly, the exposure class has a huge effect on the propagation
time for a given city and scenario. In this case, the % Reduction in
propagation time was as high as 74.7% for the C4 exposure case.
Table 3
Mean corrosion densities.
It can be inferred that there would be significant reductions in prop-
agation time for concrete subjected to more severe exposure condi-
Exposure class Mean (lA/cm2) tions in the other cities in this study as well. However, overall, it is
C1-Dry, permanently wet 0.1 important to note that in this case, the propagation time was only
C2-Wet, rarely dry 0.345 about 10% of the initiation time, and therefore, the overall % reduc-
C3-Moderate humidity 0.172
C4-Cyclic wet–dry 0.431
tion in time to cracking would not be too significant. While the
values presented in this paper are still valid, as the methodology
780 S. Talukdar, N. Banthia / Construction and Building Materials 40 (2013) 775–782

Table 5
Time to corrosion initiation – constructed in year 2030.

Time to initiation Time to cracking Ultimate carbonation

(years) (years) depth (mm)
London
Control >70 N/A 9
A1FI >70 N/A 13
NYC
Control 57.2 65.4 23
A1FI 42.4 48.8 31
Mumbai
Fig. 13. NYC carbonation depths, constructed in 2030. Control 53.9 56.6 24
A1FI 41.5 43.8 31
Sydney
Control 52.7 59.2 25
A1FI 40.4 45.4 32
Toronto
Control >70 N/A 14
A1FI >70 N/A 19
Vancouver
Control >70 N/A 10
A1FI >70 N/A 14

perature increases continue as per an A1FI scenario until the year

2030 where they stabilize. The structures would then be subjected
to these stabilized, but elevated levels. The justification for this is
Fig. 14. Sydney carbonation depths, constructed in 2030. that analysis of data over the past decade indicates that the A1FI
scenario is actually the one being followed and possibly exceeded
throughout the world [20], and therefore, we assume that it will
take until at least the year 2030 before any sort of mitigation/sta-
bilization takes effect. In the other ‘Worst Case’ scenario we as-
sume that we allow climate change to continue unabated past
2030 all the way to 2100 as per the A1FI scenario, and consider
the results (Figs. 11–16) (Table 5).
Analysis of the forecasts for structures built in 2030 reveal two
important points. First, Fig. 17 shows that when comparing each
respective scenario, structures built in 2030 carbonate almost the
same amount in 70 years as a structure built in 2000 carbonates
in 100 years. This confirms that the majority of carbonation occur-
Fig. 15. Toronto carbonation depths, constructed in 2030.
ring propagates after the year 2030. Second, there is a noticeable
decrease in the time to cracking for the control scenario vs the
A1FI scenario (Fig. 18).
Results of the simulations for structures constructed in 2030
may cause concern. When comparing the results for NYC, Mumbai
and Sydney, notice that the corrosion initiation times for the Con-
trol and A1FI scenarios in 2030 are worse than even the A1FI sce-
nario in the year 2000. Furthermore, the worst case climate change
scenario for the 2030 cases causes the time to cracking to decrease
to under 45 years. In cities susceptible to carbonation damage
(Mumbai, NYC, Sydney), the average reduction in life span appears
to be approximately 15–20 years due to climate change, would
need to be accounted for at the time of design/construction, as it
would cause an earlier time to repair, which would occur within
Fig. 16. Vancouver carbonation depths, constructed in 2030. the design lifespan of the structure.
In general, throughout North America, the state of our infra-
is still sound, it may be worthwhile to do a probabilistic analysis for
structure is considered to be quite poor. Even before considering
the different exposure classes for the different cities, and determine
the effects of climate change, our infrastructure is not being con-
what is the most predominant exposure class within each city as
structed and maintained properly, and is generally not able to
the data becomes available, to determine just how much faster
achieve its design life as prescribed by building codes. In this paper,
the time to cracking is under more harsh exposure conditions.
it has been shown that climate change would play a role in
decreasing the service life of an idealized structure due to carbon-
3.3. Structures built in the year 2030 ation induced corrosion. If we were to consider factors such as
cracking due to weathering, improper placement, or overloading,
The scenarios were repeated for structures which will be con- along with the presence of chlorides in marine environments, or
structed in the year 2030. However, in this instance we consider environments where deicers are regularly used, it is safe to
only two climate change scenarios which are schematically repre- conclude that climate change effects would reduce service lives
sented in Fig. 10. In the control scenario, CO2 emissions, and tem- even further than what they are today.
S. Talukdar, N. Banthia / Construction and Building Materials 40 (2013) 775–782 781

Fig. 17. Carbonation depths for various cities in the year 2100.

Fig. 18. Time to cracking for various cities.

782 S. Talukdar, N. Banthia / Construction and Building Materials 40 (2013) 775–782

4. Conclusions [2] Talukdar S, Banthia N, Grace J. Carbonation in concrete infrastructure in the

context of global climate change: Part 1, Experimental results and model
development. Cem Concr Compos 2012;34(8):924–30.
(1) Global climate change will affect the progression of carbon- [3] Tutti K. Corrosion of steel in concrete. Report 4-82, Swedish cement and
ation induced corrosion in our concrete infrastructure. We concrete research institute, Sweden; 1982.
[4] Yoon IS, Copuroglu O, Park KB. Effect of global climatic change on carbonation
will see much higher ultimate carbonation depths in the
progress of concrete. Atmos Environ 2007;41(34):7274–85.
long term. [5] Chernin L, Val DV. Prediction of corrosion-induced cover cracking in reinforced
(2) Concrete exposure conditions have a significant effect on concrete structures. Constr Build Mater 2011;25(4):1854–69.
[6] El Maaddawy T, Soudki K. A model for prediction of time from corrosion
corrosion propagation times in concrete. The results for time
initiation to corrosion cracking. Cem Concr Compos 2007;29(3):168–75.
to cracking presented in this paper are for concrete which is [7] ACI Committee 318. Building code requirements for structural concrete (ACI
only subjected to a dry exposure class, and therefore, times 318-08) and commentary (ACI 318R-08). Farmington Hills (MI): ACI; 2008.
to cracking are somewhat overestimated, meaning the [8] Pour-Ghaz M, Isgor OB, Ghods P. The effect of temperature on the corrosion of
steel in concrete. Part 1: Simluated Polarization Resistance Tests and model
effects of climate change on time to crack initiation may development. Corros Sci 2009;51(2):415–25.
be being underestimated. For harsher exposure classes, [9] Hussain RR, Ishida T. Enhanced electro-chemical corrosion model for
propagation times will be shorter. For one scenario reinforced concrete under severe coupled action of chloride and
temperature. Constr Build Mater 2011;25(3):1305–15.
described, the propagation time was overestimated by [10] Otsuki N, Madlangbayan MS, Nishida T, Saito T, Baccay MA. Temperature
approximately 75%. It may be worth conducting a probabi- dependency of chloride induced corrosion in concrete. J Adv Concr Technol
listic analysis for different cities, and different exposure clas- 2009;7(1):41–50.
[11] Jaggi S, Bohni H, Elsener B. Macrocell corrosion of steel in concrete –
ses to estimate the overall further reduction in times to experiments and numerical modelling. Riva di Garda, Italy: Eurocorr 2001;
cracking due to carbonation induced corrosion. 2001.
(3) Although climate change will have unnoticeable effect in the [12] Ostvik JM. Thermal aspects of corrosion of steel in concrete: effect of low
temperature on the resistivity and the cathodic reaction rate. PhD thesis.
near future on durability of concrete structures constructed
Norwegian university of science and technology, faculty of engineering science
in the Year 2000, the real effects of climate change will and technology, department of structural engineering.
become evident after approximately 30 years. Therefore, [13] European Committee for Standardization. Concrete – Part 1: Specification,
performance, production and conformity (European standard EN206-1). CEN,
structures being constructed in 2020–2030 may have to
Brussels, Belgium; 2000.
begin taking into account degradation due to climate change [14] Andrade C, Alonso C. Corrosion rate monitoring in the lab and on site. Constr
when they are being designed. Build Mater 1996;10(5):315–28.
(4) For the worst case scenario, for concrete structures con- [15] Stewart MG, Ping J. Life-cycle cost assessment of climate change assessment of
climate change adaption measures to minimise carbonation-induced corrosion
structed in the year 2030, in areas where carbonation risks. Int J Eng Under Uncertainty: Hazards Assessment Mitigation 2010;2(1–
induced corrosion would be a concern (moderate humidity, 2):35–45.
higher temperatures), for a dry exposure class, we can [16] DuraCrete. DuraCrete-probabilistic performance based durability design of
concrete structures. EU-brite EuRam III. Contract BRPR-CT95-0132. Project
expect structures to begin to show a reduction in serviceable BE95-1347/R4-5; 1998.
lifespan due to climate change of approximately [17] IPCC. Special report on emissions scenarios. Cambridge: Cambridge University
15–20 years, with signs of damage being apparent within Press; 2000.
[18] IPCC. Contribution of working group I to the third assessment report of the
40–45 years of construction. intergovernmental panel on climate change. Cambridge: Cambridge
University Press; 2001.
[19] Bastidas-Arteaga E, Chateauneuf A, Sanchez-Silva M, Bressolette P, Schoefs F.
Influence of weather and global warming in chloride ingress into concrete: a
References
stochastic approach. Struct Saf 2010;32(4):238–49.
[20] Schneider S. The worst case scenario. Nature 2009;458:1104–5.
[1] Talukdar S, Banthia N, Grace J, Cohen S. Carbonation in concrete infrastructure
in the context of global climate change: Part 2, Canadian urban simulations.
Cem Concr Compos 2012;34(8):931–5.