International Journal of Disaster Risk Reduction 1 (2012) 4454

Contents lists available at SciVerse ScienceDirect

International Journal of Disaster Risk Reduction

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

Assessing community resilience to climate-related disasters in

Chennai, India
Jonas Joerin a,n, Rajib Shaw a, Yukiko Takeuchi a, Ramasamy Krishnamurthy b
a
Graduate School of Global Environmental Studies, Kyoto University, Yoshida Honmachi, Sakyo-ku, Kyoto 606 8501, Japan
b
Department of Applied Geology, University of Madras, Guindy Campus, Chennai, India

a r t i c l e in f o abstract

Article history: This article compares the resilience of two communities in Chennai, India, to climate-
Received 9 January 2012 related disasters. The selected communities have similar exposure to natural hazards
Received in revised form (cyclones and river-based oods due to their close proximity to the sea. Both areas are
30 May 2012
predominantly residential. Results from a household survey, assessing the physical,
Accepted 30 May 2012
social and economic resilience of individuals through a Climate-related Disaster
Available online 12 June 2012
Community Resilience Framework (CDCRF), reveal that people living in the vicinity of
Keywords: rivers and canals are at higher risk from impacts (damages on house, diseases) of oods
Community resilience compared to others. However, despite their experience to past ood-related disasters,
Climate-related disasters
they have not been able to enhance their coping capacity due to their limited adaptive
Chennai
capacity. Thus, their resilience is limited to absorb, manage and bounce back future
climate-related disasters (particularly oods). In collaboration with other stakeholders,
mainly the Corporation of Chennai (Municipality), community-driven participatory
solutions are concluded to have benecial effect in enhancing the resilience of
communities to climate-related disasters.
& 2012 Elsevier Ltd. All rights reserved.

1. Introduction effective disaster management systems at national [37]

In recent years, terms such as building or making
cities resilient (e.g. UN 201011 World Disaster Reduction
Campaign: Making Cities Resilient) have gained wide
attraction to denote amongst others, efforts taken by
various stakeholders (authorities, communities, private
sector, etc.) to address issues of urbanisation and climate
change, prevalent in many cities in developing countries
to avoid or minimise impacts from natural hazards [30].
Increasing the capacity of urban systems to manage
disasters corresponds to ambitions stated in the Hyogo
Framework for Action which was adopted by 168 UN
nations in 2005 [36]. Taking decisive action by creating
n
Corresponding author.
E-mail addresses: jonas.joerin@gmail.com (J. Joerin),
shaw.rajib.5u@kyoto-u.ac.jp (R. Shaw),
takeuchi.yukiko.5m@kyoto-u.ac.jp (Y. Takeuchi),
rrkrishnamurthy@gmail.com (R. Krishnamurthy).
and local levels [23] is crucial to generate the platform for
ensuring communities are made disaster resilient.
In this paper, Chennai (India), in particular two com-
munities (Adikesvapuram and South Kodambakkam), is
presented as a case study location which experienced
large population growth (urbanisation) rates (average
1.72% p.a.) over the past decades [9], and is directly
exposed to various natural hazards (cyclones and intense
rainfall) due to its coastal location. Following projections
given by the IPCC, weather events are likely to become
less predictable and also more intense and severe in the
future [17]. This is expected to have serious consequences
for Chennai as its low-lying surface and vicinity to the sea
makes it susceptible to ooding and water-logging in
many parts throughout the city already after sporadic
rainfall events [28].
By adopting the concept of resilience [6,10] to describe
the resilience of a community to climate-related disasters,
the objective is to understand to what extent individuals

2212-4209/$ - see front matter & 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.ijdrr.2012.05.006
 J. Joerin et al. / International Journal of Disaster Risk Reduction 1 (2012) 4454
are prepared and capable to respond to intense natural
hazards. Communities are identied as key actors in
shaping the resilience of an urban system [27,39] there-
fore, their ability to absorb, manage or bounce back
following a disaster [35] is crucial for the functioning of
a city. Based on a city-wide resilience matrix named
Climate Disaster Resilience Index (CDRI) [19], the aspects
of ve dimensions (physical, social, economic, institu-
tional and natural) were assessed. This paper, now pro-
poses a modied Climate-related Disaster Community
Resilience Framework (CDCRF) consisting of physical,
social and economic aspects to disclose the resilience of
communities at the household or micro-level in two
selected areas, Adikesvapuram (Ward 79) and South
Kodambakkam (Ward 131), of Chennai. The following
research questions and specic hypotheses in italics are
attempted to be answered in this article:
 Are households located in the vicinity of natural
hazards (rivers, canals) less resilient?
J Households located in the vicinity of rivers and canals
are less resilient.
 Do affected households learn from climate-related
disasters and enhance their resilience?
J Households learn from a climate-related disaster after
having been affected and neither enhance their resilience.
 How can community resilience to climate-related dis-
asters be enhanced?
To answer these questions and hypotheses, a household
survey and interviews with community leaders have been
undertaken in the two before mentioned communities in
Chennai.
2. Community resilience to climate-related disasters
Although the concept of resilience has its origin
from the ecological eld of science [16], works from
various authors [1,2,7,13] have shown the linkage of
resilience to explain socio-ecological systems in which
communities respond to disturbances or disasters
Community Condition (Capacity)
Disaster
Pre-Disaster
Coping
Capacity
Fig. 1. Community disaster resilience framework (graphical).
45
within a natural environment. Accordingly, concepts of
community resilience to disasters were evolved and
applied by various scholars [6,35,1,3,22,26,29,34,38]
who further include political and economic aspects in
the conceptualisation of community resilience, apart from
the social and natural context in which communities are
embedded in. The technical understanding of community
resilience is described as the communitys ability to
absorb, manage and bounce back after a disaster. Simi-
larly, community resilience is explained [10] in form of a
life-cycle where the post-disaster phase is characterised
by the ability of a community to strengthen its coping
through adaptive capacity following a disaster [31]. This
allows communities to enhance their responsive (coping
and recovery) capacity (resilience) to manage future
disasters (Fig. 1).
However, all these theoretical aspects of resilience
depend on the actual context in which a community is
located. For an individual person, in particular, physical,
social and economic aspects may decide to what
extent he/she can respond to a disaster [18]. For example,
the extent dening how well an individual person is
integrated (social capital) into his/her community and
can count on support from neighbours during a disaster
[21] is one of many aspects of social resilience.
Thus, enhancing resilience of communities requires
them to be better prepared to disasters [10,5]. In this
context, the term adaptive capacity [14] describes the
ability of socio-ecological systems, including commu-
nities, to learn and/or improve their capacity to
manage a disturbance (disaster) either through reaction
following a disaster experience, or in a proactive
manner where a future stress or change is anticipated
before it occurs. This follows the argument that higher
experience of disasters, mainly oods, enhances the pre-
paredness of people based on a learning effect that would
take place among people after they experience such
events [25]. This notion shall be integrated into the
CDCRF. Finally, to enhance community resilience, social
learning helps a community to increase its awareness,
skills and ability to confront a future disaster in a
collective approach [1,24].
Post-Disaster
Community Disaster
Evidence of Adaptive Capacity
Resilience Scenarios:
Capacity equal as before Disaster C - Ideal
B
Not recovered from Disaster A
Time
Recovery
Capacity
46 J. Joerin et al. / International Journal of Disaster Risk Reduction 1 (2012) 4454
However, before applying aspects of resilience into a
practical CDCRF, the previous thoughts (e.g. learning from
a disaster) are conceptualised and integrated into a
graphical representation emphasising the theoretical
approach of the later proposed CDCRF.
This community disaster resilience framework (Fig. 1) has
been developed to highlight the capacity or condition of a
community during different time phases before, during and
after a disaster. Thereby, the existence of adaptive capacity is
crucial to enhance a communitys ability (resilience) to
confront and manage future disasters. Accordingly, three
scenarios can evolve due to a disaster:
 A: the communitys coping capacity is not sufcient to
absorb a hazard. As a result, the community suffers
from damages and losses and its condition becomes
weakwhich is the evidence that the hazard could not
be absorbed by the community and the consequence is
a disaster. During the recovery phase a community
can get back to its pre-disaster condition (middle way),
but also either not fully recover (bottom line) or learn
and adapt (ideal) from the disaster (upper line), and
enhance its resilience (capacity).
 B: the communitys coping capacity is not sufcient to
absorb a hazard. As a result, the community suffers from
damages and losses (disaster) and its condition becomes
weaker than before the hazard occurred. Unlike scenario
A, the community cannot exercise recovery action and
thus, remains vulnerable to future natural hazards.
 C: ideal scenario: the communitys coping capacity is
sufcient enough to absorb a natural hazard (rainfall,
storm, earthquake, etc.) and therefore, no damages or
losses are recorded. Although, the community may not
have been affected by disaster impacts, it may still
exercise a certain amount of adaptation (adaptive
capacity) to strengthen its coping capacity for poten-
tially more intense hazards expected to occur in the
future; otherwise, the capacity remains the same as
before the natural hazard took place.
The graphical representation of community disaster
resilience emphasises that the levels of a communitys
coping and adaptive capacity are crucial in absorbing,
managing and bouncing back from a disaster event.
Particularly, the ability or capacity of a community to
adapt to disasters is crucial in enhancing their own
resilience.
Box 1Dimensions and parameters of the Climate-related Disaster Community Resilience Framework (CDCRF).
Dimensions Physical Social
Parameters (P) Electricity Health
Water Education and awareness
Sanitation and Solid waste Social capital and preparedness
2.1. Climate-related community disaster resilience
framework
To link the theoretical understanding of resilience into
practice, an earlier developed Climate Disaster Resilience
Index (CDRI) [19,20], consisting of physical, social, eco-
nomic, institutional and natural dimensions quantita-
tively assesses how urban systems, in particular
communities, may cope to climate-related disasters, such
as oods, landslides, cyclones, etc. In this paper, this
framework is further modied to be applied at the
community level respectively, household or micro-level
of an urban area. Box 1 shows the identied dimensions
and parameters of this framework. In order to integrate,
the above mentioned key elements of resilience into, for
example, physical parameters such as electricity, water or
sanitation and solid waste, the current supply of electri-
city, water, etc., is assessed among households of a
community to understand their physical capacity (coping
capacity). Furthermore, since the selected case study sites
in Chennai are located in hazard-prone areas, this frame-
work anticipates that certain households may have
experienced climate-related disasters in the past.
Moreover, the presented CDCRF aims to understand
quantitatively whether affected households take action
(adaptive capacity) to enhance their resilience as a reac-
tion to experienced disasters, or not. As a result, resilience
in the following case studies is understood as a combina-
tion of coping and recovery (adaptive) capacity.
The linkage between the community disaster resili-
ence framework and the CDCRF is, as follows: to achieve
scenario C (Fig. 1) in the community disaster resilience
framework, the parameters from the three dimensions of
the CDCRF need to be met at a high level, for example, full
knowledge among households on impacts of climate
change in the parameter of education and awareness
(social dimension). Although, in reality, not all parameters
are expected to score high among all households and
therefore, they are asked whether they take action (adap-
tive capacity) to enhance their knowledge to be better
prepared to respond to future climate-related disasters.
This aspect of taking action has particular importance for
those households who were affected by climate-related
disasters as their resilience (coping capacity) obviously
was not sufcient to absorb the disaster. Thus, it is
expected that they have a particular incentive to take
action in order to avoid future impacts from natural
Economic
Income and employment situation
Household assets
Finance and saving
 J. Joerin et al. / International Journal of Disaster Risk Reduction 1 (2012) 4454
hazards. In the case, where disaster affected households
take action, they follow scenario A (upper line), according
to the community disaster resilience framework (Fig. 1)
which is again a desired scenario if they demonstrate
adaptive capacity that would increase their resilience
beyond the level they had before the disaster event.
To conclude shortly, the CDCRF has the objective to
address the need raised by various scholars [10,15,24] to
develop models that can quantitatively assess the resi-
lience of communities to disasters. Therefore, the poten-
tial of this approach is to better understand and disclose
the strengths and weaknesses of communities to with-
stand climate-related disasters.
The limitations of this framework include: rstly, the
number of parameters in the physical, social and eco-
nomic dimension may be further increased, however, to
understand the resilience or ability of people to respond
to disasters, the selected parameters are expected to be
sufcient; secondly, this approach assesses the resilience
of communities during non-disaster times, implicating
that attributes of community resilience, to absorb, man-
age and to bounce back following a disaster are projected
based on their current capacity; nally, the presented
resilience framework neglects potential involvement from
other stakeholders (e.g local government) to support
communities during a disaster. Instead, the focus is put
entirely on how individual households are likely to
manage climate-related disasters and whether they have
learned from previous experiences through enhancing
their coping capacity during adaptation processes.
3. Case studies from two communities of Chennai
The selected case study sites (see Fig. 2), Ward 79
(Adikesvapuram) and Ward 131 (South Kodambakkam),
are both located along major rivers of Chennai and are
Fig. 2. Location of selected Wards 79 and 131 in Chennai.
47
predominantly residential. However, Ward 79 belongs to
the historical part of Chennai built during the 19th
century whereas, Ward 131 is considered a new devel-
oped area since it has been urbanised in the latter part of
the 20th century. The area sizes of Ward 79 and 131 are
0.45 km2, respectively, 1.26 km2. The population size in
2011 was 18,190 in Ward 79 and 58,400 in Ward 131.
As a result, the population densities in both wards
were beyond 40,000 people per square kilometre. These
population densities are considerably high compared to
Chennais average of 26,762 [9]. The disaster proles of
both wards are similar not only because they are located
at a river, their relatively close proximity to each other
makes them equally prone to impacts of cyclones in form
of intense rainfall events causing potential sporadic ood-
ing in Chennai during the post-monsoon period from
October to December [28,11]). Although Chennai suffered
from past cyclone-related oods in 2005, 2008 and 2010,
residents from Ward 131 claim that major oods occurred
in the years 1976, 1983, 2006, 2008 and 2009, mostly
after dams of lakes in the up-stream of the Adyar River
were opened-up to avoid overowing after intense rain-
falls. Thus, the ooding was to some extent man-made. In
addition to big events, yearly small-scale ooding lead to
temporary water-logging in both wards and perceived by
some residents as ood events, as these type of events
have as well potential to cause damages to people and
infrastructures.
3.1. Data collection and approach
To understand the resilience of communities in Ward
79 and 131, a household survey was undertaken during
MayJune 2011. In order to get a spatially equal repre-
sentation of household responses in each ward, every
third house, a combination of systematic and random
48 J. Joerin et al. / International Journal of Disaster Risk Reduction 1 (2012) 4454
sampling technique was adopted, was selected to be
surveyed. Furthermore, a statistically representative
number of samples were targeted to get enough house-
hold responses that would allow in an aggregated form to
reveal the view of the entire community. Thus, a sufcient
number of household perspectives/responses are assumed
to generate an average view of the conditions of a larger
number of people who are equally familiar with the local
culture, traditions and contexts that characterise the term
community. The idea that a single ward can sense a
community follows Chavis and Wandersman [8] who
distinguish between place-bound (with clear boundaries)
and interest-based groups with no clear bound-
aries. Hence, in this study the selected two wards are
both functioning as two sample or cluster communities
situated within an equal cultural context, provided a sta-
tistically sufcient number of household views are gen-
erated to reveal the collective character of these two
selected communities.
Thus, the sampling frame applied in both communities
was, as follows: every third residential house was selected
and within that, one household chosen. However, since
one residential house has in many cases more than one
household, the selected sample size was less than 1/3 of
all household in both selected communities. Accordingly,
the sample size for Ward 79 totalled at 336 where 195 or
58% of the households responded. This response rate
again highlights that a total 8.4% of all households of
Ward 79 took part in the survey. In the case of Ward 131,
the sample size was 1,112 and the response rate was
59.4% or 661 households. Accordingly, 8.7% of all house-
holds of Ward 131 took part in this survey. Although the
survey targeted all households in both wards, it must be
noted that residents from slum areas were excluded from
this study due to difculties in accessing them. The
omission of slum households should, however, only have
minor impacts to the results as their number is negligible,
particularly in Ward 131. Nonetheless, in Ward 79 slum
households are directly located at the river which makes
them highly vulnerable to potential river-based ooding
and thus, their views would have been supportive to the
overall ndingsthis is a limitation of this study. Addi-
tionally, the collected number of 8.4% in Ward 79 respec-
tively 8.7% in Ward 131 is clearly lower than the
percentage that could be assumed if every third house
(see above) would represent just one household. How-
ever, in an urban area, several households may live in a
single house which is why the sample sizes and collected
responses are lower than an expected one-third of all
households. Households were approached during all days
of a week usually between 4 and 7 pm to conduct face-to-
face interviews by lling-up a standardised questionnaire
which was analysed using various statistical tools avail-
able in Microsoft Excel.
In addition to the household surveys in Ward 79 and
131, interviews have been conducted with local commu-
nity leaders in a focus group discussion who are repre-
senting households in community-based groups, such as
Residential Welfare Associations (RWAs) and women
associations. The focus group discussions were organised
during a stakeholder workshop which took place at a
public primary school in Ward 131 in August 2011 where,
apart from local residents functioning as community
leaders, representatives from the local government parti-
cipated to provide a platform for exchanging and sharing
concerns and ideas. The key aim of this workshop was to
validate the survey results and to develop potential path-
ways for solutions that would enhance the resilience of
urban communities of Ward 131 (Ward 79 was not
considered in the focus group discussion). In the focus
group discussions, two groups were formed where one
group consisted of ve and the other of three local
community leaders. However, since the collected data
(drawing of resilience enhancing pathways) from the
focus group discussions is not directly relevant to the
CDCRF study, the presentation of detailed results is not
provided. Nonetheless, the conduction of focus group
discussions was necessary to validate the ndings made
later in the results and discussion sections.
3.2. Overall response
At rst, results from the household surveys in Ward 79
and 131 point out that ood was in more than 92% of all
responses the climate-related disaster type that affected
households. Thus, other climate-related natural hazards
such as storms, droughts or heat waves are hardly
experienced by residents in the two selected neighbour-
hoods. Secondly, 28.7% in Ward 79 and 43.6% of all
households in Ward 131 claim to have experienced a
climate-related disaster, mostly ood, in the past, but
have not necessarily been affected. In an attempt to
correlate responses from households who answered to
have experienced a climate-related disaster and those
whose houses were damaged in the past, using Pearson
correlation method in Microsoft Excel, show relatively
high correlation coefcients of r 0.82 in Ward 79 and
r 0.84 in Ward 131. In other words, 78.6% of all house-
holds who claim to have experienced a climate-related
disaster in Ward 79 and 85.1% in Ward 131 state that
their houses were also damaged due to such events.
Hence, disaster experience from oods is to a large extent
perceived by residents in form of damaged houses.
In order to facilitate the analysis of the household
survey, the following results concentrate rst on compar-
ing the households whose houses were damaged and
those who claim no past damages to their houses.
Localising damaged and non-damaged houses (see
Figs. 3 and 4), using ArcGIS, highlights that residents
located in the vicinity of waterbodies (rivers and canal)
are more susceptible from ooding compared to residents
living in the inner parts. This is the case in both selected
communities. This may not be surprising, however, pro-
vides an ideal setting to examine whether these house-
holds are becoming more prepared or resilient to face
future oods or not, which aims to address the second
research question and hypothesis. Finally, initial results
point out that more households in Ward 131 (37.1%)
claim damages on their houses compared to Ward 79
(23.6%), but are the residents who experienced disasters
from the Ward 131 less resilient?
 J. Joerin et al. / International Journal of Disaster Risk Reduction 1 (2012) 4454 49

Fig. 3. Location of damaged and not damaged houses due to climate-related disasters in the past in Ward 79.

Fig. 4. Location of damaged and not damaged houses due to climate-related disasters in the past in Ward 131.

By looking at the overall characteristics of results (see
Table 1), similar patterns are visible between the two
communities based on two-sided t-tests measuring the
statistical signicance between two variables. The results
show that more people who always lived (no migration)
in Ward 131 claim disaster experiences that damaged
their houses. The reason for this is two-fold: on one hand,
they have lived in average longer in the neighbourhood
compared to those who migrated, but on the other are not
willing to relocate away from their living location. This
latter statement cannot be statistically proven, but is
based on various discussions from focus group discus-
sions held at a workshop in August 2011 and talks to
residents living in such areas during the time when the
50 J. Joerin et al. / International Journal of Disaster Risk Reduction 1 (2012) 4454

Table 1
Overall characteristics of HouseHolds (HH) in Wards 79 and 131 (bold values are signicantly different p o 0.05).

Key characteristics Ward 79 Ward 131

House not damaged House damaged House not damaged House damaged
(n 149) in % (n 46) in % (n 416) in % (n 245) in %

Sex
Male 44.3 28.3 36.3 29.8
Female 55.7 71.7 63.7 70.2

Age
1524 12.1 10.9 20.7 11.0
2564 72.5 82.6 67.3 67.3
More than 65 15.4 6.5 12.0 21.6

Marital status
Single 16.1 13.0 17.3 11.0
Married 83.9 87.0 82.7 89.0

Household size
Average # (not in %) 4.6 4.9 4.4 4.3

Migration
From Tamil Nadu State, but outside of 7.4 8.7 34.4 24.9
Chennai
From outside of Tamil Nadu State 4.0 2.2 5.8 2.9
Within Chennai 42.3 39.1 37.0 31.0
No migration 46.3 50.0 22.8 41.2

Mother Tongue
Tamil 80.5 93.5 89.4 91.0

Languages (speaking)
Tamil 98.7 95.7 99.3 99.2
English 53.0 13.0 58.9 40.8

Languages (reading)
Tamil 98.7 97.8 98.6 98.4
English 54.4 13.0 59.9 41.2

survey was carried out. In both wards, residents who have
experienced a disaster (mainly ood), have signicantly
lower English abilities (speaking and reading) compared
to those who claim no damage to their houses due to
climate-related disasters (oods). This is an indication
that less privileged residents, in terms of school educa-
tion, are also less resilient.
3.3. Physical resilience
The physical resilience (Table 2) reveals that house-
holds who experienced damaged houses due to climate-
related disasters suffer more from power cuts (in both
communities) than others and are in particular less
satised with the electricity provision in Ward 79. Other
ndings emphasise on the aspect that residents with
disaster experiences due to damaged houses are possibly
less privileged. For example, they have in both commu-
nities signicantly lower provision of water storage tanks.
Moreover, the garbage for these residents in Ward 131 is
collected among fewer households and also less fre-
quently. In terms of adaptive capacity, residents who
have experienced damages to their houses due to disas-
ters are not taking more action compared to those who
claim no disaster damages to their houses. This is likely
due to the relatively high provision of basic services,
meaning electricity, water, sanitation and solid waste
management services are supplied at a high level.
To summarise briey, households with disaster experi-
ences are not more resilient in relation to physical aspects
compared to others who do not claim damages to their
houses due to a climate-related disasters.
3.4. Social resilience
Results from the household surveys (Table 3) in the
two selected communities in Chennai highlight that
residents with disaster experience (house damaged) are
signicantly more affected by waterborne and vector-
borne diseases compared to those who have not been
impacted by damages on their houses due to climate-
related disasters.
Unlike the nding that residents with disaster experi-
ence (house damaged) have lower English skills, their
perception about knowing the impacts of climate change
and also their awareness on how to respond to climate-
related disasters is not signicantly different compared to
households who claim no damages to their houses due to
climate-related disasters. Regarding aspects of social
capital and disaster preparedness, households with dis-
aster experience (house damaged) in Ward 79 are on one
hand more actively involved in Community-based Orga-
nisations (CBOs) than residents who have not experienced
climate-related disaster damages to their houses, but on
the other, they are signicantly less prepared to face such
disasters as fewer households possess sufcient basic
 J. Joerin et al. / International Journal of Disaster Risk Reduction 1 (2012) 4454 51

Table 2
Physical resilience of HouseHolds (HH) in Wards 79 and 131 (bold values are signicantly different p o 0.05).

Parameters Key characteristics Ward 79 Ward 131

House not damaged House damaged House not damaged House damaged
(n 149) in % (n 46) in % (n416) in % (n 245) in %

Electricity HH supplied with Yes 100 100 99.5 99.2

electricity?
Power cuts: More than once a 36.2 63.0 34.5 49.8
day
Once a day 63.1 37.0 65.5 49.0
Less than once a 0.7 0.0 0.0 0.0
day
HH satised with No 36.9 69.1 37.3 42.0
electricity provision?
if no, HH take action? No action (n 55) 60.0 (n 18) 88.9 (n155) 69.0 (n 103) 67.0

Water HH supplied with piped No 6.7 21.7 6.0 9.4

water
If yes, HH has water Yes (n 139) 84.9 (n 36) 58.3 (n391) 80.6 (n 222) 69.4
storage tank?
HH satised with water No 5.4 6.5 7.5 13.9
provision?
If no, HH take action? No action (n 8) 87.5 (n 3) 66.7 (n31) 38.7 (n 34) 35.3

Sanitation and HH connected to a Closed drainage 98.0 95.7 99.8 98.8

solid waste drainage system?
HH satised with No 13.4 34.8 5.0 10.6
sanitary provision?
If no, HH take action? No action (n 20) 55.0 (n 16) 56.3 (n21) 95.2 (n 26) 84.6
HHs garbage collected? No 5.4 2.2 2.4 13.5
If yes, how often: Once every day 86.5 86.7 84.0 75.5
Once every three 9.2 4.4 14.0 22.2
days
Less than once 4.3 8.9 2.0 2.3
every three days
HH satised with waste No 15.4 8.7 11.3 23.3
management?
If no, HH take action? No action (n 23) 91.3 (n 4) 100.0 (n47) 76.6 (n 57) 61.4

emergency equipments (medical kit, ash light, food, etc.)
to manage climate-related disasters.
Looking at aspects of adaptive capacity, in none of the
social parameters, residents with disaster experience
(house damaged), in both communities, show efforts to
take action to increase their resilience to climate-related
disasters. Since there is no result indicating that residents
who were affected by a past climate-related disaster took
specic action to become better prepared to a future
climate-related disaster, they show insufcient adaptive
capacity and thus, remain less resilient than others.
However, this conclusion shall not mean that all house-
holds who have not been, or not yet, affected by such a
disaster would be automatically more resilient. Since
based on ndings from the spatial analysis (Figs. 3 and
4), disaster affected households are pre-dominantly living
in more hazardous areas compared to others and may also
be economically poorer than others, see next section on
economic resilience, they are more vulnerable to natural
hazards.
However, the results also point out that a learning
effect cannot be found in form of taking action to become
better prepared. This may be due to nancial constraints,
especially among urban poor, however, certain variables
such as participating in CBOs or the willingness to provide
support to other people during future disaster events
would not require money. Although, this explanation
shall again not imply that the social resilience of less
privileged households is only denominated by their
nancial capacity since the social status and integration
into the wider society are likely to be further factors
hindering them to actively increase their social capital
and thus, ability to enhance their resilience.
3.5. Economic resilience
The economic resilience shows patterns (Table 4) that
households with disaster experience (house damaged)
derive more income from the informal sector which means
during disaster times the income generation may be less
secured compared to employment in the formal sector.
Accordingly, it may not surprise that among them less
people are satised with their employment situation, which
is likely to provide irregular income in the informal sector.
Regarding, the adaptive capacity of these households, it is
limited as signicantly less action is taken by those living in
Ward 131 to apply for jobs or search for new income
sources. Regarding household assets, residents with disaster
experience (house damaged) have less access to computer
with Internet compared to those who have not been
affected by damages on their houses due to climate-related
disasters. This exemplies again that residents with disaster
52 J. Joerin et al. / International Journal of Disaster Risk Reduction 1 (2012) 4454

Table 3
Social resilience of HouseHolds (HH) in Wards 79 and 131 (bold values are signicantly different p o0.05).

Parameters Key characteristics Ward 79 Ward 131

House not damaged House damaged House not damaged House damaged
(n 149) in % (n 46) in % (n 416) in % (n245) in %

Health Affected by WD in the past ve Yes 10.1 45.7 13.9 33.9

years?
If yes, take action? No (n 15) 13.3 (n 21) 4.8 (n 58) 6.9 (n83) 2.4
action
Affected by VD in the past ve Yes 21.5 54.3 24.5 38.8
years?
If yes, take action? No (n 32) 3.1 (n 25) 0.0 (n 102) 8.8 (n95) 9.5
action

Awareness Knowledgeable about impacts of No 55.7 69.6 52.6 55.1

climate change?
If no, take action? No (n 83) 77.1 (n 32) 90.6 (n 219) 72.1 (n135) 76.3
action
Knowledgeable on how to respond Not 75.8 73.9 72.8 79.6
to a CRD? fully
If not fully, take action? No (n 113) 65.5 (n 34) 88.2 (n 303) 54.5 (n195) 54.9
action

Social capital and Participation in CBOs? Yes 8.7 23.9 30.5 33.9
preparedness HH satised with the services No 26.2 23.9 12.3 24.5
provided by Municipality?
If no, take action? No (n 39) 94.9 (n 11) 100.0 (n 51) 86.3 (n60) 83.3
action
HH prepared for a CRD (disaster Yes 43.6 13.0 36.1 32.7
supply kit)?
HH capable to provide voluntary Yes 49.7 45.7 63.2 58.8
support during a CRD?

Table 4
Economic resilience of HouseHolds (HH) in Wards 79 and 131 (bold values are signicantly different p o 0.05).

Parameters Key characteristics Ward 79 Ward 131

House not damaged House damaged House not damaged House damaged
(n 149) in % (n 46) in % (n416) in % (n245) in %

Income and HH gets income from Not fully 63.1 89.1 57.9 56.3
employment formal sector
If not fully, from where? Informal sector (n 94) 85.1 (n 41) 92.7 (n 241) 71.4 (n138) 87.0
Formal and (n 94) 14.9 (n 41) 7.3 (n 241) 28.6 (n138) 13.0
informal sector
Satised with No 22.1 41.3 14.2 26.5
employment
If no, take action? No action (n 33) 51.5 (n 19) 68.4 (n 59) 11.5 (n75) 78.5

Household HH equipped with HH Yes 100.0 97.8 99.8 98.4

assets assets
If yes, with which Radio 36.2 (n 45) 17.8 (n415) 27.0 (n241) 20.3
devices? TV 100.0 (n 45) 100.0 (n415) 99.3 (n241) 99.6
Computer 28.2 (n 45) 4.4 (n 415) 35.7 (n241) 26.1
Internet
Key HH assets Yes 69.1 67.4 87.7 87.8
protected from CRD

Finance and House insured against No 97.3 100.0 95.9 93.9

savings CRD?
If no, take action? No action (n 145) 93.8 97.8 (n399) 94.9 (n230) 89.1
HH saves money for No 83.9 91.3 82.0 89.8
future CRD?
If no, get support? Get no support (n 125) 51.2 (n 42) 42.9 (n 341) 31.1 (n220) 43.2

experience (house damaged) are less privileged to afford money or insure their house to climate-related disaster is
services, such as Internet at houselower coping capacity. either equally low or again lower in Ward 131 (adaptive
Aspects of nance and savings point out that the ability of capacity) compared to non-affected (house damaged)
residents with disaster experience (house damaged) to save households.
 J. Joerin et al. / International Journal of Disaster Risk Reduction 1 (2012) 4454
To summarise, households with disaster experience
(house damaged) are less resilient in all the three para-
meters of economic resilience compared to residents who
have not been affected by climate-related disasters (house
damaged).
3.6. Summary of ndings from household surveys
In an attempt to quantitatively assess the resilience of
communities to climate-related disasters in two neigh-
bourhoods of Chennai with similar characteristics (expo-
sure to natural hazards and land-use), the ability of
households to learn from disaster experiences is mea-
sured as low. Findings from the household surveys in
Wards 79 and 131 show that households who experi-
enced disasters (house damaged) are less privileged, for
example, in terms of level of education, access to Internet
from home or having jobs in the formal sector to get
regular income compared to those who have not experi-
enced intense disasters (damage of house). Hence, more
wealthy and educated people are more resilient to cli-
mate-related disasters than others. Moreover, households
located in the vicinity of waterbodies are more likely to
experience intense impacts (damage on houses) from
ood-related disasters compared to those living further
away from such hazards.
The adaptive capacity of the two communities to
take action is limited, following climate-related disasters,
which would allow them to increase their resilience,
based on the results from the three assessed dimensions
(physical, social and economic). Therefore, previously
affected (damage on house) households are likely to
experience again future ood-related disasters as their
resilience has not improved following past experiences.
Although one may argue that those households who were
previously affected are also economically less privileged,
other aspects detached from economic ones, for example,
participation in CBOs or the willingness to provide volun-
tary support during future disasters do not show that
disaster affected households would take action to make
them better prepared, or more resilient, against future
climate-related disasters. Thus, a learning effect from the
disaster event cannot be seen among affected households.
The reasons for this absence of learning effect are beyond
the abilities of the CDCRF to provide evidence, but are
denitely needed to be investigated in further research.
4. Discussion
Assessing the climate-related disaster resilience of
communities through the CDCRF allows integrating the-
oretical aspects about how resilience to disturbances
(disasters) ought to be understood among communities.
Moving away from qualitative descriptions of community
resilience [33], the CDCRF intends to highlight strengths
and weaknesses of a community in a quantitative
approach that would ease comparisons between different
communities in a specied urban area. Furthermore, in
line with aspirations from various scholars [12,18],
the outcomes of the CDCRF aim to reveal the current
capacities of communities to trigger contextualised
53
community-led planning and implementation of local
responses in form of actions to enhance the community
resilience to disasters. This is crucial, particularly in the
context of Indian cities where local governments usually
have a strong power to control processes within commu-
nities [4]. Thus, it is not surprising that ndings, in
particular, in the physical dimension show that house-
holds take little action to enhance their resilience since
traditionally urban services are provided by governmental
authorities. As a result, the assessed communities in
Chennai expect the local government (Corporation of
Chennai) to take the lead in improving their resilience
and therefore, make them highly dependent on other
stakeholders about their destiny. Although local residen-
tial welfare and ladies associations exist in both assessed
communities of Chennai, these CBOs have so far little
inuence in decision-making processes in their neigh-
bourhoods, according to results arising from focus group
discussions and individual interviews. Also, young people,
in particular, are few in numbers to participate in these
organisations due to other priorities, such as job or family
issues. The fact that different CBOs, largely middle-class
run organisations, already exist in both assessed commu-
nities of Chennai offers the potential to give greater
mandate to such organisations to bring forward the needs
of the local people. Hence, the traditionally top-down
driven local municipality of Chennai (Corporation of
Chennai) may take more into account local concerns in
processes of planning and decision-making.
Linking the ndings from the CDCRF to the conceptual
ideas developed in the community disaster resilience
framework, results emphasised limited learning abilities of
disaster affected (house damaged) households who are
mostly located in the vicinity of water bodies. In other
words, the adaptive capacity of affected households did not
increase their community condition to a higher level. Thus,
households who were affected mostly by oods are likely
again be hit in the future by strong natural hazards.
Despite the nding that the economic condition of disaster
affected households is lower compared to non-affected
households, other aspects detached from nancial needs
neither show efforts in form of taking action or willingness
of households to enhance their resilience to climate-related
disasters. Thus, the rst hypothesis dened in the introduc-
tion is conrmed by the results from the CDCRF study. In
contrast, the results from the CDCRF cannot accept the
second hypothesis that affected households would learn
and enhance their resilience following a disaster. Finally,
since disaster affected households obviously did not with-
stand to natural hazards in the past and further do not show
adaptive capacities higher than non-affected households, it
must be concluded that their resilience is lower compared to
others.
To conclude, the applied CDCRF reveals that enhanced
efforts need to be undertaken by various stakeholders to
address problems at the micro-level of a community.
These problems, which require to be further identied
through additional research, are then required to be
reected in local decision-making processes to make
communities more resilient to future climate-related
disasters.
54 J. Joerin et al. / International Journal of Disaster Risk Reduction 1 (2012) 4454
Acknowledgements
The authors would like to thank the Japan Society for
the Promotion of Science (JSPS) and the Global Centre of
Excellence (GCOE) Human Security Engineering (HSE)
Program for their nancial support for this study. The
authors would also like to thank Mr. Mathavan from the
local municipality (Corporation of Chennai) for his kind
support in providing data and information to this article.
References
[1] Adger WN. Social and ecological resilience: are they related?
Progress in Human Geography 2000;24(3):34764.
[2] Adger WN, Arnell NW, Tompkins EL. Successful adaptation to
climate change across scales. Global Environmental Change
2005;15:7786.
[3] Allen KM. Community-based disaster preparedness and climate
adaptation: local-capacity building in the Philippines. Disasters
2006;30(1):81101.
[4] Baud I, Nainan N. Negotiated spaces for representation in Mumbai:
ward committees, advanced locality management and the politics
of middle-class activism. Environment and Urbanization
2008;20(2):48399.
[5] Berkes F, Colding J, Folke C. Navigating socialecological systems:
building resilience for complexity and change. Cambridge: Cam-
bridge University Press; 2003 (393 p.).
[6] Bruneau M, Chang SE, Eguchi RT, Lee GC, ORourke TD, Reinhorn
AM, et al. A framework to quantitatively assess and enhance the
seismic resilience of communities. Earthquake Spectra 2003;19(4):
73352.
[7] Carpenter S, Walker B, Anderies JM, Abel N. From metaphor to
measurement: resilience of what to what? Ecosystems 2001;4:76581.
[8] Chavis DM, Wandersman A. Sense of community in the urban
environment: a catalyst for participation and community develop-
ment. American Journal of Community Psychology 1990;18(1):
5581.
[9] Chennai Metropolitan Development Authority (CMDA). Second
master plan for Chennai metropolitan area, 2026, CMDA, vol. 3.
Chennai; 2008. 303 p.
[10] Cutter SL, Barnes L, Berry M, Burton C, Evans E, Tate E, et al. A
place-based model for understanding community resilience to
natural disasters. Global Environmental Change 2008;18:598606.
[11] Drescher A, Glaser R, Pfeiffer C, Vencatesan J, Schliermann-Kraus E,
Glaser S, et al. Risk assessment of extreme precipitation in the
coastal areas of Chennai as an element of catastrophe prevention, 8.
Forum DKKV/CEDIM: disaster reduction in climate change 15./
16.10.2007. Karlsruhe; 2007. 5 p.
[12] Few R. Flooding, vulnerability and coping strategies: local
responses to a global threat. Progress in Development Studies
2003;3(1):4358.
[13] Folke C. Resilience: the emergence of a perspective for social
ecological system analyses. Global Environmental Change 2006;16:
25367.
[14] Gallopin GC. Linkages between vulnerability, resilience, and adap-
tive capacity. Global Environmental Change 2006;16:293303.
[15] Godschalk DR. Urban hazard mitigation: creating resilient cities.
Natural Hazards Review 2003;4(3):13643.
[16] Holling CS. Resilience and stability of ecological systems. Annual
Review of Ecological Systems 1973;4:123.
[17] Intergovernmental Panel on Climate Change (IPCC). Summary for
policymakers. In: Solomon S, Qin D, Manning M, Chen Z, Marquis M,
Averyt KB, Tignor M, Miller HL, editors. Climate Change 2007: the
physical science basis, Contribution of working group I to the fourth
assessment report of the Intergovernmental Panel on Climate Change.
Cambridge: Cambridge University Press; 2007. p. 118.
[18] Jabeen H, Johnson C, Allen A. Built-in resilience: learning from
grassroots coping strategies for climate variability. Environment
and Urbanization 2010;22(2):41531.
[19] Joerin J, Shaw R. Mapping climate and disaster resilience in cities.
In: Shaw R, Sharma A, editors. Climate and disaster resilience in
local governments. London: Emerald Publishers; 2011. p. 4761.
[20] Joerin J, Shaw R, Takeuchi Y, Krishnamurthy R. Action-oriented
resilience assessment of communities in Chennai, India. Environ-
mental Hazards, http://dx.doi.org/10.1080/17477891.2012.689248,
in press.
[21] Kadushin C. Too much investment in social capital? Social Net-
works 2004;26:7590.
[22] Klein RJT, Nicholls RJ, Thomalla F. Resilience to natural hazards:
how useful is this concept? Environmental Hazards 2003;5:3545.
[23] Kyoto University (KU) and United Nations International Strategy
for Disaster Reduction (UNISDR). A guide for implementing the
hyogo framework for action by local stakeholders. Accessible at:
/http://www.unisdr.org/les/13101_ImplementingtheHFA.pdfS;
2010.
[24] Lopez-Marrero T, Tschakert P. From theory to practice: building
more resilient communities in ood-prone areas. Environment and
Urbanization 2011;23(1):22949.
[25] Mishra S, Suar D. Do lessons people learn determine disaster
cognition and preparedness? Psychology and Developing Societies
2007;19(2):14359.
[26] Paton D, Johnston D. Disasters and communities: vulnerability,
resilience, and preparedness. Disaster Prevention and Management
2001;10:2707.
[27] Pelling Mark. The vulnerability of cities: natural disasters and social
resilience. London: Earthscan; 2003 (256 p.).
[28] Revi A. Climate change risk: an adaptation and mitigation agenda
for Indian cities. Environment and Urbanization 2008;20(1):
20729.
[29] Rose A. Economic resilience to natural and man-made disasters:
multidisciplinary origins and contextual dimensions. Environmen-
tal Hazards 2007;7:38398.
[30] Satterthwaite D, Huq S, Pelling M, Reid H, Romero Lankao P.
Adapting to climate change in urban areas: the possibilities and
constraints in low- and middle-income countries. Human settle-
ments climate change and cities discussion series 1. London: IIED;
2007. 112 p.
[31] Smit B, Wandel L. Adaptation, adaptive capacity and vulnerability.
Global Environmental Change 2006;16(3):28292.
[33] Tanner T, Mitchell T, Polack E, Guenther B. Urban governance for
adaptation: assessing climate change resilience in ten Asian cities.
Brighton: Institute of Development Studies; 2009 (47 p.).
[34] Tobin GA, Whiteford LA. Community resilience and volcano hazard:
the eruption of Tungurahua and evacuation of the faldas in
Ecuador. Disasters 2002;26(1):2848.
[35] Twigg J. Characteristics of a disaster-resilient community: a gui-
dance note. Beneld: DFID Disaster Risk Reduction Interagency
Coordination Group; 2007 (38 p.).
[36] United Nations International Strategy for Disaster Reduction
(UNISDR). Hyogo framework for action 20052015. Accessible at
/http://www.unisdr.org/2005/wcdr/intergover/ofcial-doc/L-docs/
Hyogo-framework-for-action-english.pdfS; 2005.
[37] United Nations International Strategy for Disaster Reduction
(UNISDR). Words into action: a guide for implementing the hyogo
framework. Accessible at /http://www.unisdr.org/les/594_10382.
pdfS;2007.
[38] Wamsler C. Bridging the gaps: stakeholder-based strategies for risk
reduction and nancing for the urban poor. Environment and
Urbanization 2007;19(1):11542.
[39] Wisner B, Blaikie P, Cannon T, Davis I. In: At risk: natural hazards.
Peoples vulnerability and disasters. 2nd ed.London: Routledge;
2004 (320 p.).