Climate Change Vulnerability FCM

Working Paper

248

CLIMATE CHANGE VULNERABILITY
ASSESSMENT FOR SUSTAINABLE
LIVELIHOODS USING FUZZY COGNITIVE
MAPPING APPROACH

Pramod K. Singh and Abhishek Nair

The purpose of the Working Paper Series (WPS) is to provide an
opportunity to IRMA faculty, visiting fellows, and students to sound out
their ideas and research work before publication and to get feedback
and comments from their peer group. Therefore, a working paper is to
be considered as a pre-publication document of the Institute.

Institute of Rural Management Anand
Post Box No. 60, Anand, Gujarat (India)
Phones: (02692) 263260, 260246, 260391, 261502
Fax: 02692-260188  Email: [email protected]
Website: www.irma.ac.in

June 2013

Climate Change Vulnerability Assessment for Sustainable Livelihoods using
Fuzzy Cognitive Mapping Approach
1

Pramod K. Singh and Abhishek Nair

2

Abstract
In order to move towards climate resilient pathways local vulnerability assessments based
on peoples’ perception are required where there are large data constraints. In this paper
we have developed a new climate change vulnerability index for sustainable livelihoods to
estimate the vulnerability of poor agro-pastoralists in the Bhilwara and Tonk districts of
Rajasthan, India. The existing studies of livelihood vulnerability assessments often do not
capture the dynamics and functioning of interconnected systems. We used a semiquantitative fuzzy cognitive mapping approach to understand interconnected interactions
occurring in the climate-human-environment interaction space. This approach enables the
quantification of peoples’ perceptions of perturbations to a given climate exposure and
their adaptation mechanisms in relative terms. This was combined with a sustainable
livelihood approach for an improved understanding of asset classes sensitive to climate
change and assets serving as adaptive capacities. Data collection involved group
discussions during which 35–38 groups drew cognitive maps in each district on climaterelated perturbations and adaptation mechanisms. Data from cognitive maps were
aggregated using a composite index and differential vulnerabilities were compared.
Results indicate that Tonk may be more vulnerable to climate change due to lower
adaptive capacity than Bhilwara over all three seasons. Asset classes most susceptible to
harm are natural assets while physical assets serve as a strong adaptive mechanism
against climate change in both the districts.
Keywords: Vulnerability, adaptation, fuzzy cognitive mapping, livelihoods

1

Associate Professor, Institute of Rural Management, Anand-388001, Gujarat, India
E-mail : [email protected]

2

Research Associate, Institute of Rural Management, Anand-388001, Gujarat, India
E-mail: [email protected]

1

1. INTRODUCTION
The climate-change issue is global and long-term, involving complex interactions between
climatic, socio-economic, environmental, technological, institutional, and political
processes. With livelihoods being increasingly affected by climate change; Cruz et al.
(2007) and Stern (2007) have stated that it adversely impacts people’s health, safety, and
livelihoods with the poor people in poor countries expected to suffer first and foremost.
Agriculture-based livelihoods are among the sectors most impacted by climate change.
Rural people in the Asia-Pacific region face serious loss of high-value agricultural lands,
significant land degradation, and reduced crop yields owed to increasing temperatures
and varying precipitation patterns (IFAD and The Global Mechanism 2009).
Climate change is already threatening agriculture-based livelihoods in Rajasthan along
with escalating competition for resources including water, land and energy from nonagricultural sectors concomitant with a rising food demand (Government of Rajasthan
2011). Climate change projections for Rajasthan under the A1B SRES scenario for the
near-term of 2021–2050 project temperature increases by around 2–2.5ºC
(Gopalakrishnan et al. 2011) and overall rainfall reduction could lead to water shortages
(Singh et al., 2010). Several studies have explicated the seriousness of recurring droughts
in the region with far-reaching and deep impact on local livelihoods. (Sivakumar and
Kerbart 2004; Government of Rajasthan 2011). In order to increase the resilience of
livelihoods based on agriculture there is a need to understand local vulnerability of the
poor for improved adaptability to climate change.
Today, climate change vulnerability analysis incorporates several aspects of livelihood
frameworks (Schroter et al. 2005; Eakin and Bojorquez-Tapia 2008; Park et al. 2009;
Orencio and Fujii 2013; Li, Y et al., 2013). Studies are increasingly documenting the local
knowledge vis-à-vis climate change and climate variability, its impacts, and locally
developed knowledge and practices of resource use (Hennessey et al. 2007; Green et al.
2010; Leonard et al. 2013). Throughout the world, peoples’ perceptions concerning
impacts of climate change and adaptation strategies are under discussion with
governments and intergovernmental and non-governmental organisations (Hunter and
Leonard, 2010; Petheram et al. 2010; Leonard et al. 2013). Most vulnerability assessment
studies have been conducted on a macro scale in India (Obrien et al. 2004, ISNC 2008;
INCAA 2011). Local vulnerability assessments capable of capturing the dynamics of an
interconnected interacting system need to be identified against a set of indicators that
gauge vulnerability as an outcome of various factors. Since there is a lack of data and
scientific knowledge of vulnerability on a local scale, we proposed presenting the
vulnerability of two districts (Bhilwara and Tonk) in Rajasthan based on peoples’
perceptions through a multi-step fuzzy cognitive mapping (FCM) approach. This
vulnerability assessment yields relative values. FCM approach is a semi-quantitative tool
(Kosko 1986; Ozesmi and Ozesmi 2004; Vliet et al. 2009; Reckien et al. 2010) that
enables capturing the interconnected interactions in a climate-human-environment system
to depict vulnerability.
2

It is important to understand communities and their forms of adaptation, as each
adaptation option and decisions about governance and institutional arrangements are
underpinned by a set of values associated to certain worldviews that shape what is
considered to be worthwhile adaptation action and what is not (Spence et al. 2011; Wolf
and Moser 2011; Jacob et al. 2010; Leonard et al. 2013). In this study the FCM approach
has been used for the documentation of peoples’ understanding of perturbations due to
climate change and adaptation mechanisms. The remainder of the paper provides a brief
overview of the evolution of vulnerability analysis in the context of climate change.
Proceeding on to discuss the FCM approach and building the climate change vulnerability
index for sustainable livelihoods we have outlined our research findings structured into
analysis of the cognitive maps, tacit knowledge of climate related perturbations and
adaptation mechanisms, and the vulnerability of the region.

1.1. VULNERABILITY ANALYSIS: EVOLUTION AND PRACTICE
Climate change brandishes interactions between the environment and humans rendering
it a complex socio-ecological system. Vulnerability arising from climate change is often in
a continuous state of flux; both the biophysical and social processes that shape local
conditions and the ability to cope are themselves dynamic (O’Brien et al. 2005).
Vulnerability includes exposure, sensitivity to perturbations, and the capacity to adapt
(Adger 2006). Vulnerability assessments have evolved through time; the risk hazard (RH)
model of vulnerability assessment is well-represented in the work of the Intergovernmental
Panel on Climate Change. These approaches tend to consider negative outcomes as
functions of both biophysical risk factors and the 'potential for loss' of a specific exposed
population (Eakin and Luers 2006). Further investigation into the RH models reveals the
inadequacy of capturing the dynamics of a system functioning. Vulnerability in RH models
has been discussed as a linear process and as an outcome of factors that begins with
exposure and determines sensitivity or potential impacts (Burton et al. 1978; Cutter 1996;
Mustafa 1998; Brooks et al. 2005).
Vulnerability assessments that subsequently evolved into the 'Pressure and Release'
(PAR) models as defined by Blaike et al. (1994), is a function of the perturbation, stressor,
or stress, and the vulnerability of the exposed unit. This model emphasises the underlying
causes of disaster and the social production of risk (Blaike et al. 1994; Turner et al. 2003a;
Eakin and Luers 2006). Besides the dimensions of RH models, the PAR model
incorporates facets of political economy and political ecology. These perspectives focus on
the political dimensions of vulnerability while highlighting social inequities and points of
conflict within societies (Eakin and Luers 2006). The PAR model does not address issues
pertaining to resilience and the system’s ability to bounce back from perturbation in a
coupled human-biophysical system. Vulnerability assessments need to be comprehensive;
the PAR model does not encompass interactions between human-ecological systems.
According to Turner et al. (2003a) the PAR model seems insufficient for the broader
concerns of sustainability science. Ecological resilience enters vulnerability analysis and
3

views it as a coupled human-biophysical system in which both systems interact and are
interconnected with one another. Eakin and Luers (2006) claim that vulnerability is seen as
the dynamic property of a system in which humans are constantly interacting with the
biophysical environment. Ecological resilience is used to characterise a system’s ability to
rebound, absorb change and disturbance, and yet maintain a reference state after
disturbance (Hollings 1996; Turner et al. 2003a). Incorporating ecological resilience is
critical to vulnerability analysis since it tries to capture 'real vulnerability'.
The amalgam of ecological resilience and political economy with hazards has thus
addressed the concerns of vulnerability. Previous studies have confirmed that assessing
vulnerability in a system is often difficult and intricate (Cutter et al. 2003; Eakin and Luers
2006) with no single approach established yet (UNEP 2002; Orencio and Fujii 2013). A
diverse set of approaches, methods, and tools allow for a systematic integration and
investigation of interactions between humans and their social and biophysical environment
in the light of exposure, sensitivity, and adaptive capacity (Turner et al. 2003a; Turner et
al. 2003; O’Brien et al., 2004; Vincent, 2004; Brooks et al. 2005; Polsky et al. 2007; Eakin
and Bojorquez-Tapia 2008; Hahn et al. 2009). The key differences between studies
include scale, selection methods, group and aggregate indicators, and methods for
displaying results (Hahn et al. 2009). In recent years, an indicator and metrics based
analysis has been linking system dynamics quantitatively. For example, Luers et al.
(2003b) tried to link biophysical and economic attributes to vulnerability outcomes. Global
environmental change demands research indicator based analysis in policy circles for
identifying, ranking, or distinguishing between units of analysis for informing resource
allocation or targeting support programmes or other inter entions ( akin and Bo o r ue Tapia 2008). Vulnerability assessments, which use indicators, have been conducted at
national (Cutter et al. 2000; Moss et al. 2001; O’ Brien et al. 2004; Brooks et al. 2005) and
household levels (Alwang et al. 2001; Vincent 2007; Eakin and Bojorquez-Tapia 2008;
Hahn et al. 2009). These vulnerability assessments link attributes to outcomes without
capturing the true essence of an interacting system. The main problem with these
vulnerability assessments is the lack of interconnectedness between attributes.
1.2. Livelihood Vulnerability Analysis in the Climate Context
Political ecology, ecological resilience, and sustainable livelihood approaches have all
proved effective in identifying vulnerable systems. The sustainable livelihood approach
allows identifying the sensitivity of assets and entitlements and critical assets for coping
with and adapting to risks, and the linking of livelihood strategies to opportunities and
constraints of the broader institutional and biophysical environment (Kelly and Adger 2000;
Eakin and Luers 2006). Several studies have incorporated or combined the sustainable
livelihood approach in vulnerability analysis, which provides general lessons concerning
the varying implications of socioeconomic and environmental shocks. Besides, it is useful
for understanding the advantages of certain strategies and combinations of assets for
addressing risks (Vasquez-Leon et al. 2003; Eakin 2005; Eakin and Bojorquez-Tapia
2008; Li et al. 2013; Gautham et al. 2013; Orencio and Fujii 2013).
4

Hahn et al. (2009) developed a climate change vulnerability index based on a set of
indicators and collected data using structured questionnaires to understand vulnerability of
livelihoods to climate change. The indicators portraying vulnerability are rigid and do not
clearly showcase interactions in the system. The indicators have been aggregated and the
IPCC vulnerability formula deployed to decide the vulnerability of a region. In a similar
vein, exists a study by Eakin and Bojorquez-Tapia (2008), in which the interactions have
been better represented but changes that occur due to climate change and the
'interconnectedness' of relationships have not been captured. Relevant to livelihood and
the climate system is a high degree of interaction and interconnectedness between
climate-human-environment systems that needs to be understood for a better assessment
of vulnerability.
While the existing vulnerability analysis looks at climate-human-environment interactions
they are rigid, linear, and/ or hierarchical giving a general description of vulnerability. In an
interaction space there may be interconnected pathways. Though these vulnerability
assessments are able to capture some interactions between the climate-humanenvironment systems, the dynamics and function of interconnected systems remain
uncaptured. The above indicator methods of analysis do not capture relations between
concepts. These indicators do not, at the same time, adequately address the dynamic
nature of vulnerability in its manifestation and causes (Leichenko and O’Brien 2002; Vogel
and O’Brien 2004; akin and Luers 2006). Indices are limited in their application due to
high subjectivity in the selection of variables and their relative weights, availability of data
at various scales, and difficulty of testing or validating different metrics (Luers et al. 2003;
Orencio and Fujii 2013).The concept of 'interconnectedness' is pivotal for understanding
that the true functioning of a dynamic system is not captured in these vulnerability
assessments. According to previous studies determining vulnerability in a system is often
difficult and intricate (Cutter et al. 2003; Eakin and Luers 2006; Orencio and Fujii 2013).
We emphasise, therefore, the imperativeness of understanding interconnected
relationships in an interaction space, during vulnerability analysis, in order to capture the
true essence of vulnerability.
Climate change adds complexity to the livelihood vulnerability analysis. The sustainable
livelihood (SL) framework as proposed by Chambers and Conway (1992), Scoons (1998),
and Department of International Development (1999) looks at five types of household
assets. Aiming to render sustainable livelihood frameworks more comprehensive we
added another asset class to this framework—the organisational asset. The new climate
change vulnerability index for sustainable livelihood combines the fuzzy cognitive mapping
(FCM) approach with the proposed sustainable livelihood framework to explain the
community’s understanding of climate-related perturbations (sensitivity) and adaptation
mechanisms against climate stressors. The perceptions, information, views, opinions, and
observations revealed during group discussions were captured using an FCM approach
and explained through the proposed sustainable livelihood framework. This is to showcase
5

assets providing resilience against shocks and those sensitive to a particular exposure/
shock.
Our approach differs from previous vulnerability assessments as it tries to document
ulnerability based on people’s perceptions and uantifying its weights using the FCM
approach. This approach explains interconnected relationships occurring in a dynamic
climate-human-environment interaction space.
2. METHODOLOGY
2.1. The Fuzzy Cognitive Mapping Approach
The FCM approach, a powerful tool in environmental decision-making and management,
tries capturing the functioning of a complex system based on people’s understanding. We
employed the FCM approach to understand various climate-related perturbations on
human, economic, and ecological systems along with factors creating resilience against
shocks in a climate-human-environment interaction space. FCMs tries depicting
interconnected relationships between variables. Each variable represents characteristics
of the system and interconnections between these variables better depicts the dynamics
and interactions within the system (Papageorgiou and Kontogianni 2012). FCMs indicate
the importance of causal functions between variables of the system using quantitative
numbers given by interviewee groups (Kosko 1986, Özesmi and Özesmi 2004). The
variables could be physical, measurable entities and qualitative, aggregate or abstract
ideas including ethical, political or aesthetic issues (Reckien et al. 2010). These models
have been developed to look at climatic, environmental, ecological problems.
Local knowledge renders FCMs valuable in supplementing and complementing scientific
data, particularly where human behaviour needs to be understood and problems are
complex, where many parties are involved and no straight-forward solution is deducible
(Reckien et al. 2010; Papageorgiou and Kontogianni 2012). FCMs, semi-quantitative in
nature, are used extensively in soft knowledge domains (Ferarrini, 2011; Vliet et al., 2009;
Kok, 2000). The FCM map consists of a set of nodes representing variables connected
through links denoted by arrows (Özesmi and Özesmi 2004; Harary et al. 1965).
Relationships between these variables are labelled with positive or negative polarities and
weights assigned. These polarities describe the arrangement of the system with the
positive sign indicating a direct relationships and the negative sign an inverse one. The
weights assigned to the connections vary from 0 to 1, defining the strength of the
relationship between the variables (Vliet et al. 2009; Ozesmi and Ozesmi 2004; Reckien et
al. 2010). FCM explains complex interactions occurring in a dynamic system. Therefore, to
facilitate ease in depicting the interconnected interactions that occur, FCM are graphically
represented through Cognitive Interpretive Diagrams (CIDs). FCM aids the visualisation of
interrelated variables affecting one another while representing feedback; it is a graphical
representation in the form of CIDs (Özesmi and Özesmi 2004).

6

Recent studies have indicated the potential of fuzzy cognitive mapping regarding
investigating the role of feedback mechanisms in coupled climate–human–environment
systems (Murungweni et al. 2011; Reckien et al. 2010). This approach is flexible, avoiding
frequently under-identification and non-convergence of solutions, a problem with most
studies involving structured questionnaires. Detailed descriptions of the approach and its
application may be found in Özesmi and Özesmi (2004); Isak et al. (2009) and
Papageorgiou and Kontogianni (2012). For this study we adopted a partial multi-step FCM
approach, suggested by Ozesmi and Ozesmi (2004), for collecting data required for the
climate change vulnerability index for sustainable livelihoods. It involves the following
steps:
i.
Drawing single fuzzy cognitive maps by each community group
ii.
Determining if the sample size is adequate using Monte-Carlo simulations
iii.
Coding the cognitive maps into adjacency matrices
iv.
Augmenting single cognitive maps and then adding them to form a stakeholder
social cognitive maps
2.2. Obtaining individual cognitive maps and creating social cognitive maps
Looking at the livelihoods of agro-pastoral communities and dealing with sustainable
livelihoods we have realised that the poor are most affected by climate change. We
selected marginal (> 0.3 hectare) and small scale (0.3–0.6 hectare) farmers with few cattle
as our community groups. We divided men and women separately to understand genderbased perceptions of climate-related perturbations and coping strategies. First, we
obtained near consensus on changes in summer and winter temperatures and
precipitation patterns over the past six years to a decade. We chose 6–10 years as the
recall window for climate change having realised respondents’ probable inability to
accurately report climate change of an earlier period. Other studies too have used a 6–10
year recall window to understand climate and weather events based on peoples'
perceptions (Fowler 2002; Hahn et al. 2009). We demonstrated the construction of a fuzzy
cognitive map to the community. The latter were divided into 4-5 member groups. Each
group drew fuzzy cognitive maps based on the group’s perception of impacts and coping
strategies. Fuzzy cognitive maps were drawn for each season. The driving questions
behind obtaining fuzzy cognitive maps are listed below:
i.
ii.
iii.
iv.

What changes in the climate have you observed in the past decade over the
summer, winter, and rainy seasons?
How have your various livelihoods components been affected due to changes in
summer and winter temperatures and rainfall patterns in the last decade?
What are the consequential impacts arising from the direct impacts?
What are the coping mechanisms used to respond to these impacts?

The cognitive maps were aggregated, analysed via networks statistics, and refined to
remove biases. Analyses of complex cognitive maps were conducted through graph
theoretical indices. Each fuzzy cognitive maps may be mathematically represented
7

through adjacency matrices in the form of A (D) = [aij] (Harary et al. 1965; Özesmi and
Özesmi (2004); Reckien et al. 2010) where variable vi is listed in the vertical axis and the
variables vj are listed in the horizontal axis to form a square matrix. When a connection
exists, a value between -1 and 1 are coded into the square matrix. Each individual
cognitive map is coded into an augmented matrix and normalised. A negative–positive–
neutral calculus is used to compute compound values for differing directions in logical
structures.
2.3. Sampling
Monte Carlo techniques have been used to determine sample size through accumulation
curves. Average accumulation curves of the total number of maps versus the number of
new variables were added per map to determine how the variables accumulated. FCMs
were created with different groups until the representative population was sampled
sufficiently. Monte Carlo simulations were conducted using STATA to acquire the
accumulation curve. We gathered 38 fuzzy cognitive maps from the Bhilwara district and
35 from the Tonk district. The average accumulation curve stabilised at 32 maps as
shown in Figure 1.

2.4. Calculating sensitivity and adaptive capacity
The sustainable livelihood approach enables the quantification of sensitivity and adaptive
capacity under various asset classes. As mentioned earlier, we combined the FCM
approach with a modified sustainable livelihood framework to understand sensitive asset
classes and those serving as adaptive capacities. Data collected from the fuzzy cognitive
maps provide variables, arising from the functioning of a complex system. The fuzzy
cognitive maps give a clear indication of variables sensitive to climate change and those
serving as adaptive capacities and separated. Calculating vulnerability is a complex
process in an interaction space with interconnected variables. While calculating sensitivity
of a certain variable every corresponding variable (driver) causing its sensitivity is
accounted for. Similarly, every adaptation (influence of resilience) reducing the sensitivity
of a particular variable is accounted for calculating adaptive capacity. Another important
interaction includes adaptation practices that increase sensitivity, also considered under
sensitivity analysis. Sensitivity and adaptive capacity analysis are conducted separately
using a weighted balance approach (Sullivan et. al 2002) where each sub-component
contributes equally to the overall index although having a varied number of subcomponents. For sensitivity analysis the sensitivity of each variable is calculated first using
simple averages as shown in Eq. (1).

S vi 

1 n
 Di
n x1 x

8

Eq. (1)

where

S vi

is the average sensitivity of ith variable,

Di are the drivers of sensitivity of ith

S
variable and n is the number of drivers of each variable ( vi ) that causes sensitivity. Once
sensitivity of each variable is obtained the sensitivity of each assets’ class is calculated
using balanced weighted averages as shown in Eq. (2).
i m

AC j 

 ND S
i

i 1

vij

 ND

Eq. (2)

where AC j is the sensitivity of jth assets class, S vij is the average sensitivity of the ith
variable in jth asset class, NDi is the number of drivers of the ith sensitivity variable and m
is the number of variables in jth asset class. Finally the overall sensitivity is calculated using
Eq. (3).
i n

OSs 

 NV AC
i 1

j

j

 NV

Eq. (3)

where OS s is the overall sensitivity for each season, AC j is average sensitivity of the jth
asset class, NV j is the number of variables in the jth asset class and n is the number of
assets

Similarly first adaptive capacity of each variable is calculated using Eq. (4)

Avi 

1 n
 IRi
n x1 x

Eq. (4)

where, Avi is average adaptive capacity of ith variable, IRi is the influence of resilience
by ith variable and n is the number of variables influenced by each variable ( Avi ) that
increases resilience. Once adaptive capacity of each variable is obtained the adaptive

9

capacity of each assets’ class is calculated using balanced weighted averages as shown
in Eq. (5).
i m

AC x 

 ND A
i 1

i

vix

 ND

Eq. (5)

where, AC x is the adaptive capacity of xth assets class, Avix is the average adaptive
capacity of the ith variable in xth asset class, NDi is the number of influenced variables by
the ith adaptive capacity variable and m is the number of variables in xth asset class.
Finally the overall adaptive capacity is calculated using Eq. (6)
i m

OACs 

 NV AC
i 1

x

x

 NV

Eq. (6)

where, OACs is the overall adaptive capacity for each season, AC x is average adaptive
capacity of the xth asset class, NVx is number of variables in the xth asset class and n is
the number of assets.
2.5. Illustrations for calculating sensitivity and adaptive capacity
For illustrative purposes detailed calculations of sensitivity and adaptive capacity for one of
the districts in Rajasthan in summer season has been calculated as shown in Annexure
Tables A1 and A2. Sensitivity and adaptive capacity of the population is calculated for
each season separately using the above mentioned formulae (Eq. 1 to 6).

2.6. Calculating exposure
Exposure or factors causing sensitivity to livelihood security include changes in summer
and winter temperatures and precipitation patterns. Exposure is calculated based on 20
years’ data collected from the Climate Research Unit database. Exposure is calculated for
each season using Eq. (7).

10

Es 

Os  Omin 

Omax  Omin 

Eq. (7)

where, E s is the exposure of each season, Os is the ten-year moving point averages of
increase in temperature and decline in rainfall during the latest year recorded. Omin is the
minimum value and Omax is the maximum value during the observation period.

2.7. Calculating the Climate Change Livelihood Vulnerability Index for
Sustainable Livelihoods
Defining vulnerability is crucial for calculating the vulnerability index. Today there are
common terms in place that help define vulnerability comprising exposure, sensitivity and
adaptive capacity. According to Adger (2006), vulnerability includes exposure and
sensitivity to perturbations or external stresses and adapting capacity. Exposure refers to
the exposure of a system of interest to stimuli acting on that system (Preston and StaffordSmith 2009). According to the IPCC (2001), sensitivity is the degree to which a system is
affected directly or indirectly, either adversely or beneficially, by climate-related stimuli,
directly and/or indirectly. We have defined adaptive capacity as the ability of a system to
manage sensitivity to climatic influences. Having calculated exposure, sensitivity, and
adaptive capacity we combined the three using Eq. (8) to calculate climate change
livelihood vulnerability index for sustainable livelihoods. Vulnerability of population is
calculated for each season separately using Eq. (8).

CCLVI SLs  Es  OSs  OACs 

Eq. (8)

Where CCLVI SL is the climate change vulnerability index for sustainable livelihood for
each season, E s is the exposure OS s is the overall sensitivity and OACs is the overall
adaptive capacity calculated for each season respectively. The adaptation practices
observed at the local level only contribute towards reducing sensitivity not exposure. In the
context of exposure, various climate models have projected an increase in temperature
and change in precipitation patterns in the Indian sub-continent (INCAA 2010; ISNC 2008).
Exposure to a climate hazard is not reduced owing to current adaptation practices. We
have, thus, subtracted adaptive capacity from sensitivity and multiplied it with exposure.
Here, we assume that as exposure increases, the need to adapt will increase as
adaptation is a process for existence. The CCLVISL lies in the range of -1 to 1 where -1 is
the least vulnerable, 0 vulnerable, and +1 is the most vulnerable.
11

3. AREA OF STUDY
The climate change vulnerability assessment for sustainable livelihoods was piloted in two
districts of Rajasthan, namely Bhilwara and Tonk. Bhilwara is located between 25°00' and
27°50' North latitude and between 74°03' and 75°25' East longitude, while Tonk lies
between 25°41' and 26°34' North latitude between 75°07’ and 76°19’ ast longitude. Area
of Bhilwara is 10455.26 km2. It is the 12th largest district in Rajasthan with a population of
2.41 million and a population density of 230 persons/km2 (Census of India 2011). Its
climate is characterised by dry summers and severe winters with maximum temperatures
reaching 45°C and minimum 7°C. Average rainfall in the district is 79.60 cm (Government
of Rajasthan 2011). Tonk, the 18th largest district in Rajasthan, has a land area of 7194.36
km2 with a population of 1.42 million and population density of 198 persons/km2 (Census
of India 2011). Its climate is generally dry except during the short south-west monsoon
season. The maximum temperature during summer is 45oC and the minimum temperature
in winter 8oC. Average annual rainfall in the district is around 61.36 cm (Government of
Rajasthan 2011). Crop cultivation and livestock rearing are major livelihoods of rural
population in Bhilwara and Tonk districts.
Of the 33 districts in Rajasthan Bhilwara ranks 15th in the Human Development Index
(0.633) (Directorate of Economics and Statistics 2008). Tonk, however, has a lower HDI of
0.571 ranking 24th in the state (Directorate of Economics and Statistics 2008). Bhilwara’s
total literacy rate stands at 62.71% (male: 77.16%, female: 47.93%) and Tonk’s at 62.48%
(male: 78.27% female: 46.01%) literacy rate (Census of India 2011). Bhilwara has a 7.13%
forest area to the total geographical area of the district whereas Tonk’s forest cover
stretches only upto 3.83% of its total geographical area (Directorate of Economics and
Statistics 2008). About 64% of the entire population of Bhilwara and 68.7% of the entire
population of Tonk are engaged in agriculture. The net irrigated area to net sown area in
Bhilwara is 30.25%; in Tonk this is 39.70% (Census of India 2011).
4. RESULTS AND DISCUSSIONS
4.1. Fuzzy Cognitive Maps: Interpretation of Networks using Graph Theory
A total number of 38 cognitive maps were obtained from Bhilwara during summer and the
rains, while 32 cognitive maps were analysed during the winter season. In Tonk, 35
cognitive maps were analysed during summer and 38 cognitive maps during winter and
the rainy season. Table 1 gives an overview of the stakeholder groups interviewed.
Table 1: Overview of interviews conducted
Region

No. of
Villages

Valid maps/
Stakeholder group

Increased
Summer
Temperature
38

Increased Winter
Temperature

Decreased
Rainfall

Bhilwara

9

Marginal Farmers

32

38

Tonk

9

Marginal Farmers

26

26

26

Small Scale Farmers

9

12

12

12

4.1.1. Analysis of individual cognitive maps
Maps obtained from marginal farmers of Tonk have the maximum number of variables per
map followed by small scale farmers in the region for all seasons. These communities see
more interconnected relationships with climate change. The highest relationship between
variables (average density) has been observed by marginal communities in Bhilwara
(Table 2). The number of emitter (outdegrees) and receiver (indegree) variables
represents the strength of outgoing and incoming variables (Reckien et al. 2010). The
higher number of receivers per factor indicates that the map constitutes a large number of
outcomes and implications observed by marginal farmers in Bhilwara over stakeholders in
Tonk for all three seasons. Emitters per factor indicate complexity or number of
relationships with the maximum observed by small scale farmers in Tonk followed by
marginal farmers in Bhilwara for all three seasons (Tables 3 and 4).
Table 2: Network statistics of a single map of marginal farmers in Bhilwara
Network statistics

Summer

Winter

Rainfall

Average variables

15.42

11.35

13.00

MAX variables

26.00

21.00

22.00

MIN variables

6.00

6.00

4.00

Average. connections/variables

1.06

1.10

1.06

Average Density

0.07

0.11

0.09

Average receivers/variables

0.32

0.29

0.28

Average emitters/variables

0.33

0.34

0.33

Average receiver/emitters

0.95

0.86

0.85

Size of cognitive maps

Density of maps

Complexity - influence and diversity

Table 3: Network statistics of a single map of marginal farmers in Tonk
Network statistics

Summer

Winter Rainfall

Size of cognitive maps
Average variables

17.81

14.15

15.62

MAX variables

28.00

22.00

24.00

MIN variables

10.00

7.00

9.00

Average. connections/variables

1.17

1.15

1.13

Average Density

0.07

0.08

0.08

Average receivers/variables

0.27

0.26

0.24

Average emitters/variables

0.31

0.36

0.32

Average receiver/emitters

0.89

0.73

0.73

Density of maps

Complexity - influence and diversity

13

Table 4: Network statistics of a single map of small farmers in Tonk
Network statistics

Summer

Winter

Rainfall

Average variables

17.33

14.17

15.75

MAX variables

20.00

21.00

25.00

MIN variables

14.00

10.00

11.00

Average. connections/variables

1.03

1.13

1.11

Average Density

0.06

0.08

0.07

Average receivers/variables

0.28

0.22

0.23

Average emitters/variables

0.33

0.35

0.35

Average receiver/emitters

0.86

0.63

0.65

Size of cognitive maps

Density of maps

Complexity - influence and diversity

4.1.2. Analysis of cognitive interpretive diagrams (CIDs)
We interviewed both marginal and small scale farmers (except in Bhilwara) for the three
seasons. This gave us nine CIDs; the summary of network statistics appears in Tables 5, 6
and 7.
Table 5: Network statistics of Cognitive Interpretive Diagrams of marginal farmers in
Bhilwara
Network statistics

Summer

Winter

Rainfall

Size of cognitive maps
Number of variables in CID

66

59

64

1.74

1.90

1.68

Number of connections in CID

156

136

154

New connection/ map in average

4.11

4.39

4.05

0.036

0.039

0.038

15

11

11

0.39

0.35

0.29

22

17

23

New emitter/ map in average

0.58

0.55

0.61

Receiver/ emitter in CID

0.68

0.65

0.48

New variable/ per map in average
Density of map

Average. Density in CID
Complexity - influence diversity
Number of receivers in CID
New receiver/ map in average
Number of emitter in CID

14

Table 6: Network statistics of Cognitive Interpretive Diagrams of marginal farmers in
Tonk
Network statistics

Summer

Winter

Rainfall

Size of cognitive maps
Number of variables in CID

57

54

59

New variable/ per map in average
Density of map

2.19

2.08

2.27

Number of connections in CID

140

118

140

New connection/ map in average

5.38

4.54

5.38

0.043

0.040

0.040

9

8

9

0.35

0.31

0.35

Average. Density in CID
Complexity - influence diversity
Number of receivers in CID
New receiver/ map in average
Number of emitter in CID

19

22

20

New emitter/ map in average

0.73

0.85

0.77

Receiver/ emitter in CID

0.47

0.36

0.45

Table 7: Network statistics of Cognitive Interpretive Diagrams of small farmers in
Tonk
Network statistics

Summer

Winter

Rainfall

38

41

52

4.22

3.42

4.33

68

82

99

7.56

6.83

8.25

0.047

0.049

0.037

0.89

0.33

0.58

12

15

19

New emitter/ map in average

1.33

1.25

1.58

Receiver/ emitter in CID

0.67

0.27

0.37

Size of cognitive maps
Number of variables in CID
New variable/ per map in average
Density of map
Number of connections in CID
New connection/ map in average
Average. Density in CID
Complexity - influence diversity
Number of receivers in CID
New receiver/ map in average
Number of emitter in CID

Marginal farmers from Bhilwara revealed the maximum number of variables for all three
seasons. New variables added per cognitive map stood at maximum with small scale
farmers in Tonk, but the number of variables added per cognitive map in each group was
an average of 2–4 variables (Tables 5, 6 and 7). As far the CID is concerned, the size of
each variable indicates centrality or the importance of the variable, which is the summation
of indegrees and outdegrees. The density of CIDs occurs highest amongst small scale
15

farmers from Tonk over all three seasons. They also perceive higher relationships
amongst the variables. Tonk’s small scale farmers show higher complexity i.e. a larger
number of outcomes and relationships during summer, while complexity of cognitive maps
amongst Tonk’s marginal farmers has highest occurrence during winter and rainy seasons,
followed by stakeholders from Bhilwara (Tables 5, 6 and 7).
4.2. Perception of climate-related perturbations and adaptive mechanisms in
Bhilwara
. Agro-pastoralists in Bhilwara perceiving the inexorability of climate change, have
observed an increase in summer and winter temperatures with a decline in precipitation.
The most evident impact of increased summer temperature is decline in water availability
while decline in agricultural produce is the chief impact under increased winter
temperature conditions. The general perception of climate-related perturbations (direct and
indirect) on livelihoods due to increased summer and winter temperatures is mainly based
on natural assets, physical and financial assets (Table 8) including water availability,
agricultural produce, groundwater, fodder, forest degradation, human health, money
(income), and the declining number and deteriorating health of livestock (Figures 2 and 3).
Declining rainfall in Bhilwara over the past decade has had its impact felt on agricultural
produce and water reserves causing increased droughts in the region (Figure 4).
Agriculture has been severely affected due to declining soil quality and soil moisture. This
has not only affected food availability but also livestock since crop residue serves as
fodder. Livestock decline causes decreased manure supply leading to decreased crop
produce. Milk production too is affected because of the livestock’s deteriorating health.
Declining soil quality and moisture has led to forest degradation causing, in the bargain, a
rapid decline in fuel wood and forest products. The waning of essential non-timber forest
products (NTFP’s) including alternative sources of fodder (green leaves) has impoverished
the population further. Increased temperatures have caused a proliferation in the
population of mosquitoes resulting in malaria outbreaks during the monsoon season. To
grasp the sensitivity of each asset classes due to climate change see Table 8.
Table 8: Assets that are sensitive to climate change and assets that serve as
adaptive capacities as perceived by farmers in Bhilwara

Livelihood
asset class
Natural
Human
Physical
Financial
Social
Organizational

Increased summer
Increased winter
Decreased Rainfall
temperature
temperature
Adaptive
Adaptive
Adaptive
Sensitivity capacity
Sensitivity capacity Sensitivity capacity
0.75
0.71
0.71
0.78
0.75
0.56
0.69
0.56
0.69
0.61
0.67
0.67
0.78
0.58
0.72
0.67
0.53
0.69
0.83
1.00
0.78
0.80
0.82
0.00
0.68
0.40
0.84
0.00
0.48
0.48
0.00
0.75
0.00
0.70
0.00
0.73

16

Of the 38 groups interviewed 17 comprised women in Bhilwara. Each village except for
Bagjana and Chitamba, was represented by at least one women’s group. Increased
frequency of heat waves has added to the stress on women’s health (dehydration) and
work efficiency. Women who were interviewed mentioned about rising temperatures
leading to decreased soil moisture and increased soil temperatures causing blisters on
their feet. A few interesting impacts mentioned by women’s groups included declining
NTFPs and increased soil hardness and mortality. Alcoholism issues too surfaced.
Increased crop failure and limited opportunities from government-sponsored employment
programmes limits incomes from meeting household demands and undermining the selfconfidence of many men who have turned to alcohol depleting their financial reserves.
This, according to the women, is among the main factors behind the decline of produce.

Adaptations: Physical and natural assets have contributed to adaptations including
irrigation, farm bunding, well constructions, animal, and human health care. Organisational
assets have mainly facilitated resilience through watershed, afforestation, and pastureland
development committees (Table 8). Institutional environments in the region have also
contributed towards absorbing climate shocks although they were not envisioned as
setting up resilience against climate change. Implications of the institutional environments
including the Mahatma Gandhi National Rural Employment Guarantee Scheme
(MNREGS) have contributed towards providing alternative sources of income with crop
production being severely affected. In Bhilwara, where agricultural produce has declined,
watershed activities under the MNREGS are being undertaken. These initiatives have not
only provided alternative sources of income but also improved water resources to some
extent by slowing down the runoff, reducing soil erosion, and improving soil moisture. In
some villages biogas plants have been provided at subsidised rates to reduce dependency
on fuel wood with residue from the biogas plant used as manure. Other local institutional
arrangements include regulations on grazing and collecting of fodder from common
pasturelands to optimise resource utilisation. Various village-level forest committees
conduct afforestion activities while others deal with forest protection. Financial assets in
terms of access to credit and loans have also increased adaptive capacity during summer
and winter.
The usage of chemical pesticides and fertilisers for enhancing crop produce, however, is
still rampant even with stakeholders perceiving a reduction in soil quality and increase in
soil hardness with no alternatives available. Migration, in some cases, also increases
sensitivity which is a reluctant choice. Access to market for food and fodder is another
alternative that helps cope with declining yields; but it only burdens already declining
financial resources. These may be portrayed as mal adaptations as they further stress
livelihoods. People’s perception indicates that climate change burdens livelihood

17

strategies; vulnerability of these populations is set to increase if climate resilient
development pathways are not adopted.
We have calculated vulnerability over three seasons: summer, winter, and rainfall.
Vulnerability is not constant and can change from time to time. Figure 5 shows the
relationship between exposure, sensitivity and adaptive capacity for the three seasons in
Bhilwara. Vulnerability due to increased summer temperature stands at 0.0728, due to
increased winter temperature at 0.023, and due to rainfall at 0.021. Lying in the range of
being vulnerable to climate change and cradled by the semi-arid region of region of
Rajasthan, Bhilwara is more susceptible to increased summer temperatures compared to
other seasons.
4.3. Perception of climate related perturbations and resilience mechanisms in
Tonk
In a bid to gauge Tonk’s ulnerability inter iews were conducted with marginal and small
scale farmers. . Their perceived impacts pointed to financial assets being largely impacted
followed by natural and physical assets attributable to increased summer temperatures
(Figures 6, 7 and Table 9). Both groups had similar views on impacts and adaptations.
Forest degradation was the main concern of the groups interviewed followed by income,
livestock health, fodder, drinking water availability, and human health. Stakeholders
interviewed in the district were highly dependent on forest resources, it being their main
source of fodder. Forest degradation has led to increased human-wildlife conflicts. Indirect
impacts have mainly been felt on financial resources and income. Raised winter
temperatures have led to the decline of agricultural produce, forest degradation, fodder,
and water availability within natural assets (Figures 8 and 9). Increased winter
temperatures have impacted human health and financial resources. Farmers associate
increased winter temperatures with increased malarial incidences as they perceive an
increase in mosquito populations.
Table 9: Assets that are sensitive to climate change and assets that serve as
adaptive capacities as perceived by farmers in Tonk
Increased summer
Increased winter
Decreased Rainfall
temperature
temperature
Adaptive
Adaptive
Adaptive
Sensitivity capacity Sensitivity capacity
Sensitivity capacity
Natural
0.81
0.56
0.69
0.74
0.73
0.66
Human
0.66
0.66
0.76
0.48
0.74
0.47
Physical
0.80
0.57
0.63
0.54
0.65
0.56
Financial
0.82
0.23
0.69
0.37
0.72
0.55
Social
0.75
0.24
0.70
0.45
0.73
0.64
Organizational
0.90
0.62
0.00
0.55
1.00
0.55
Livelihood
asset class

The impacts and adaptations highlighted by these groups do not differ from one another’s
assessments; they all have a similar understanding of climate-related impacts and coping
18

strategies. The major impacts due to declining rainfall have been attributed to agricultural
produce, water availability, forest degradation, and fodder in natural assets (Figures 10
and 11). Declining human health due to malarial incidences and increasing human wildlife
conflicts are the major impacts faced by both groups. Significantly, financial assets and
income are indirect consequences of decreased rainfall.
Of the 35 groups interviewed, six comprised women. Increasing summer and winter
temperatures have directly affected water availability in the region. Scarcity of water
compels villagers, especially women, to travel long distances (at least 5-6 km or even
beyond 9-10 km) to fetch water for drinking and household purposes. Another point
(mostly highlighted by the women) was the relationship between agricultural produce and
education. Declining crop produce has led to depressed incomes, making it difficult for the
villagers to meet their children’s school fees. Parents have been encouraging their children
to work instead of allowing them to pursue their education.
Adaptations: Tonk is relatively weak in terms of coping with climate change owing to the
lack of financial and organisational assets aiding the strengthening livelihoods (Table 9).
Most adaptive mechanisms are autonomous; tube wells and bore wells are deepened in
order to tackle irrigation problems. While local water markets help meet dire situations they
encumber financial resources. Due to lowered water availability farmers have had to shift
to less water-intensive crops. Environmentally unsustainable practices are conducted to
enhance agricultural produce through chemical fertilisers including urea, diammonium
phosphates, super phosphates, insecticides, and pesticides. Impacts of MNREGS are
visible even though they do not contribute significantly towards reducing shocks against
climate change. There have been a few watershed activities in this region under the
MNREGS involving the creation of check dams and farm bunding and providing alternative
livelihood strategies. Milk cooperatives have also facilitated market linkages by providing
alternative sources of income. Some non governmental organisations in this region have
created self help groups (SHGs) to control dairy activities and increase women’s savings
while helping improve their living conditions. The adaptive capacity and adaptation in this
region is low with the population portraying signs of vulnerability. Immediate action on
adaptation is required to help communities’ cope better with climate change in Tonk.
The overall vulnerability of the district was calculated based on perceptions of both small
and marginal farmers. Figure 12 shows the relationship between exposure, sensitivity, and
adaptive capacity. The overall vulnerability for Tonk is higher than for Bhilwara for all three
seasons. The vulnerability due to increased summer temperature stands at 0.1668, for
increased winter temperature at 0.1198, and for decreased rainfall at 0.0498. This shows
that Tonk’s vulnerability is higher than that of Bhilwara’s, because of lower adaptive
capacity attributable to lower assets’ base, lack of planned adaptation measures, and
greater natural aridity.

19

4.4. Climate Change Vulnerability Assessment for Sustainable Livelihoods:
Bhilwara versus Tonk
The data depict that exposure component is higher during summer and winter due to
increasing temperature. Exposure component of decreased rainfall in the study area is
lower than increased temperatures since it lies in the semi-arid region. There is not much
difference in the exposure component between Bhilwara and Tonk. Perception of
perturbation arising out of the increase in summer and winter temperatures and decreased
rainfall months is relatively high for both districts. A small decline in rainfall is considered a
major perturbation as the study areas lie in the drought prone semi-arid region of NorthWestern India. Adaptive mechanisms and capacities are not similar in Bhilwara and Tonk.
Bhilwara portrays higher adaptation to climate change compared to Tonk. This shows that
Bhilwara has a higher capacity to cope with climate change compared to Tonk. Impacts
across both regions due to climate change are similar but it is adaptive capacity that
determines the region’s vulnerability.
4.5. Strengths and Limitations to this Approach
There have numerous discussions regarding the strengths and limitations of fuzzy
cognitive maps. Some of the main strengths include being easy to build, yielding
quantitative results, allowing a feedback process, and having the capability of being made
by anyone (Kosko 1987; Ozesmi and Ozesmi 2004). FCMs can deal with a large number
of variables that are not well-defined in the context of limited scientific knowledge but are
aided by local/ expert knowledge (Kosko 1987; Ozesmi and Ozesmi 2004) as they can
capture the interconnected interactions occurring in a dynamic and interacting system that
are not captured in indicator-based assessments. It is easier to obtain the essence of a
system functioning through fuzzy cognitive maps (Taber 1991; Ozesmi and Ozesmi 2004).
Similar results may be obtained with smaller samples as opposed to other sampling
techniques (Ozesmi and Ozesmi 2004) since FCM captures the richness of interactions
within stakeholder groups. Cognitive maps of different stakeholder groups can be
combined (Kosko 1992; Ozesmi and Ozesmi 2004). Under-identification in a major
problem with indicator-based methods and techniques including the analytical hierarchal
process (AHP) and structured equation modeling (SEM), which does not arise with FCM
with its unlimited number of variables and loops (Ozesmi and Ozesmi 2004).
A chief drawback of this method is its inability to provide real-value parameter estimations
which could allow inferential statistical tests and deal with co-occurrences of multiple
causes (Schneider et al. 1998; Ozesmi and Ozesmi 2004). The assignment of
directionality from less to more vulnerable involves normative judgments (Vincent 2007;
Hahn et al. 2009).

20

5. CONCLUSIONS
Fuzzy cognitive maps capture the dynamics of a system functioning and may be regarded
as a system dynamic method. Crucial to grasping real vulnerability is capturing the
interconnected relationships present in an interacting space like the dynamic climatehuman-environment system. These interconnected relationships portray true vulnerability
as compared to linear indicator-based vulnerability assessments describing an interacting
space without showcasing interconnected relationships. The FCM-based climate change
vulnerability index for sustainable livelihoods depicts peoples' understanding on how
sensitive they are to climate change and how they respond to it. These values may not
provide statistical inference but act as the representation of a belief system in relative
terms. This not only supplements findings where there is a lack of scientific data but also
contributes in comparing understanding of people to scientific data, while being a powerful
tool to understand human behaviour.
Peoples' perception of vulnerability needs to be understood; this may be better quantified
in relative terms. Relative vulnerability assessments may be carried out on a larger scale,
not necessarily at household-levels warranting ease while comprehending peoples'
perception of vulnerability. For communities that perceive the prevalence of higher
vulnerability further investigation of vulnerability may be conducted using indicator-based
assessments that are linear and provide outcome-based results. Thus, a balance between
quantitative methods of vulnerability assessments and semi-qualitative methods like FCM
may be desirable. Such studies can open up new avenues for research including
understanding the insecurities of people due to food, water security, environmental
degradation, pollution, and climate change.

21

Figure 1: Accumulation curve

15
10
5
0

Number of Variables

20

Accumulation Curve for Samples

0

10

20

30

Number of Fuzzy Cognitive Maps

22

40

Natural Assets

Human Assets

Social Assets

Figure 2: Perception of impacts and adaptations due to increased summer temperature in Bhilwara

Organisational Assets

LEGEND
Perturbations
Adaptations

23

Financial Assets

Physical Assets

Figure 3: Perception of impacts and adaptations due to increased winter temperature in Bhilwara

LEGEND
Perturbations

Social Assets

Adaptations

Human Assets

Natural Assets

Organisational Assets

24

Financial Assets

Physical Assets

Figure 4: Perception of impacts and adaptations due to declining rainfall in Bhilwara
Social Assets

LEGEND

Natural Assets

Perturbations
Adaptations
Human Assets

Organisational Assets

25

Financial Assets

Physical Assets

Figure 5: Climate Change Vulnerability in Bhilwara
Sensitivity
0.8
0.7
Summer

0.6
0.5

Winter

0.4

Rainfall

0.3
0.2
0.1
0

Exposure

Adaptive Capacity

26

Human Assets

Physical Assets

Organisational Assets

27

Financial Assets

Natural Assets

Social Assets

Figure 6: Perception of marginal farmers on impacts and adaptations due to increased summer temperature in Tonk

Figure 7: Perception of small scale farmers on impacts and adaptations due to increased summer temperature in Tonk
LEGEND

Social Assets

Perturbations
Adaptations
Human Assets
Physical Assets

Organisational Assets

28

Financial Assets

Natural Assets

Figure 8: Perception of marginal farmers on impacts and adaptations due to increased winter temperature in Tonk
LEGEND
Perturbations
Adaptations
Social Assets

Human Assets

Physical Assets

Organisational Assets

29

Financial Assets

Natural Assets

Figure 9: Perception of small scale farmers on impacts and adaptations due to increased winter temperature in Tonk
LEGEND
Perturbations
Adaptations
Human Assets
Physical Assets

Organisational Assets

30

Financial Assets

Natural Assets

Figure 10: Perception of marginal farmers on impacts and adaptations due to decreased rainfall in Tonk

Social Assets

LEGEND

Adaptations

Human Assets

Natural Assets

Organisational Assets

31

Physical Assets

Financial Assets

Perturbations

Figure 11: Perception of small scale farmers on impacts and adaptations due to decreased rainfall in Tonk
LEGEND
Perturbations
Adaptations
Natural Assets

Social Assets

Human Assets

Organisational Assets

32

Financial Assets

Physical Assets

Figure 12: Climate Change Vulnerability in Tonk
Sensitivity
0.8
0.7
0.6

Summer

0.5

Winter

0.4

Rianfall

0.3
0.2
0.1
0

Exposure

Adaptive Capacity

33

REFERENCES
Alwang, J., Siegel, P. and Jorgensen S. L., 2005. Vulnerability: a view from different
disciplines. In: Social Protection Unit, World Bank, Social Protection
Discussion paper series115, Washington DC: Social Protection Unit, World
Bank, Washington, DC.
Adger, W.N., 2006. Vulnerability. Global Environmental Change, 16(3), pp.268-281.
Blaikie, P., Cannon, T., Davis, I. and B. Wisner., 1994. At Risk: natural hazards,
people’s vulnerability and disasters. [e-book] London: Routledge. Available
at: <http://www.preventionweb.net/files/670_72351.pdf> [Accessed 1 May
2013].
Burton, I., Kates, R.W. and White, G.F., 1978. The Environment as Hazard. New York:
Guilford Publications.
Brooks, N., Adger, W.N. and Kelly, P.M., 2005. The determinants of vulnerability and
adaptive capacity at the national level and the implications for adaptation.
Global Environmental Change, 15, pp.151–163.
Census of India, 2011, Population of India, New Delhi: Office of the Registrar General
and Census Commissioner, India
Chambers, R. and Conway, G., 1992. Sustainable rural livelihoods: Practical concepts
for the 21st century. In: IDS Discussion Paper 296, Institute of Development
Studies, Brighton.
Cruz, R.V., Harasawa, H., Lal, M., Wu, S., Anokhin, Y., Punsalmaa, B., Honda, Y.,
Jafari, M., Li, C. and Huu Ninh, N., 2007: Asia. Climate Change 2007:
Impacts, Adaptation and Vulnerability. Contribution of Working Group II to
the Fourth Assessment Report of the Intergovernmental Panel on Climate
Change, M.L. Parry, O.F. Canziani, J.P. Palutikof, P.J. van der Linden and
C.E. Hanson, Eds., Cambridge University Press, Cambridge, UK, 469-506.
Cutter, S.L., 1996. Vulnerability to environmental hazards. Progress in Human
Geography, 20, pp.529–539.
Cutter, S.L., Mitchell, J.T. and Scott M.S., 2000. Revealing the vulnerability of people
and places: a case study of Georgetown County, South Carolina. Annals of
the Association of American Geographers, 90, pp.713–737.
Cutter, S. Boruff, B. and Shirley, W., 2003. Social Vulnerability to Environmental
Hazards. Social Science Quarterly, 84(2), pp.242-261.
Department of International Development, 1999. Sustainable Livelihoods Guidance
Sheets. [pdf] Available at: <
http://www.eldis.org/vfile/upload/1/document/0901/section2.pdf> [Accessed
20 April 2013]
Directorate of Economics and Statistics, 2008, Bhilwara and Tonk profile, Jaipur:
Directorate of Economics and Statistics, Government of Rajasthan,
Available at: <http://www.statistics.rajasthan.gov.in/socio_Bhilwara.aspx>
[Accessed 3 April 2013].
Eakin, H., 2005. Institutional Change, Climate Risk, and Rural Vulnerability: Cases
from Central Mexico. World Development, 33(11), pp.1923–1938.

34

Eakin, H. and Leurs, L., 2006. Assessing the Vulnerability of Socio-Environmental
Systems. Annual Reviews, pp.365-394.
Eakin, H. and Bojorequez-Tapia, A. L., 2008. Insights into the composition of
household vulnerability from multicriteria decision analysis. Global
Environmental Change, 18(1), pp.112-127.
Ferrarini, A., 2011, Fuzzy cognitive maps outmatch loop analysis in dynamic modeling
of ecological systems. Computational Ecology and Software, 1(1), pp.55–
59.
Fowler, F. J., 2002. Survey Research Methods. 3rd ed. California: Sage Publications,
Inc.
Gautam, M., et al., 2013. Climate change in arid lands and Native American
socioeconomic vulnerability: The case of the Pyramid Lake Paiute Tribe.
Climatic Change.
Gopalakrishnan, R., Jayaraman, M. and Ravindranath, N. H., 2011. Regional Climate
Modeling results for Rajasthan state. In: Science-based policy options for
climate change adaptation in Rajasthan. Jaipur, India 24-25 February 2011.
Jaipur: Rajasthan State Pollution Control Board.
Government of Rajasthan, 2011, Rajasthan State Action Plan on Climate Change,
Government of Rajasthan.
Government of Rajasthan, 2011, Rajasthan Foundation. Rajasthan Foundation,
Government of Rajasthan: Available at
<http://rajasthanfoundation.gov.in/tonk.htm> [Accessed 3 April 2013].
Green, D., Billy, J. and Tapim, A., 2010. Indigenous Australians’ knowledge of weather
and climate. Climatic Change, 100, p.337–354.
Hahn, M., Riederer, N. and Foster, S., 2009. The Livelihood Vulnerability Index: A
pragmatic approach to assessing risks from climate variability and changeA case study in Mozambique. Global Environmental Change, 19(1), pp.7488.
Harary, F., Norman R. Z., and Cartwright D., 1965. Structural Models: An Introduction
to the Theory of Directed Graphs. New York: John Wiley & Sons.
Hennessey, K., Fitzharris, B., Bates, B. C., Harvey, N., Howden, S. M., Hughes, L., et
al., 2007. Australia and New Zealand. In: M. L. Parry, O. F. Canziani, J. P.
Palutikof, P. J. Van Der Linden, C. E. Hanson, eds. 2007. Climate Change
2007: Impacts, Adaptation and Vulnerability. Cambridge: Cambridge
University Press. pp. 507–540
Holling, C. S., 1996. Engineering resilience versus ecological resilience. In: P. Schulze
ed. 1996. Engineering Within Ecological Constraints, Washington DC:
National Academy of Science. pp. 31–44.
Hunter, N., Leonard, S., 2010. Indigenous weather knowledge and bio temporal
indicators of climate change. In: National Climate Change Adaptation
Research Facility (NCCARF) and Commonwealth Scientific and Industrial
Research Organization (CSIRO), 2010 International Climate Change
Adaptation Conference: Climate Adaptation Futures, Gold Coast,
Queensland, Australia, 29 June 2010. Queensland: NCCARF and CSIRO.
IFAD and the Global Mechanism, 2009, Climate change impacts in the Asia/ Pacific
Region. In: International Fund for Agricultural Development. Available at: <
35

http://asia.ifad.org/resources?p_p_id=1_WAR_resource_libraryportlet&_1_
WAR_resource_libraryportlet_jspPage=%2F%2Fhtml%2Fresource_library
%2Fentry_detail.jsp&_1_WAR_resource_libraryportlet_entryId=3177>
[Accessed 4 May 2013].
IPCC 2001, Climate Change: Impacts, Adaptation and Vulnerability. IPCC Third
Assessment Report, Cambridge University Press.
Indian Network for Climate Change Assessment (INCAA), 2010, Climate Change and
India: A 4X4 Assessment: A Sectoral and Regional Analysis for 2030,
Ministry of Environment & Forests, Government of India.
India’s Second National Communication (ISNC), 2008, India's Initial National
Communication to the United Nations Framework Convention on Climate
Change, Ministry of Environment & Forest, Government of India.
Isak, K., Wildenberg, M., Adamescu, M., Skov, F., De Blust G. and Varjopuro R., 2009.
Manual for applying Fuzzy Cognitive Mapping – experiences from ALTERNet. Alter-Net final report. Available at
http://www.fcmappers.net/joomla/index.php?option=com_content&view=arti
cle&id=66:fcm-in-alter-net&catid=39:current-projects&Itemid=37 [Accessed
8 May 2013]
Jacob, C., McDaniels, T. and Hinch, S.G., 2010. Indigenous culture and adaptation to
climate change: sockeye salmon and the Statimc people. Mitigation and
Adaptation Strategies for Global Change, 15, pp.859–876.
Kelly, P.M. and Adger, W. N., 2000. Theory and practice in assessing vulnerability to
climate change and facilitating adaptation. Climate Change, 47(4), pp.325–
352.
Kok, J. Titus, M. and Wind, H. G., 2000. Application of fuzzy sets and cognitive maps
to incorporate social science scenarios in integrated assessment models.
Integrated Assessment, 1, pp.177-188.
Kosko, B., 1987. Adaptive inference in fuzzy knowledge networks. Proceedings of the
First IEEE International Conference on Neural Networks (ICNN-86), 2, pp.
261–268.
Kosko, B., 1986. Fuzzy Cognitive Maps. International Journal of Man-Machine Studies,
24, pp.65-75.
Kosko, B., 1992. Fuzzy associative memory systems. A. Kandeled, Fuzzy Expert
Systems. Boca Raton: CRC Press, pp.135–162.
Leichenko, R., O’Brien, K., Aandahl, G. Tompkins, H. and Javed A., 2004. Mapping
Vulnerability To Multiple Stressors: A Technical Memorandum. Oslo,
Norway: Center for International Climate and Environmental Research
(CICER).
Leichenko, R. and O’Brien, K., 2002. The dynamics of rural vulnerability to global
Change: the case of Southern Africa. Mitigation and Adaptation Strategies
for Global Change, 7, pp.1-18.
Leonarda, S., Parsons, M., Olawsky, K., and Kofoda, F., 2013, The role of culture and
traditional knowledge in climate change adaptation: Insights from East
Kimberley, Australia. Global Environmental Change, 23, pp.623-632

36

Luers, A., Lobell, D., Sklar, L.S., Addams, C.L. and Matson, P.M., 2003. A method for
quantifying vulnerability, applied to the Yaqui Valley, Mexico. Global
Environmental Change, 13(4), pp.255–267.
Li, Y., Conway, D., Wu, Y., Gao, Q., Rothausen, S. Xiong, W., Ju, H., and Lin, E.,
2013. Rural livelihoods and climate variability in Ningxia, Northwest China.
Climatic Change, [e-journal] Available through: springerlink.com
website<http://link.springer.com/content/pdf/10.1007%2Fs10584-013-07659.pdf> [Accessed 4 May 2013].
Moss, R. H., Brenkert, A. L. and Malone, E. L., 2001. Vulnerability to climate change: a
quantitative approach. Richland, Washington: Pacific Northwest National
Laboratory. Available at: <
http://www.globalchange.umd.edu/data/publications/Vulnerability_to_Climat
e_Change.pdf> [Accessed 6 May 2013].
Mustafa, D., 1998. Structural causes of vulnerability to flood hazard in Pakistan.
Economic Geography, 74, pp.289–305.
Murungweni, C., Wijk, M. T., Andersson, J. A., Smaling, E. M. A. and Giller, K.E., 2011,
Application of fuzzy cognitive mapping in livelihood vulnerability analysis,
Ecology and Society, 16(4), p.8.
O’Brien, K.L., Leichenko, R., Kelkarc, U., Venemad, H., Aandahl, G., Tompkins, H.,
Javed, A., Bhadwal, S., Barg, S., Nygaard, L. and West, J., 2004. Mapping
vulnerability to multiple stressors: climate change and globalization in India.
Global Environmental Change, 14, pp.303–313.
O’Brien, K.L., Eriksen, S., Schjolden, A. and Nygaard, L., 2005. What’s in a word?
Interpretations of vulnerability in climate change research, Oslo: Center for
International Climate and Environmental Research (CICERO), Available at:
< http://www.cicero.uio.no/media/2682.pdf> [Accessed 25 April 2013].
Orencio, P. and Fijji, M., 2013. An index to determine vulnerability of communities in a
coastalzone: a case study of baler, aurora, Philippines. Hokkaido
University.
Ozesmi, U. and Ozesmi, L., 2004. Ecological models based on people’s knowledge: a
multi-step fuzzy cognitive mapping approach. Ecological Modelling, 176,
pp.43-64.
Papageorgiou, E. and Kontogianni, A., 2012. Using Fuzzy Cognitive Mapping in
Environmental Decision Making and Management: A Methodological Primer
and an Application. In: S. S. Young and S. E. Silvern, eds. 2012.
International Perspectives on Global Environmental Change. s.l.: InTech.
Ch.21.
Park, S., Howden, M., Booth, T., Stokes, C., Webster, T., Crimp, S., Pearson, L.,
Attard, S. and Jovanovic, T., 2009. Assessing the vulnerability of rural
livelihoods in the Pacific to climate change. Canberra: CSIRO Sustainable
Ecosystems.
Petheram, L., Zander, K.K., Campbell, B.M., High, C. and Stacey, N., 2010. Strange
changes: indigenous perspectives on climate change and adaptation in NE
Arnhem Land (Australia). Global Environmental Change, 20, pp.681–692.
Polsky, C., Neff, R. and Yarnal, B., 2007. Building comparable global change
vulnerability assessments: the vulnerability scoping diagram. Global
Environmental Change, 17, pp.472–485.
37

Preston, B.L. and Stafford-Smith, M., 2009, Framing vulnerability and adaptive
capacity assessment: Discussion paper. Canberra: CSIRO. Available at
<http://www.csiro.au/org/ClimateAdaptationFlagship.html> [Accessed 6
March 2013].
Reckien, D., Wildenberg, M. and Deb, K., 2010. Understanding Climate Change
Impacts and Adaptation Options in Developing Countries' Megacities. In: K.
Otto-Zimmermann, ed. 2010, Resilient cities: Cities and Adaptation to
Climate Change Proceedings of the Global Forum 2010, s.l. Ch. 3
Scoones, I., 1998. Sustainable Rural Livelihoods: A Framework for Analysis. Institute
of Development Studies, Brighton (IDSB). Working paper 72, Brighton:
IDSB.
Schroter, D., Polsky, C. and Patt, A.G., 2005, Assessing vulnerabilities to the effects of
global change: An eight step approach. Mitigation and Adaptation
Strategies for Global Change, 10, pp.573-596.
Schneider, M., Shnaider, E., Kandel, A. and Chew, G., 1998. Automatic construction of
FCMs. Fuzzy Sets Systems, 93, pp.161–172.
Sivakumar, S. and E. Kerbart, 2004. Drought, sustenance and livelihoods: 'akal' survey
in Rajasthan, India. Economic and Political Weekly, 39(3), pp.285-294.
Singh, R. D. and Kumar C. P. 2010. Impact of Climate Change on Groundwater
Resources. In: Ministry of Water Resources, The 2nd National Ground
Water Congress. New Delhi 22 March 2010. New Delhi: Ministry of Water
Resources
Spence, A., Poortinga, W., Butler, C. and Pidgeon, N. F., 2011. Perceptions of climate
change and willingness to save energy related to flood experience. Nature:
Climate Change, 1, pp.46–49.
Stern, N., 2007, The Stern Review: Economics of Climate Change, Washington DC:
Cabinet Office - HM Treasury.
Sullivan, C., 2002, Calculating a water poverty index. World Development, 30,
pp.1195–1210.
Taber, W.R., 1991. Knowledge processing with fuzzy cognitive maps. Expert System
Application, 2, pp.83–87.
Turner et al., 2003a. Illustrating the coupled human environment system for
vulnerability analysis: Three case studies. Proceedings of the National
Academy of Sciences, 100(14), pp.8080-8085.
Turner, L. B., Kasperson, R. E., Matson, P. A., McCarthy, J. J., Corell, R. W.,
Christensen, L., Eckley, N., Kasperson, J. X., Luers, A., Martello, M. L.,
Polsky, C. Pulsipher, A., and Schiller, A., 2003b. A Framework for
Vulnerability Analysis in Sustainability Science. Proceedings of the National
Academy of Sciences, 100(14), pp.8074–8079.
United Nations Environment Programme (UNEP), 2002. Assessing Human
Vulnerability to Environmental Change Concepts, Issues, Methods and
Case Studies. [pdf] Kenya: UNEP. Available at:
<http://www.unep.org/geo/GEO3/pdfs/AssessingHumanVulnerabilityC.pdf >
[Accessed on 20 April 2013]

38

V´asquez-L´eon M, West CT, Finan TJ.2003. A comparative assessment of climate
vulnerability: agriculture and ranching on both sides of the US-Mexico
border. Global Environmental Change, 13, pp.159–173
Vliet, M., Kok, K., and Veldkamp, T., 2009. Linking stakeholders and modellers in
scenario studies: The use of Fuzzy Cognitive Maps as a communication
and learning tool. Futures, 42(1), pp.1-14.
Vogel, C. and O’Brien K. L., 2004.Vulnerability and global environmental change:
rhetoric and reality. AVISO, 13, pp.1–8.
Vincent, K., 2004. Creating an index of social vulnerability to climate change for Africa.
Tyndall Centre for Climate Change Research (TCCCR) and School of
Environmental Sciences (SES), University of East Anglia, Working Paper
56, TCCCR and SES, University of East Anglia.
Vincent, K., 2007. Uncertainty in adaptive capacity and the importance of scale. Global
Environmental Change, 17, pp.12–24.
Wolf, J. and Moser, S., 2011. Individual understandings, perceptions, and engagement
with climate change: insights from in-depth studies across the world.
WIRES Climate Change, 2, pp.547–569.

39

Annexure
Table A1: Sensitivity analysis for Bhilwara
Overall
Sensitivity

Asset
Classes

0.7421967

Natural
assets

Sensitivity
of assets

Variables

Sensitivity
of Variables

Drivers of
sensitivity

Value of
drivers

0.7526
Agricultural
productivity

0.8274
Crop failure
Groundwater
Human wildlife
conflict
IST
Soil fertility
Soil moisture
Water Availability

Air quality

Drinking
water

0.925
0.657
0.8
0.67
0.89

0.7
Forest
degradation

Crop failure

0.9
0.95

0.7

0.8875
Groundwater
Heat waves
IST
Water Availability

0.7
0.9
0.95
1

Groundwater
IST
Water Availability

0.675
0.6
0.775

0.6833

Fodder

0.772

Food
Availability

Forest
degradation

Agricultural
productivity
Forest
degradation
IST
Water Availability

0.467
0.871
0.95

Agricultural
productivity
Water Availability

0.84
0.7

0.77

0.7025
IST
Soil hardness

Fuel wood

0.44
0.3

0.742
IST
Water Availability

40

0.705
0.7

0.37
Forest
degradation
IST

Groundwater

0.8

0.784
0.7

Heat waves
Livestock
number

Livestock
health

Milk
production

Milk products

0.73

Fodder
Forest
degradation
Heat waves
IST
Livestock health
NTFP
Water Availability

0.72
0.8
0.7
0.833
0.85
0.8
0.85

0.7912
Fodder
Forest
degradation
Heat waves
IST
Water Availability

0.756

Fodder
Livestock
Livestock health

0.475
0.64
0.744

0.9
0.7
0.7
0.9

0.6197

0.8
0.4

NTFP

0.7

Rainfall

0.73

0.7933

Mosquitoes

Pest invasion

IST

Milk production

0.8

IST

0.4

Forest
degradation

0.7

IST

0.7

IST

0.9

0.7
0.9

Soil fertility

0.635
IST
Pest invasion

Soil hardness

0.8
IST

Soil moisture

0.8

0.76
IST
Water Availability

Surface
water

0.67
0.6

0.72
0.8

0.9

Water
Availability

41

Groundwater

0.9

Groundwater
IST
Rainfall

0.8
0.857
0.9

0.8643

Water quality

Wildlife

Soil moisture

0.9

Groundwater
IST

0.7
0.9

0.8

0.567
Forest
degradation

Manure

Fodder
Livestock
Human
assets

0.6
0.75

0.6906
Education

0.746
Money

Employment
Human
health

0.746

0.5
Work efficiency

0.5

0.5789

Human
wildlife
conflict

Drinking water
Heat waves
IST
Milk products
Water Availability
Water quality
Hand pump

0.2
0.672
0.68
0.4
0.7
0.7
0.7

Forest
degradation

0.925

0.925

Mortality

Work
efficiency

Nutrition

0.6
Heat waves
Human health

0.6
0.6

Heat waves
Human health
IST

0.7
0.683
0.816

0.733

1
Money

Daily wage
labour

Social
assets

0.567

0.675

1

0.8333
IST
Money
Work efficiency

0.8
0.95
0.75

Drinking water
Water Availability

0.85
1

Money

0.65

0.68
Drudgery

Marriage

42

0.925

0.65

Physical
assets

Migration

0.1

Migration

0.8

0.1

IST

0.8

IST

0.8

Soil moisture

0.7

Groundwater

0.7

IST

0.9

0.775
Electricity
supply

0.8

Ploughing

0.7

Irrigation
Biogas gas
plants
Financial
assets

Money

0.7

0.9

0.8291
Money

0.813
Agricultural
productivity
Crop failure
Food Availability
Groundwater
Human health
Livestock
Milk production
Milk products
Alternative
sources of fodder
Animal care
Fertilizer
Food from market
Medical care
Tube well boring
Water supply
Well construction
Well deepening

Poverty

0.76
0.9
0.6
1
0.82
0.775
0.683
0.8
0.833
0.7
1
0.7
0.75
0.9
0.9
1
0.7

0.9

Loan

Money

0.9

Money

1

1

IST stand for Increased summer temperature
Note: Coping mechanisms that cause sensitivity are shown in italics

43

Table A2: Adaptive Capacity Analysis for Bhilwara

Overall
adaptive
capacity
0.6345

Asset
Classes
Natural
assets

Adaptive
capacity of
assets

Influencers of
resilience
(variables)

Adaptive
capacity
of
variables

Influenced
variables

0.70625
Cow dung cakes
Livestock feed

Manure

0.45
Fuel wood

0.45

Fodder
Milk
production

0.7

0.7

0.7

0.7

Plantation

Shift in cropping
pattern

Agricultural
productivity
Soil moisture

0.7
0.7

Forest
degradation

0.8

0.8

0.8
Fodder
Human
wildlife
conflict

Physical
assets

Value of
influenced
variables

1

0.6

0.58316
Alternative
sources of
fodder

Animal care

Dam

0.4935
Fodder
Livestock
health

0.487

Livestock
Livestock
health

0.35

0.5

0.4875

0.625

0.5
Groundwater
Water
Availability

Fertilizer

Food from
market

44

0.2
0.8

0.62
Agricultural
productivity

0.62

Food
Availability

0.5

0.5

Biogas gas
plants

0.65
Fuel wood

Hand pump

0.5165
Drinking
water
Water
Availability

Irrigation

0.71
1

0.8
0.3

0.635
Agricultural
productivity
Drinking
water
Water
Availability

Organiza
tional
assets

0.6

0.55
Drinking
water
Water
Availability

Water supply

0.4

0.855
Human
health
Mortality

Tube well boring

0.633

0.6
Soil fertility

Medical care

0.65

0.7
0.525
0.68

0.7491
Afforestation
committees

0.7765
Forest
degradation
Air quality
Fodder
Water
Availability

Farm bunding
committees

Soil moisture
conservation
committees

45

0.9

0.70625
Agricultural
productivity
Fodder
Water
Availability
Irrigation

Pastureland
development
committees

0.706
0.8
0.7

1
0.3
0.525
1

0.86
Fodder

0.86

Forest

0.7

0.7

degradation
Human
assets

0.56
Daily wage
labour

Night watch

Social
assets

Money
Migration

0.33
0.4

Human
wildlife
conflict

0.95

Money

0.4

Drinking
water

0.4

0.95

0.4
Migration
Social capital

Financial
assets

0.365

0.4
0.4

1
Loans

1
Money

46

1