Socio-Ecological Vulnerability Assessment of Drought and Flood-Saline Prone Regions in Rural Bangladesh

Abstract

The current study attempted to quantitatively measure the vulnerability status of selected regions in Bangladesh impacted by climate change. Three upazilas were selected in the drought prone region of Rajshahi, while another three upazilas were assessed in the saline-flood prone Barisal region. The Exposure, Sensitivity and Adaptive capacity of each upazila was measured through socio-demographic, agro-economic and infrastructural indicators inspired by the literature, RiceClima reports but also elicited from a household survey in the examined areas. The technique of Principal Component Analysis was used for the assessment of the indicators while descriptive statistics also helped for a better understanding of the current situation in the two regions.

The findings indicated that the drought prone Rajshahi upazilas (North Bangladesh) are more exposed to inefficient irrigation management and lack of access to household’s utilities (water, electricity). The flood and saline prone upazilas of the Barisal region in South Bangladesh lack transportation, agricultural, education and health infrastructure on a regional level. In both regions, the introduction of cash crops and the improvement of market conditions in agriculture are deemed as necessary actions.

Full text

RICE CLIMA PROJECT
(Project nr. 2011/000174)
Report
Socio-Ecological Vulnerability Assessment of Drought and
Flood-Saline Prone Regions in Rural Bangladesh
satisfying Deliverable 3.1 of Work Package 3
(January 2014)
Report Contributors:
Stefanos Xenarios, Bioforsk
Golam Wahed Sarker, BRRI
Attila Nemes, Bioforsk
Udaya Sekhar Nagothu, Bioforsk
Jatish Chandra Biswas, BRRI
Md Maniruzzaman, BRRI
Funded by
Norwegian Embassy of Bangladesh

RiceClima – Deliverable 3.1
2

RiceClima – Deliverable 3.1
3
Ackowledgements
The Norwegian Institute for Agriculture and Environmental Research (Bioforsk), the Center for
Environmental and Geographic Information Services (CEGIS) and the Bangladesh Rice Research
Institute (BRRI) are collaborating towards the completion of the multidisciplinary project
‘Climate change impacts, vulnerability and adaptation: Sustaining rice production in
Bangladesh.’
We thank the Royal Norwegian Embassy, Dhaka and the Ministry of Foreign Affairs, Norway
facilitating this study with the financial support, and especially Mr. Arne Haug, Counsellor,
Deputy Head of the Mission and Mr.Morshed Ahmed, Senior Advisor (Development Affairs) for
their guidance and cooperation. We would also like to thank the external reviewer for the very
constructive comments.

RiceClima – Deliverable 3.1
4
Contents
Summary ................................................................................................................................................. 6
1. Introduction ......................................................................................................................................... 7
2. Methodology ....................................................................................................................................... 8
2.1. Defining the vulnerability concept in a changing climate ............................................................... 8
2.2 The operational dimensions of vulnerability ................................................................................... 9
2.2.1 The ‘’starting’’ and ‘’ending’’ points of operational vulnerability .............................................. 9
2.2.2 Principal Component Analysis and Farmer’s preferences ....................................................... 13
3. Case Study ......................................................................................................................................... 16
3.1 Vulnerability of Bangladesh .......................................................................................................... 16
3.2 Drought (Rajshahi Region) and flood-saline (Barisal) regions ........................................................ 17
3.2.1 General Description ............................................................................................................... 17
3.2.2 Agricultural practices in Rajshahi and Barisal regions ............................................................. 20
3.3 Primary and Secondary Data ........................................................................................................ 21
4. Results ............................................................................................................................................... 22
4.1 Descriptive results ........................................................................................................................ 22
4.2 PCA Results .................................................................................................................................. 30
4.3 Farmers’ preferences ................................................................................................................... 32
5. Discussion .......................................................................................................................................... 34
6. Concluding remarks ........................................................................................................................... 36
References ............................................................................................................................................. 37
Annex 1. Excluded indicators in vulnerability assessment ...................................................................... 40
Annex 2. Factor Analysis in PCA.............................................................................................................. 41
Table 1. Factor Analysis for Adaptive Capacity Indicators ................................................................... 41
Table 2. Factor Analysis for Sensitivity and Exposure Indicators ......................................................... 42
Annex 3. Standardizes Values ................................................................................................................. 43
Table 1. Adaptive Capacity ................................................................................................................. 43
Table 2. Exposure – Sensitivity .......................................................................................................... 44

RiceClima – Deliverable 3.1
5
Tables
Table 1. Exposure Indicators .................................................................................................................. 11
Table 2. Sensitivity Indicators ................................................................................................................. 11
Table 3. Indicators of Adaptive Capacity ................................................................................................. 12
Table 4. Example of Standardized Adaptive Indicators ........................................................................... 15
Table 5. Sampling distribution ................................................................................................................ 22
Table 6. Farm ownership status (ha) ...................................................................................................... 23
Table 7. Farm sizes based on cultivated land (ha) ................................................................................... 23
Table 8. Major crops grown and yield level in Rajshahi region ................................................................ 24
Table 9. Major crops grown and yield level in Barisal region .................................................................. 25
Table 10. Location wise crops grown and gross margin (Tk/ha) in Rajshahi region .................................. 26
Table 11. Location wise crops grown and gross margin (Tk/ha) in Barisal region..................................... 27
Table 12. Source-wise irrigation coverage under study sites ................................................................... 28
Table 13. Information about rice disease incidence level ........................................................................ 29
Table 14. Information about rice insect’s incidence level........................................................................ 30
Table 15. Significance of Vulnerability Indicators ................................................................................... 31
Table 16. Suggestions for improvement of adaptive capacity ................................................................. 33
Figures
Figure 1. Boundaries of vulnerability and climate change, Source: Fellman, 2012 ..................................... 8
Figure 2. Rajshahi Region, Source: CEGIS (2013) ..................................................................................... 17
Figure 3. Barisal Region, Source: CEGIS (2013) ....................................................................................... 18
Figure 4. Sluice gate in Amtoli upazila (Source: Field Trip in Barisal region, February 2012) .................... 19
Figure 5. The growth period of cultivated crops in Rajshahi and Barisal regions ..................................... 21
Figure 6. PCA Assessment Results .......................................................................................................... 32
Figure 7. Vulnerability Scenarios for Barisal region ................................................................................. 35

RiceClima – Deliverable 3.1
6
Summary
The current study attempted to quantitatively measure the vulnerability status of selected
regions in Bangladesh impacted by climate change. Three upazilas were selected in the drought
prone region of Rajshahi, while another three upazilas were assessed in the saline-flood prone
Barisal region. The Exposure, Sensitivity and Adaptive capacity of each upazila was measured
through socio-demographic, agro-economic and infrastructural indicators inspired by the
literature, RiceClima reports but also elicited from a household survey in the examined areas.
The technique of Principal Component Analysis was used for the assessment of the indicators
while descriptive statistics also helped for a better understanding of the current situation in the
two regions.
The findings indicated that the drought prone Rajshahi upazilas (North Bangladesh) are more
exposed to inefficient irrigation management and lack of access to household’s utilities (water,
electricity). The flood and saline prone upazilas of the Barisal region in South Bangladesh lack
transportation, agricultural, education and health infrastructure on a regional level. In both
regions, the introduction of cash crops and the improvement of market conditions in agriculture
are deemed as necessary actions.

RiceClima – Deliverable 3.1
7
1. Introduction
It is predicted that climate change will aggravate the presence of sudden (e.g. cyclones, floods
etc.) and chronic (e.g. erosion) hazards to agrarian communities in developing countries. The
degree of exposure, sensitivity and adaptive capacity to climate change determines the
vulnerability level of a community (Nelson et al., 2010a). The agrarian population in Bangladesh
is ranked by many studies to be of the most vulnerable in the world due to the poor socio-
economic features, the unique geophysical location and the high exposure to climate change
effects (Ramamasy and Bass, 2007).
However, the measurement and interpretation of vulnerability indices is argued to be a rather
difficult undertaking (O’Brien et al., 2004). First, it is rather arduous to define the vulnerability of
an agrarian community within some administrative boundaries only. The climate change
impacts affect larger scale areas - geographical regions (Fussel, 2007) and thus it is difficult to
tell the differences between administrative units. Further, there can be multiple threats at
various scales occurring simultaneously in social and natural aspects, which makes the
identification and impact-value assessment quite dubious. Additionally, an impact from climate
change can be instantaneous or may develop slowly over time, and thus the vulnerability
assessment may become a rather puzzling process (Nelson et al., 2010b).
Although there may be difficulties in determining the assessment parameters of vulnerability,
the biophysical and socioeconomic disciplines seem to adopt two distinctively different
approaches. The “end-point” approach is more welcomed among biophysicists while the “start-
point” notion prevails in socio-economics. The “end-point” approach may, for example examine
future climate scenarios by evaluating - through modeling - its biophysical impacts and
suggesting potential adaptive options. The “start-point” deploys the existing inequalities within
a society which are deemed to further exacerbate when exposed to climate change (Smit and
Wandel, 2006).
In our study, we attempted to borrow elements from both domains for the development of a
socio-ecological vulnerability assessment in flood-saline and drought prone areas of Bangladesh.
The northern drought prone Rajshahi and the southern flood-saline prone Barisal regions were
selected as study areas and three sub-regions (upazilas) were adopted in each region.
Demographic, agro-economic and infrastructure related indicators were introduced as assumed
signals of social vulnerability, along with the results of climatic and hydrological models as
biophysical indicators. Principal component analysis (PCA) was employed for the valuation of
the vulnerability levels in each of the examined upazilas. Also, farmers’ preferences were
elicited for a better clarification of potential adaptation measures to be taken against climate
change.

RiceClima – Deliverable 3.1
8
2. Methodology
2.1. Defining the vulnerability concept in a changing climate
According to the definition of the Intergovernmental Panel on Climate Change (IPCC), the
leading international body for the assessment of climate change, the vulnerability to climate
change could be synopsized as the “degree to which a system is susceptible to, and unable to
cope with, adverse effects of climate change, including climate variability and extremes” (IPCC,
2001, Glossary).
The vulnerability concept is highly dependent on the exposure, sensitivity and adaptive capacity
of a system to cope with weather extremes. There is a multitude of interpretations pertaining to
the affecting parameters of vulnerability but we currently borrow the definitions given by the
IPCC which stipulates that the exposure relates to ―”the nature and degree to which a system is
exposed to significant climatic variations”‖ (IPCC, 2001, Glossary). The sensitivity on the other
hand, reveals the “degree to which a system is affected, either adversely or beneficially, by
climate variability or change. The effect may be direct (e.g., a change in crop yield in response to
a change in the mean, range or variability of temperature) or indirect (e.g., damages caused by
an increase in the frequency of coastal flooding due to sea level rise) (IPCC, 2007, Glossary)”.
Finally, the adaptive capacity is dictated as “ the ability (or potential) of a system to successfully
adjust to climate change (including climate variability and extremes) to (i) moderate potential
damages, (ii) to take advantage of opportunities, and/or (iii) to cope with the consequences
(IPCC, 2007, Glossary).
Although the components of vulnerability are well described in IPCC it still remains difficult to
define the multifaceted nature of vulnerability. Both natural and social scientists agree that the
vulnerability is multi-dimensional and differential which means that it is perceived differently
across physical space and between various social groups (Cardona et al., 2012). It is also scale
and time-dependent because various socioeconomic and biophysical impacts unequal in
magnitude, may appear at the same time. Moreover, it is highly dynamic because the impacts
may appear instantaneously or aggregated within the years (Vogel and O’Brien, 2004,
Devisscher et al., 2012). Although the fuzzy nature of vulnerability is highly acknowledged there
is a strong effort to define the boundaries of a vulnerable system. In this report, we have
adopted the following diagrammatic concept of vulnerability as presented below:
Figure 1. Boundaries of vulnerability and climate change, Source: Fellman, 2012

RiceClima – Deliverable 3.1
9
As shown in Figure 1, when the climate is in a stationary mode (left part of the figure), there are
still some weather abnormalities which could be however managed within the coping range of
an agrarian community. For example, higher temperatures or heavier rainfalls could be
observed for some days in rural Bangladesh even when there is a stationary climate. The
farmers have developed the relevant mechanisms to cope with weather fluctuations and
overcome the relevant problems occurring from such weather events.
In the case of climate change however, the weather extremes may become more frequent and
with higher intensity (right part of figure 1). In this case, the coping rage of a socio-ecological
system becomes more limited and it is much dependent on the exposure and sensitivity to the
changing climate. It is then that the adaptive capacity should be enhanced which actually
represents the potential of a system to better adapt in climate change. In other words, the
higher the adaptive capacity, the lower the vulnerability is. On the contrary, the synergy
between exposure and sensitivity will augment the vulnerability levels.
In simple mathematical terms, the vulnerability of climate change can be expressed as below:
𝑉=𝐴 − (𝐸+𝑆)… (1) where
𝑉 = Vulnerability, 𝐴= Adaptive Capacity, 𝐸 = Exposure, 𝑆 = Sensitivity
The operational dimensions of vulnerability often depend on the biophysical and socio-
economic perspectives attributed in each case.
2.2 The operational dimensions of vulnerability
2.2.1 The ‘’starting’’ and ‘’ending’’ points of operational vulnerability
The operational dimensions of vulnerability are differently interpreted by social and biophysical
sciences. The social sciences mostly perceive vulnerability as a situation where the existent
inequalities between developed and developing regions will further exacerbate (O’Brien et al.,
2004). The inherent social and economic differences will make it very hard for communities in
developing countries to cope with the external pressures and climate change. As a result, the
people from developing regions will be further marginalized and restrained from economic
wealth. This vulnerability dimension is mostly acknowledged as a ‘’starting point’’ and as such is
nowadays acknowledged from all scientific disciplines. The input data for the ‘’starting point”
perception are mostly indicators pertaining to the areas of socio-demographics, economic
wealth, infrastructural facilities and information access.
It is frequent that in developing regions the indicators are processed with operational tools used
for poverty analysis. The reason is that a given set of adverse phenomena such as weather
extremes could decrease consumption below a minimum poverty level. Hence, a poverty
analysis could somehow reflect the vulnerability aspects as well the distributional effects and
inequality aspects of an agrarian community in Bangladesh for instance, which is hampered

RiceClima – Deliverable 3.1
10
from extensive droughts or floods. It is however noted that the poverty analysis is mostly
focused on the consumption levels which can hardly represent the socio-ecological vulnerability
by climate change (Brouwer et al., 2007). Instead, statistical exploratory techniques such as
components and factor analysis, generalized linear and non-linear models are nowadays
introduced to this purpose.
The biophysical disciplines put much emphasis on the physical affects while the socio-economic
aspects tend to be of secondary importance. Future emissions coupled with projected
population trends and other technological aspects generate different climate change scenarios
(Eakin and Luers, 2006). The adaptive capacity of an ecosystem is determined through the
robustness and resilience conditions of an ecosystem to cope with the magnitude of the climate
change impacts (Anderies et al., 2004). The biophysical perspective is mostly acknowledged as
the ‘’end point’’ approach.
The operational tools applied in such cases are - more often than not - different climatic models.
The current models have been much evolved so as to forecast climate change on regional and
global scales with a degree of uncertainty (Gallopin, 2006). The most frequent parameters
examined are the temperature, precipitation, wind speed, sunshine exposure and humidity.
However, there are considerable limitations in our understanding of the climate system and the
precision of biophysical parameters especially on a regional level. This becomes more distinctive
in the case of developing countries where the biophysical indicators for the regions are scarce
and often unreliable. (Basak, 2011).
Our study introduces a mixture of theory and data-driven approaches for the development of a
quantitative regional assessment in two regions of rural Bangladesh. In particular, we borrow
elements from both the socio-economic and biophysical perspectives for the construction of a
vulnerability assessment. To this end, we introduced indicators already applied in a multitude of
biophysical and socio-economic studies for the development of vulnerability indexes (Abson et
al., 2012; Deressa et al., 2008; Fellman 2012; Piya et al. 2012). These indicators were sourced
from published sources of similar projects, RiceClima reports and individual research papers.
However, the agricultural conditions in the flood and drought prone areas of Bangladesh should
be also investigated with indicators pertaining to the peculiarities of the case study areas. For
this reason, we also had to adapt our vulnerability assessment for the inclusion of
representative indicators from the selected areas. A household survey was conducted for this
data-driven approach as it is presented in details in the following Section.
Overall, we introduced three groups of vulnerability indicators corresponding to the areas of
exposure, sensitivity and adaptive capacity, respectively. The exposure group in Table 1
represents a set of various biophysical and technical indicators originated from RiceClima
reports. It should be mentioned that the values of the Exposure indicators represent the
weighted mean of a 30-years observations in the selected upazilas.

RiceClima – Deliverable 3.1
11
Table 1. Exposure Indicators
EXPOSURE INDICATORS
Code Abbreviation Unit Explanatory Note TD.-
DD
1
T_annual
Celcius
Mean Temperature for All year
TD
2
P_annual
mm
Mean Precipitation for All year
3
Yloss_Aus
%
Yield Loss compared to the potential yield without
irrigation for T.Aus period
DD
4
Yloss_aman
Yield Loss compared to the potential yield without
irrigation for T.Aman period
5
YL_slight_aus
Indicated level of slight loss in % of years for T.Aus
period
6
YL_mod_aus
Indicated level of moderate loss in % of years for
T.Aus period
7
YL_severe_aus
Indicated level of severe loss in % of years for T.Aus
period
8
YL_slight_aman
Indicated level of slight loss in % of years for T.Aman
period
9
YL_mod_aman
Indicated level of moderate loss in % of years for
T.Aman period
10
YL_severe_aman
Indicated level of severe loss in % of years for
T.Aman period
11
NIR_Aus
mm
Net irrigation requirements for T.Aus period
12
NIR_Aman
Net irrigation requirements for T.Aman period
13
NIR_Boro
Net irrigation requirements for Boro period
Note: TD= Theory-Driven Indicators; DD=Data-Driven Indicators
In Table 2, the sensitivity indicators suggested for our study are displayed. As advised by the
Bangladesh Rice Research Institute (BRRI), the growing of winter rice (boro) or keeping fallow
land in winter time are considered as more sensitive practices to drought conditions than
cultivating water resistant crops. Also the small and tenant farmers are suggested by literature
reviews to be suitable sensitivity indicators for agricultural vulnerability assessments (Biswas et
al., 2009).
Table 2. Sensitivity Indicators
Code
Abbreviation
Unit
Explanatory Note
TD-DD.
1
Cropping Pattern 1
%
Boro- Fallow-T.Aman
DD
2
Cropping Pattern 2
%
Fallow-T. Aus-T.Aman
3
HYV Boro
tn/ha
Rice variety for dry (winter)period
4
Small Farm
%
Small Farmers
TD
5
Tenant Farm
%
Tenancy Farming
Note: Tn/ha= Tonnes per hectare, HYV= High Yield Variety

RiceClima – Deliverable 3.1
12
It is noted that the biophysical indicators used for the sensitivity and exposure measurements
are not related to any future climate change predictions but to past observations. Finally, the
adaptive capacity indicators are displayed in Table 3 as below:
Table 3. Indicators of Adaptive Capacity
C. Indicator Unit Explanatory Note
TD-
DD
C. Indicator Unit Explanatory Note
TD-
DD
SOCIO-DEMOGRAPHIC INDICATORS 12
Access b.
house-
Electr.
Nos.
Access to brick-made
housing- electricity
DD
1 Age
Years
Mean age of adult
family members
TD
13
Infr.Healt
h
Community clinics
per population
TD
2 Schooling
years
Mean schooling years
of adult family
members
14 Infr.post. Post services per
population
3 Farm Exp. Mean Farm experience 15 Infr.veter.
Veterinary centers
per population
4 Family
Size Nos. Mean Family Size 16 Infr.coop
Cooperatives per
population
5 Own Farm % Owning Farmland 17 Infr.agr.ex
t.
Agricultural
extensions per
population
AGRO-ECONOMIC INDICATORS 18 Infr.finan.
Financial schemes
per population
6 Farm Size Ha Mean Farm size per
household
19 Infr.school
Schools per
population
7 Crop
Intens. %
Ratio between the
gross cropped area and
cultivated land
DD 20 Infr.coll. Colleges per
population
8 BCR All
Nos.
Benefit Cost Ratio
crops/ha
TD
21 In-migrat.
People migrating to
the upazila per
population
9 Livestock Livestock amount with
weighted averages
22 Local m. Km
Distance from local
markets
10 Inc. Av. Tk/hs
d
Mean income per
household
23 Bigger m. km
Distance from bigger
markets
INFRASTRUCTURE INDICATORS 24 Hosp. Km KM
Distance from
Hospitals
11
Access
Tub.-Latr.
Nos.
Access to tubewell and
Latrine
DD 25 Town km km Distance from towns
Note : C. = Code; Nos.= Number; TD= Th eory-Driven Indicators; DD=Data-Driven Indicators

RiceClima – Deliverable 3.1
13
As presented in Table 3, 6 indicators are attributed to the socio-demographic situation (1-5, 21),
5 indicators describe agro-economic activities (6-10) and 14 (11-20, 22-25) indicators refer to
infrastructure access. The relatively small number of agro-economic indicators is due to the
summation of individual indicators in some cases. For instance, the Benefit-Cost Ratio of crops
per hectare represents the average ratio of all the cultivated crops (e.g. different rice varieties,
vegetables etc.) on a per hectare basis. Similarly, the Livestock indicator represents the total
amount of livestock (i.e. cows, goat, poultry) given different weights for each animal due to the
various economic importance.
It is noted that there were additional meaningful indicators, like the irrigation management, the
insect and disease frequency and others to be introduced in the vulnerability assessment.
However, the absence of sufficient and appropriate data obstructed their use in the
vulnerability assessment. A description of these indicators is presented in Annex 1.
2.2.2 Principal Component Analysis and Farmer’s preferences
We employ the Principal Component Analysis (PCA) to identify the potential significance of the
adaptive capacity, sensitivity and exposure indicators for the assessment of vulnerability in
selected drought and saline-flood prone areas of Bangladesh. The PCA is a technique presented
in many applications of statistical and econometric inference. PCA has been also extensively
applied in socioeconomic and biophysical vulnerability assessments in regional, national and
global level (Deressa et al., 2008; Abson et al., 2012, Piya et al. 2012; Borja-Vega and De la
Fuente, 2013).
The objective of PCA is to explain potential relations between a large set of independent
variables (in our case indicators) with a latent dependent variable which in our case is the
vulnerability level of each upazila. The comparative advantage of PCA over other exploratory
techniques is that it can rearrange the independent variables for the simplification of the
analysis without losing significant information. This is achieved by lowering the dimensions of
the original data to few principal components.
The components are tested for potential correlations with each independent variable
(indicator), known as factor loadings which are equivalent to standardized regression
coefficients (β weights) in multiple regressions (Beaumont, 2013). The higher values of the
factor loadings (correlation), mean a closer relationship with the principal components. The
correlation threshold for a variable to remain as a loading factor is not quite precise. As a rule of
thumb though, the correlations, positive or negative, presenting a loading factor lower than +/-
0.7 are often discarded from the analysis. The remaining correlations represent the variables
needed to develop the scoring index for the vulnerability assessment.
Also, the number of principal components to interpret the relevant variables is debatable and it
mainly depends on the grading of eigenvalues associated with each component. In practical
terms, the components presenting eigenvalue higher than 1 are approved for explaining the

RiceClima – Deliverable 3.1
14
independent variables (indicators) (Everitt and Hothorn, 2011). More often than not, the
principal components should be as many as to explain 60-70% of the variables (Abson et al.,
2012).
PCA gives also the potential to understand the overall importance of an independent variable
across all the principal components. This is named as Communality for PCA and it is equal to the
sum of all the squared factor loadings for all the principal components related to the
independent variable (indicator). This value is the same as the 𝑅2 in multiple regression. The
value ranges from zero to 1 where 1 indicates that the variable can be fully defined by the
factors. The higher the value, the higher the importance of the relevant indicator.
The data to be used in PCA should be initially standardized and checked for potential
multicollinearity between the independent variables for the avoidance of biased results.
A potential limitation of the PCA method is the weighting importance in the selected variables.
Some authors claim that the PCA may not reflect the higher significance that each variable may
possess, by failing to attribute the actual results of a vulnerability assessment. The introduction
of experts’ judgment (Kaly and Pratt 20009), correlation with past disaster events and use of
fuzzy logic (Eakin and Tapia 2008) are some suggestions for the appointment of weighting
coefficient. However, there is an allegation that the proportion of variance could also constitute
a weighting factor when calculated with the standardized values of each variable (Beaumont,
2013). Moreover, the rotation of the principal components through different techniques
(varimax, equamax) could probably offer a better explanation of the results and improve these
weighting factors. In our case, we have calculated the variances of each indicator with the
standardized values without however considering it as a weighting factor but as a part of the
vulnerability assessment. We understand that the appointment of a weighting factor is of major
importance but we consider that this demands a thorough research which is beyond the scope
of this study.
The PCA can run stepwise for each group of the indicators of exposure, sensitivity and adaptive
capacity as presented in Tables 1,2 and 3 or by merging all the indicators of the three groups in
one. We have selected the stepwise approach with slight modification in an attempt to better
implement Eq.1 in our analysis. To this end, we have run PCA model for adaptive capacity
indicators while the sensitivity and exposure indicators were merged in one group since they
are represented by a negative signalling in Eq.1.
Below, we present an indicative example of PCA assessment for the case of the Adaptive
Capacity Assessment in Godagari upazila (Rajshahi region). As shown in Table 4, all the Adaptive
Indicators have been initially standardized. We then run the PCA analysis to identify which of
the proposed indicators present a loading factor higher than +/- 0.7 and would be eligible for
the vulnerability assessment. In the example, the eligible indicators are highlighted with greyish
colour.

RiceClima – Deliverable 3.1
15
Table 4. Example of Standardized Adaptive Indicators
Age
Scho
oling
years
Farm
Exp.
Famil
y
Size
Own
Farm
Farm
Size
Crop
Inten
s.
BCR
All
Livest
ock
Score
Inc.
Av.
Acc.
Tubwl.
Latrine
Acc.
b
house
Elect.
Acc.
health
0.81
-0.41
-0.10
-0.18
-1
1
1
0.02
-1
1
1
1
-1
Infr.
post
infr.
healt
h
infr.v
et
Infr.c
oop
Infr.a
gr.ext
.
Infr.fi
nan
infr.s
chool
infr.
colleg
e
in-
migrat
Local
m.
bigger
m.
hosp.
Km
town
km
1
-0.11
1
-1
-0.17
1
1
0.26
0.17
1
-1
0.14
0.33
In turn, the factor loadings of these indicators are multiplied with the standardized values for
the calculation of the Adaptive Capacity levels as below:
= -0.416 (Schooling years) * 0.821 (Loading) + (-0.18)(Family Size)* (-0.862) (Loading)+ (-1) (Own
Farm)* 0.876 (Loading) + 1 (Farm Size) * (-0.761) (Loading) + 1 (Crop Intens.)* 0.93 (Loading) + 1
(Invc.Av.)* (-0.9) (Loading) + 1 (Acc.Tubwl- Latrine)* 0.965 (Loading) + 1 (Acc.b.house-Electr.)*
0.967 (Loading) + (-1) (Acc. Health)* (-0.967) (Loading) + (-0.11) (infr.health) * (-0.764)
(Loading)+ 1 (infr.vet) * 0.892( Loading) + (-1) (Infr.coop)* (-0.91303)(Loading) +(-0.17)
(Infr.agr.ext)* (0.809) (Loading) + 1 (Infr.finan)* (-0.943)(Loading) + 0.267 (infr.college)* (-0.823)
(Loading) + 0.178 (in-migrat)* (- 0.816)(Loading) = 1.546, which is the Adaptive Capacity Score
for Godagari upazila in our example.
In the case of indicator’s significance as represented through Communality value, we present an
example of the Schooling Year’s indicator by considering that we have only two principal
components (PC) as below:
Schooling Years = (0.821)2 (PC 1) + (0.499)2(PC 2)= 0.924, Communality Value
It is underlined that the PCA assessment can measure the relative vulnerability between the
examined areas and does not suggest some absolute vulnerability grades based upon a global
vulnerability index.
For a better clarification of PCA results, we have also attempted to elicit farmers’ preferences
with regards to the confrontation of weather extremes and improvement of their adaptive
capacity. The farmers were not asked to assess the performance of the same adaptive indicators
introduced in PCA but to express in a non-determined context their suggestions for a better
adaptation to a changing climate.

RiceClima – Deliverable 3.1
16
3. Case Study
3.1 Vulnerability of Bangladesh
Bangladesh has been repeatedly threatened by natural disasters like flood, salinity and droughts
mainly influenced by the country’s unique geophysical and climatic conditions (Nienke et al,
2006).
In particular, the mountainous ranging of the Tibetan Plateau is drained through a massive river
network spreading all over Bangladesh and finally ending up in the Bay of Bengal. The
occurrence of intense monsoonal periods often augments the drainage effects by leading to
floods mainly in the southern lowland areas (World Bank, 2010). Additionally, saline intrusions
are noticed in the south downstream areas, which are attributed to the higher sea level
elevation in the coastlands. On the other hand, less rainfall and high evaporating losses in the
northwest Bangladesh have entailed seasonal drought events with severe impacts on local
communities (Ramamasy and Bass, 2007).
The extreme events are anticipated to get aggravated by climate change as repeatedly noted in
the literature (Nguyen, 2006; Biswas et al, 2009; Winston et al, 2010). The snow melting in the
mountainous areas of the Tibetan Plateau coupled with erratic and intense monsoons are
expected to constitute the driver for increased flooding. Also, the delayed monsoon conditions
and the higher sea level intrusion are probable to lead in more frequent drought and
salinization effects (MoEF, 2009; Winston et al, 2010). To this end, Bangladesh is struggling to
cope with the current adverse weather conditions while national plans and strategies to
respond to the impacts caused by climate change are developed.
The threatening situation and the efforts undergone by Bangladesh are well quoted in a recent
outcome of the International Institute for Environmental (2013) “…Bangladesh is the most
climate vulnerable country in the world and has consistently been a leader in developing
solutions around community-based adaptation to climate change, national adaptation planning
and offering political leadership as part of the Least Developed Country (LDC) group, which
represents the least developed countries at the climate change negotiations.”
We have selected the regions of Rajshahi and Barisal in the northern and southern parts of the
country, as the most representative areas suffering from drought and flood-saline occurrences
respectively. Within each province, three sub-regions (upazilas) were chosen which could best
ascribe these opposite weather patterns’ impact on a regional level.

RiceClima – Deliverable 3.1
17
3.2 Drought (Rajshahi Region) and flood-saline (Barisal) regions
3.2.1 General Description
In Rajshahi region, the study sites are located in Godagari and Tanore upazilas (lowest
administrative unit) under Rajshahi district and Gomostapur upazila under Chapai Nawabganj
district. The area is characterized by severe drought and is located in north-western Bangladesh
between 88.100 to 88.400 longitudes and 24.200 to 25.000 latitudes (Figure 2).
The site area receives lower amount of precipitation (1500 mm) than the rest of Bangladesh,
while its cropping intensity of 191-262% is more than the national average (180%). The higher
cropping intensity may be attributed to the improved and more widely available irrigation
facilities (deep tubewells) developed by the Barind Multipurpose Development Authority
(BMDA). The number of deep tubewells (DTW) seems to be proportionate with the cropping
intensity in the study location; however, the groundwater table is declining alarmingly due to
over exploitation (CEGIS, 2013).
Figure 2. Rajshahi Region, Source: CEGIS (2013)

RiceClima – Deliverable 3.1
18
The government rules and regulations for irrigation are seldom followed at the field level in
Rajshahi. Lack of groundwater reserves, poor quality seeds, high pest prevalence, low soil
organic matter content, and extreme temperatures are the major problems for agricultural
development. Also, grazing land has decreased tremendously because of increased cropping
intensity while insect pests and diseases have made their appearance more frequently. Of late,
brick fields have also been established in place of crop fields. The removal of top soil for making
bricks is a great concern regarding future agricultural productivity.
In Barisal region, the study sites are located in Amtoli and Patharghata Upazila (lowest
administrative unit) under Barguna disrtict and Kalapara Upazila under Patharghata district. The
study area lies between 89.500 to 90.240 longitudes and 21.460 to 22.180 latitudes (Figure 3).
The study areas are mainly bounded by the Bay of Bengal in the South side, Tetulia river in the
eastern side of Kalapara upazila, Buriswar river in the western side of the Amtoli upazila,
Biskhali and Baleswar river in the eastern and western side of Patharghata upazila, respectively.
Figure 3. Barisal Region, Source: CEGIS (2013)
The area is characterized by an intermediate amount of rainfall (about 2000 mm) and with a
cropping intensity of 173-199%, which is around the national average (180%). The land type of
this area is medium low to medium high land, where maximum flooding depth is about 90 cm
during the monsoon season.

RiceClima – Deliverable 3.1
19
The study areas are within a polder system, which was constructed mainly for flood protection
and to prevent the area from saline water intrusion as presented in Figure 4. At present, the
sluice gates are not properly maintained and many of them are out of order. Moreover,
sedimentation near the sluice gates is increasing day by day, which causes drainage congestion
in the study areas.
Figure 4. Sluice gate in Amtoli upazila (Source: Field Trip in Barisal region, February 2012)
Seasonal intrusion of saline water is damaging the ecological and hydrological balance of the
studied upazilas. Additionally, inadequate saline tolerant varieties, high pest prevalence, lack of
farm machinery, and lack of training on modern crop production technologies are some of the
other bottlenecks of agricultural development (Biswas, 2009).

RiceClima – Deliverable 3.1
20
3.2.2 Agricultural practices in Rajshahi and Barisal regions
The major crops grown in Rajshahi are rice and wheat. However, currently increasing areas of
rice fields are being replaced by mango orchards due to the lower water demand and higher
profitability of the mango fruit. This may have significant implications for the future rice
production in Bangladesh. The minor crops are potato, tomato, gram, maize, and eggplant.
The major cropping patterns in Godagari was the Boro – Fallow – T. Aman (42%) followed by
Boro – T. Aus – T. Aman (38%). Similar patterns were also observed in Tanore area. However, in
Gomostapur area the highest coverage was the Boro – Fallow – T. Aman pattern (40%) followed
by Boro – fallow – T. Aman (34%).
During the last 15 years, the amount of rainfall and its distribution pattern, temperature and
drought duration, has changed unfavorably to growing traditional rice variaties. In the mid-90s
farmers mostly cultivated Kalokuchi, Shaitta, Dharial, Sonasail, Mugi, Raghusail, Magusail,
Jhingasail, BR10, BR11 and IR20 rice varieties. At present, Pariza, Sada Sawrna, Guti Sawrna,
BINA dhan7, BRRI dhan28, BRRI dhan36 and BRRI dhan39 are mostly grown.
Farmers also grow short duration rice varieties in attempt to reduce the effect of drought
conditions. Moreover, they are growing tomato, mustard, and potato to minimize the need for
irrigation water in the dry season.
In the case of Barisal, rice is the major crop. The minor crops are pulses, potato, chili, mustard,
sunflower, watermelon, groundnut and spices, etc. Pulse-Fallow-T. Aman (55%) is the major
cropping pattern followed by Winter Crops-Fallow-T. Aman (20%) in Kalapara upazila. In the
case of Amtoli upazila, Grass pea-T. Aus-T. Aman (48%) is the major pattern followed by Fallow-
T. Aus-T. Aman (24%). The dominant cropping pattern in Patharghata upazila is Fallow-Fallow-T.
Aman (40%) followed by Grass pea-Fallow/T. Aus-T. Aman pattern (25%).
Alike Rajshahi, change in climate conditions in the past few years have adversely affected rice
growing via changes to the rainfall and its distribution pattern, temperature, and drought
duration. Farmers earlier cultivated rice varieties such as Kajalsail, Sadamota, Lalmota,
Laxmibilash, Rajasail, Shaitta, Brindamoni, Rangalaxmi, Shitabhog, Kutiagni, Betichikon,
Jhingasail, Matichak etc and a few HYV rice varieties such as BR11, BR22. At present, Sadamota,
Vajan, BR11, BR22, BR23, BRRI dhan27, BRRI dhan40, BRRI dhan41 and BRRI dhan49 are
commonly grown, which cover 60-99% of the land in the T. Aman season and about 90% of the
land in the T. Aus season.
The growth periods of different rice and non-rice crops in Rajshahi and Barisal are shown in
Figure 5. As presented, the boro rice needs longer growth period than the T. Aman rice.
Mustard, potato and tomato need comparatively short growth duration. It is further presented
that not exactly the same crops are cultivated in both regions due to different geophysical and
weather conditions.

RiceClima – Deliverable 3.1
21
Crop
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Rajshahi Region
Boro Rice
T. Aus Rice
T. Aman Rice
Wheat
Maize
Mustard
Potato
Tomato
Barisal Region
Boro Rice
T. Aus Rice
T. Aman Rice
Sunflower
Pulses (Grass Pea)
Potato
Vegetables
Figure 5. The growth period of cultivated crops in Rajshahi and Barisal regions
3.3 Primary and Secondary Data
The primary data was elicited from a household survey analysis conducted in the two regions. In
each region, 100 farmers from different farm sizes (small, medium and large) were queried
through a random sampling method. The collection of the survey responses was carried out
from February to March 2013.
The survey period covered 3 agricultural crop seasons. These are: i) Kharif-I: 16 March to 30
June); ii) Kharif-II: 01 July to 15 October and iii) Rabi: 16 October to 15 March. The survey data
covered Rabi/Boro, 2011; Kharif-I, 2012 and Kharif-II, 2012 seasons. The crops cultivated in
these seasons are boro rice in October-November to harvesting time April-May, then the aus
rice during March-April to July-August and lastly the aman rice in July-August to November-
December.
The secondary data was originated from the following sources:
- Scientific publications on socioeconomic and biophysical indicators.
- Bangladesh Meteorological Office (BMO), the Directorate of Agricultural Extension
(DAE) and other government publications.
- Internal project reports of the RiceClima project on climate change scenarios,
hydrological and crop modelling.

RiceClima – Deliverable 3.1
22
4. Results
4.1 Descriptive results
The amount of respondents in the household survey were rather well balanced among the
upazilas and the villages situated in each area as presented in Table 5.
Table 5. Sampling distribution
Drought prone study area
(Rajshahi region)
Flood-Saline prone study area
(Barisal region)
Upazila
Block/Village
Farmer
(Nos.)
Upazila
Block/Village
Farmer (Nos.)
Godagari
All
30
Kalapara
All
32
Nabagram
7
Nilganj
10
Iyhy
11
Tiakhali
12
Bidirpur
12
Chokomoya
12
Tanore
All
34
Amtoli
All
34
Kalma 8
Uttar
Tiakhali
13
Kaliganj 11
Choto
Nilganj
9
Mandomala
15
Nalbania
12
Gomastapur
All
36
Patharghata
All
34
Zinarpur
11
Char Doani
12
Chotodadpur 14
Char
Lathimara
12
Rohanpur
11
Kalomega
10
Total
100
Total
100
Source: Field Survey, 2013
Further, some descriptive statistics of agronomic and economic interest are presented for a
better understanding of the socio-economic situation in the study areas. It is noted that these
descriptive results are not necessarily presented as well in the PCA assessment in the form of
indicators. In particular, many of the descriptive results just provide a better understanding of
the study sites but they would not be meaningful as indicators for the vulnerability assessment.
In few cases, as indicatively in the irrigation and pest and disease descriptive results,
vulnerability indicators could be shaped. However, the data was provided only on a regional
level and thus the indicators would be meaningless for a vulnerability assessment on a upazilla
level.

RiceClima – Deliverable 3.1
23
Farm size and landholders
The average farm size and cultivated land of sample farmers are shown in Table 6. As presented,
the total area of cultivated land in Godagari, Tanore and Gomostapur in Rajshahi region is
nearly identical to the cultivated land of the upazilas in the Barisal region. In all the Rajshahi
upazilas, the share of owned land was greater than that of the rented/mortgaged land.
Conversely, the share of rental farming lands is greater in the Barisal region, except in
Patharghata upazila. The average farm size in both regions is relatively similar between the six
upazilas although with some variations.
Table 6. Farm ownership status (ha)
Location
Farmer
(No.)
Own
land
(ha)
Rented/
Mort. in
land (ha)
Rented/
Mort. out
land (ha)
Total
cultivated
land (ha)
Average.
farm size
(ha)
a
b
c
d
e=b+c-d
e/a
Rajshahi Region
Godagari
30
29.07
21.33
1.07
49.33
1.64
Tanore
34
38.47
7.80
1.20
45.07
1.33
Gomostapur
36
28.67
16.00
-
44.67
1.24
Total
100
96.21
45.13
2.27
139.07
1.39
Barisal Region
Kalapara
32
24.67
25.60
-
50.27
1.57
Amtoli
34
23.80
24.47
0.40
47.87
1.40
Patharghata
34
35.33
4.27
-
39.60
1.16
Total
100
83.80
54.34
0.40
137.74
1.38
Source: Field Survey, 2013
Table 7 gives a more precise allocation of farm size per landholder by dividing them between
small, medium and large farmers. The number of large and small farmers seems to be higher in
Rajshahi than in Barisal.
Table 7. Farm sizes based on cultivated land (ha)
Location
Small Farm (0.61 – 1.0
ha)
Medium Farm (1.01 –
3.0 ha)
Large Farm (3.01ha and
above)
Number
Percent
Number
Percent
Number
Percent
Rajshahi Region
Godagari
8
26.67
19
63.33
3
10.00
Tanore
15
44.12
17
50.00
2
5.88
Gomostapur
20
55.56
14
38.89
2
5.55
Average
14.33
42.12
16.67
50.74
2.33
7.14

RiceClima – Deliverable 3.1
24
Barisal Region
Kalapara
8
25.00
22
68.75
2
6.25
Amtoli
13
38.24
20
58.82
1
2.94
Patharghata
15
44.12
19
55.88
-
-
Average
12.00
35.79
20.33
61.15
1
3.06
Source: Field Survey, 2013
Crop Cultivation
Tables 8 shows the major crops in Rajshahi grown in the selected sites with their existing yield
level. Different types of crops are grown between regions and their yield also varies depending
upon location. The major rice varieties grown in the study sites were Local T. Aus (Pariza), HYV
T. Aus (BRRI dhan48), HYV T. Aman BRRI dhan49 and 56), Sawrna (Guti Sawrna, Ranjit Sawrna
and Lal Sawrna), HYV Boro (BRRI dhan28 and BINA-7) etc. The major non-rice crops were Chick
Pea, Mustard, Tomato and Wheat.
Table 8. Major crops grown and yield level in Rajshahi region
Crops
Popular varieties
Area (ha)
Average
yield (t/ha)
Yield range
(t/ha)
A. Godagari site
Local T. Aus
Pariza
12.20
4.28
3.90 – 5.85
HYV T. Aus
BRRIdhan48
1.33
4.80
4.25 – 4.90
HYV T. Aman
BRRIdhan49 and
BRRIdhan56
1.80
4.54
4.10 – 5.40
Sawrna (aman)
Guti Sawrna, Ranjit
Sawrna and Lal
Sawrna
35.20
5.13
4.80 – 5.55
HYV Boro
BRRIdhan28 and
BINA7
14.54
5.30
3.90 – 6.30
Chickpea
5.53
1.16
0.90 – 1.80
Mustard
6.33
1.07
0.90 – 1.31
Tomato
5.94
21.16
16.50 – 30.00
Wheat
12.20
3.66
3.08 – 4.80
Total Cropped Area
95.07
B. Tanore site
HYV T. Aman
BRRIdhan49
6.47
5.45
4.80 – 6.00
Sawrna (aman)
Guti Sawrna, Ranjit
Sawrna and Lal
Sawrna
25.93
5.26
4.50 – 5.63
HYV Boro
BRRIdhan28 and
BINA7
26.53
4.98
3.90 – 6.00

RiceClima – Deliverable 3.1
25
Chickpea
0.53
1.80
1.3 – 2.00
Maize
4.20
5.40
4.90 – 6.10
Mustard
6.20
1.04
0.90 – 1.20
Potato
6.20
17.84
16.05 – 18.00
Wheat
4.13
3.53
3.15 - 4.80
Total
80.20
C. Gomastapur site
Local T. Aus
Pariza
12.40
5.24
4.89 – 5.70
Local Aman
Fine and aromatic
variety
5.87
2.25
1.95 – 2.55
HYV T. Aman
BRRIdhan34/36
0.27
5.62
5.40 – 5.85
Sawrna (aman)
Guti Sawrna, Ranjit
Sawrna and Lal
Sawrna
38.80
5.36
4.50 – 5.70
HYV Boro
BRRIdhan28 and
BINA7
3.47
5.39
4.80 – 6.00
Chickpea
1.07
1.65
1.05 – 1.80
Mustard
4.00
1.11
1.05 – 1.80
Wheat
12.80
3.51
3.00 – 3.60
Total
78.67
Source: Field Survey, 2013
Respectively, in Barisal region the major rice varieties were Local T. Aus (Mala China), Local T.
Aman (Kazal Shail, Sadamota, Lalmota, Vajan and Tepu), HYV T. Aman (BR11/23 and
BRRIdan40/41), HYV Boro (BRRI dhan28). The major non-rice crops were pulses and vegetables.
Table 9 shows the average yield and range of yield of each crop in Barisal, which reveals that,
minimum and maximum yield differences were high in each crop.
Table 9. Major crops grown and yield level in Barisal region
Crops
Popular varieties
Area (ha)
Average
yield (t/ha)
Yield range
(t/ha)
A. Kalapara site
Local T. Aman
Sadamota, Lalmota,
Vajan,Tepu
36.27
2.81
2.40 – 3.60
HYV T. Aman
BR11 and BRRIdan41
15.20
3.37
3.00 – 3.60
HYV Boro
BRRIdhan28
11.73
4.20
3.90 – 4.80
Pulses (Grass Pea)
1.80
1.09
0.98– 1.20
Vegetables
2.53
12.18
10.75 – 13.50
Total
67.53
B. Amtoli site

RiceClima – Deliverable 3.1
26
Local T. Aus
Mala China
9.60
3.30
3.00-3.60
Local T. Aman
Sadamota, Lalmota
27.33
2.70
2.40-3.00
HYV T. Aman
BR11, BRRI dhan40/53
16.37
3.80
3.10-4.50
HYV Boro
BRRI dhan28
2.30
4.27
3.90 – 4.80
Pulses (Grass Pea)
6.40
1.18
1.00-1.75
Vegetables
1.00
12.08
11.05– 13.25
Total
63.00
C. Patharghata site
Local T. Aman
Kajalsail, Sadamota,
Lalmota
32.53
3.45
2.40-4.50
HYV T. Aman
BR11, BR22 and BRRI
dhan44
4.33
4.60
3.40-5.80
Potato
2.47
9.75
9.00-10.50
Sunflower
1.20
1.80
1.65-1.95
Pulses (Grass Pea)
19.87
1.24
0.98-1.50
Vegetables
1.27
10.75
9.50-12.00
Total
61.67
Source: Field Survey, 2013
Crop profitability
The profitability of crop production was examined through the Benefit-Cost ratio indicator as
presented in Tables 10 and 11. In Rajshahi region, non-rice crops were more profitable (BCR
ranged from 1.37 to 2.28) than rice crops (BCR ranged from 1.15 to 1.25). Among rice crops HYV
boro rice were less profitable than aus or T.Aman rice due to the high irrigation and fertilizer
costs associated with boro rice production (Table 10).
Table 10. Location wise crops grown and gross margin (Tk/ha) in Rajshahi region
Crops
Yield
(t/ha)
Sale price
(Tk/kg)
Total
Variable
cost ( TVC)
(Tk/ha)
Gross
return
(GR)
(Tk/ha)
Gross
Margin
(GM = GR-
TVC)
(Tk/ha)
Undiscoun-
ted BCR =
GR/TVC
Godagari site
T. Aus (Pariza)
4.28
16.25
61,525
73,830
12,305
1.20
T. Aman
5.13
16.25
70,794
88,493
17,699
1.25
HYV Boro
5.30
16.12
77,552
90,736
13,184
1.17
Mustard
1.07
45.06
33,756
49,284
15,528
1.46
Tomato
21.16
8.25
76,566
1,74,570
98,004
2.28
Wheat
3.66
18.80
52,896
72,468
19,572
1.37

RiceClima – Deliverable 3.1
27
Tanore site
T. Aman
5.45
16.25
77,060
94,013
16,953
1.22
HYV Boro
5.70
16.12
84,856
97,584
12,728
1.15
Maize
5.40
12.50
41,657
72,900
31,243
1.75
Mustard
1.04
45.06
31,935
47,902
15,967
1.50
Potato
17.84
9.40
78,362
1,67,696
89,334
2.14
Wheat
3.53
18.80
49,924
69,894
19,970
1.40
Gomastapur site
T. Aus (Pariza)
5.24
16.25
73,488
90,390
16,902
1.23
T. Aman
5.42
16.25
74,202
93,495
19,293
1.26
HYV Boro
5.60
16.12
85,600
95,872
10,272
1.12
Mustard
1.11
45.06
35,753
51,127
15,374
1.43
Wheat
3.51
18.80
47,930
69,498
21,568
1.45
Source: Field Survey, 2013
In Barisal region, non-rice crops were also more profitable (BCR ranged from 2.10 to 2.75) than
rice crops (BCR ranged from 1.18 to 1.30). Among rice crops HYV T. Aman rice was more
profitable (BCR1.30) than aus rice (BCR 1.20) or boro rice ( BCR 1.8) (Table 11b). This was
happened due to rain fed cultivation practice and use of low doses of fertilizer, which incurred
low costs associated with T. Aman rice production.
Table 11. Location wise crops grown and gross margin (Tk/ha) in Barisal region
Crops
Yield
(t/ha)
Sale
price
(Tk/kg)
Total
Variable
cost ( TVC)
(Tk/ha)
Gross
return
(GR)
(Tk/ha)
Gross
Margin
(GM = GR-
TVC)
(Tk/ha)
Undiscoun-
ted BCR =
GR/TVC
Kalapara site
Local T. Aman
2.81
16.15
39,502
48,192
8,690
1.22
HYV T. Aman
3.37
15.75
43,422
56,448
13,026
1.30
HYV Boro
4.20
15.50
58,729
69,300
10,571
1.18
Pulses
1.09
35.42
18,385
38,608
20,223
2.10
Vegetables
12.18
10.14
44,911
1,23,505
78,594
2.75
Amtoli site
Local T. Aus
3.30
16.10
47,025
56,430
9,405
1.20
Local T. Aman
2.70
16.15
37,646
46,305
8,659
1.23
HYV T. Aman
3.80
15.75
50,516
63,650
13,134
1.26
HYV Boro
4.27
15.50
58,713
70,455
11,742
1.20
Pulses
1.18
36.30
19,122
42,834
23,712
2.24
Vegetables
12.08
11.50
52,423
1,38,920
86,497
2.65
Patharghata site

RiceClima – Deliverable 3.1
28
Local T. Aman
3.45
16.25
48,385
59,513
11,128
1.23
HYV T. Aman
4.60
16.50
64,919
80,500
15,581
1.24
Potato
9.75
12.10
49,989
1,17,975
67,986
2.36
Sunflower
1.80
30.50
22,941
58,500
35,559
2.55
Pulses
1.24
38.67
19,572
47,951
28,379
2.45
Vegetables
10.75
10.75
44,108
1,15,563
71,455
2.62
Source: Field Survey, 2013
Irrigation management
In turn, Table 12 shows the main source of irrigation water along with the common type of
distribution systems. In Rajshahi, groundwater is the main sources for crop irrigation and the
supply is conducted with buried pipe systems. Both Deep Tube wells (DTW) and Mini DTW are
used for irrigation. In few cases, surface water is used for irrigation purpose in some areas
adjacent to the pond and canals.
Conversely, in Barisal surface water is the main sources for crop irrigation. Irrigation water is
distributed with open canal systems. However, recently the irrigated agriculture has not
become a common practice. The sea intrusion has increased the salinity of the surface waters to
that extend that is not suitable for irrigation purposes. Low lift pumps (LLP) are used for
pumping surface water usually from small ponds where the salinity is rather low.
Table 12. Source-wise irrigation coverage under study sites
Locations
Irrigation coverage (%)
Irrigation
Device
Distribution
system
Surface water
Ground water
Rajshahi Region
Godagari
8
92
DTW and Mini
DTW Buried pipe
Tanore
5
95
Gomastapur
6
94
Barisal Region
Kalapara
93
7
LLP Open canal
Amtoli
95
5
Patharghata
96
4
Note: DTW = Deep Tubewell (forced mode pump); Mini DTW = Low capacity submergible pump; and LLP = Low Lift
Pump (suction mode pump)
Source: Field Survey, 2013

RiceClima – Deliverable 3.1
29
Rice diseases and insects
The respondents were also asked about the impact of rice diseases and its incidence level as
presented Table 13. In both Rajshahi and Barisal regions sheath blight was the most common
disease followed by blast. The present incidence level of rice blast was almost similar compared
to last 15 years incidence but presently, sheath blight emerged as a major disease for rice
because of climatic and ecological variations occurred over this time period.
Table 13. Information about rice disease incidence level
Name of
Disease
Season
Variety
Yield Loss (%)
Control Measures
15 Years
Back
Present
Rajshai Region
Sheath
blight
T. Aman
BR11
-
20-25
Chemical control (Nativo,
Hexa) and Biological
Control (Drainage Water
from field)
BRRI dhan52
Sawrna
25-30
Blast
T. Aman
Arometic rice
(Including HYV
and Local)
15-20
15-20
Nothing
Boro
BRRI dhan29
-
10-12
Chemical control
(Trooper, Nativi, Zeel etc.)
BRRI dhan28
4-5
Bacterial
blight
T. Aman
BRRI dhan52
5-6
Nothing
Boro
BRRI dhan28
5-6
Bakanae
Boro
BRRI dhan29
2-3
Uprooting of Infected
Tillers
Barisal Region
Sheath
blight
T. Aman
BR11
-
20-25
Chemical Control (Nativo,
Hexa) and Biological
Control (Drainage Water
from Field)
Sadamota
12-15
Blast
T. Aman
All Arometic
rice (BRRI
dhan34,
Sakkorkhorai,
Kalizira)
15-20
15-20
Nothing
Boro
BRRI dhan29
-
10-12
Chemical control
(Trooper, Nativi, Zeel)
BRRI dhan47
20-25
Bacterial
blight
T. Aman
BR11
5-6
Nothing
Boro
BRRI dhan28
5-6
Source: Field Survey, 2013

RiceClima – Deliverable 3.1
30
The respondents were also asked about the rice insects and its incidence level. As presented in
Table 14, the farmers’ views were almost identical between the two regions. Brown plant
hopper (BPH) was the most common insect followed by goll midge. Table 14 also shows that the
incidence level of rice hispa was higher in the past but nowadays has been drastically reduced
because of unfavourable ecosystem for its development.
Table 14. Information about rice insect’s incidence level
Name of
Insects
Season
Variety
Yield Loss (%)
Control Measures
15 Years
Back
Present
Rajshahi and Barisal Regions
Brown plant
hopper
(BPH)
Boro
BRRI
dhan29
-
25-30
Chemical control (Mipsin) and
Biological control (Drainage Water
from Field)
Yellow stem
borer
All
seasons
All
variety
5-10
5-10
Chemical control (Furadan) and
Biological control (Parching)
Rice Hispa
T.
Aman
10-20
-
Chemical control (Diazinon) and
Biological control (Leaf Clipping)
Goll midge
15-20
15-20
Chemical control
(Diazinon/Furadan) and Biological
control (Drainage Water from
Field)
Source: Field Survey, 2013
4.2 PCA Results
The results of PCA suggest that a large amount of the indicators enclosed in the Adaptive
Capacity group are satisfactorily explained (66.4%). In particular, 16 out of the 25 adaptive
variables are statistically significant and can be identified as potential drivers for the
vulnerability levels of each upazila (see Table 1, Annex 2). The crop intensity, the access to
housing facilities and the presence of financial institutions are given the highest importance.
In the case of the Exposure and Sensitivity indicators, 14 out of the 19 variables could be well
explained (84%) by the PCA analysis as potential determinants (see Table 2, Annex 2). Also, the
standardized values of all the variables from each group are presented in Annex 3.
We then assess the overall significance of each indicator through the communality values as
presented in Table 15. The five most important ones are presented for the Adaptive capacity
group while an equal amount is also denoted for the Exposure and Sensitivity groups. For the
case of Adaptive indicators, the household’s livelihood conditions are most noticeable. It is then,
the health and veterinary access as of almost equal importance while the farm ownership is also
signified. In the case of Exposure and Sensitivity group, an almost equal merit of significance is
attributed to the five most important indicators. Particularly, the indicators related to the yield

RiceClima – Deliverable 3.1
31
loss and the irrigation requirements of T.Aus rice crop season are noticed while the annual
precipitation and temperature indicators are hinted.
Table 15. Significance of Vulnerability Indicators
Adaptive Capacity
Communality
Value
Exposure-Sensitivity Communality Value
Access Tubewell_Lartrine
0.986
Yloss_Aus
0.999
Access Pacca_Electricity
0.986
YL_severe_aus
0.997
Access health
0.986
NIR_Aus
0.997
infr.vet
0.976
P_annual
0.997
Own Farm
0.965
T_annual
0.995
The scoring of the vulnerability levels for each upazilla is derived by the subtraction of the
exposure and sensitivity indicators from the adaptive capacity as presented in Figure 6. When
each group of indicators is separately examined for each upazila, the lowest adaptive capacity is
given to Amtoli while further aside follows the Kalapara, both situated in Barisal region. This
could be probably attributed to the low mean annual income and the poor performance of
infrastructural indicators in these two upazilas which seem to hamper the adaptive potential.
The poor infrastructure could be also in part responsible for the low adaptive capacity score in
Patharghata upazila while also the small farm experience seems to be a contributor. However,
the other demographic and agro-economic indicators perform much better in Patharghata than
in the two other upazilas of Barisal region and thus there is a better adaptive capacity scoring.
In the case of adaptive capacity indicators in Rajshahi region, Godagari upazila seem to score
remarkably lower than the other two upazilas but still in higher levels than the Barisal region.
This low score seems to be rendered on the limited access to household facilities (latrine,
water, electricity) while also the education and crop intensity indicators perform
comparatively lower than the two other Rajshahi upazilas. The high scoring of Tanore and more
distinctively Gomastapur appears to be the result of a satisfactory performance in most of the
demographic and agro-economic indicators.
Reversely, all the Rajshahi upazilas attain a remarkably low scoring in the exposure and
sensitivity indicators which counterbalances the positive performance of the adaptive capacity.
This is much attributed to the unfavorable climatic conditions for irrigated agriculture recorded
for the last 30 years in Rajshahi which have hindered the potential of higher agricultural
production. On the contrary, the milder climatic conditions in Barisal region and the much
lower need on irrigation have resulted in lower production loss.
Overall, the less vulnerable areas are shown in Barisal firstly by Patharghata while closely
behind follows Kalapara upazila. Unlikely, Amtoli upazila although belonging to Barisal region,
seems to perform worse than Tanore and Gomastapur in Rajshahi. The scoring of Godagari
vulnerability is noticeably the lowest among all other upazilas.

RiceClima – Deliverable 3.1
32
Figure 6. PCA Assessment Results
It is mentioned that the vulnerability scoring between upazilas does not represent an absolute
value index but the relevant performance between the areas.
4.3 Farmers’ preferences
The farmers’ preferences for the improvement of their adaptive capacity indicate a strong
inclination to the agricultural activities. As presented in Table 16 there is a clear indication of
intertwinement between the need for farming improvement and the concept of adaptation in a
changing climate. Most of the suggestions pertaining to the pricing of agricultural inputs and
products while the technological support is also of major importance. Another area of interest
is the improvement of infrastructure in irrigation systems on surface water conservation and
provision of better groundwater systems. Finally, the access to better seeds and the
arrangement of educational seminars in technologies are also suggested as priorities for a
better adaptation to climate change.
When the preferences are allocated on an upazila level, it appears that the respondents of
Patharghata upazila are in the highest desire of the suggested initiatives and especially the
freshwater conservation measures. Broadly, the upazilas belonging to Barisal region are much
more interested than Rajshahi in participating to all the relevant suggestions but for water
infrastructure. The highest grades amongst upazilas are shaded with greyish color.
1.55
9.19
10.56
-6.61
-11.41
-2.38
-9.13 -8.92
-10.69
10.02 9.42
6.69
-7.59
0.27 -0.13
3.41
-1.99
4.31
-12
-7
-2
3
8
13
Godagari Tanore Gomastapur Kalapara Amtoli Patharghata
Adaptive Capacity Exposure+Sensitivity Vulnerability

RiceClima – Deliverable 3.1
33
Table 16. Suggestions for improvement of adaptive capacity
Areas Suggestions
Rajshahi Region (%)
Barisal Region (%)
God. Tan.
Gom
.
Mean
Values
Kal. Amt. Path.
Mean
Values
Market
Availability of
agricultural inputs at
reasonable / subsidized
price (seed, fertilizer,
water, pesticides etc.)
65 68 63 65 75 68 70 71
Market
Ensure reasonable
output prices and
profitability of
agricultural
commodities
82 78 84 81 82 80 84 82
Market
Availability of farm
machineries at
subsidized price or on
rental basis (power
tillers, pumps, sprayer,
reaper, thresher etc.)
60 63 58 60 60 62 68 63
Water
Facilities
Irrigation infrastructure
development (setting of
pumps, ensure
electricity, improved
canal system etc,)
85 82 90 86 75 78 80 78
Water
Facilities
Conservation of water
(rain water harvest,
embankment, sluice
gate, canals etc.)
55 52 65 57 95 92 96 94
Seeds
Availability of new high
yielding and short
duration rice varieties
75 72 78 75 85 82 88 85
Education
Intensive farmers’
training on agricultural
production technologies
75 78 72 75 75 88 72 78

RiceClima – Deliverable 3.1
34
5. Discussion
The descriptive statistics gave an overall impression of the agricultural conditions in both
regions. Not major differences are presented between the two sites except for the irrigation
practices. The irrigation factor appears to play a major role in the production costs of Rajshahi
region which is anticipated to get more important in the future because of the decreasing water
reserves and the higher drought frequency. Further, the need to confront with the emerging
disease and insects’ incidences seems to be commonly shared between the two regions.
The PCA results have demonstrated the significance of the adaptive capacity, sensitivity and
exposure indicators for the attribution of the vulnerability assessment. In particular, the higher
scoring of Rajshahi in adaptive indicators seemed incapable of signifying a better vulnerability
status of Rajshahi over Barisal region. The average production loss of Rajshahi in the last 30
years has offset any comparative advantage emerging from the adaptive capacity performance.
There are some methodological limitations of PCA use in the current study. Initially, there is a
considerable uncertainty on the appropriateness and relevance of the suggested indicators. This
is a broader issue standing on most of the vulnerability assessments irrelevantly to the
suggested measurement approach. There is a common understanding that many indicators
might enclose a degree of subjectivity in an effort to portray case-specific conditions of
vulnerability. We acknowledge these potential biases and as a mitigation effort, we have
introduced indicators spotted in other similar vulnerability assessments by attempting to reduce
the case-specific ones.
It is also argued that the vulnerability is a dynamic concept and a static assessment like PCA
could hardly explain any future changes. To this end, it is firmly explained that we have
estimated the present vulnerability levels in each upazila based on the current demographic,
agro-economic and infrastructural indicators and past observations of biophysical parameters.
Although it is understood that any future observations may not highly deviate from the assessed
ones, it is explicitly mentioned that the vulnerability assessment does not represent any future
status of the selected upazilas.
However, there is the potential to provide some future vulnerability scenarios based on the
performance of the examined indicators. For instance, we have tried to increase the
performance of three out of the five most significant indicators presented in Table 15 for Barisal
region only. Namely, the performance of Access to Health, Veterinary and the farm ownership
was improved by 30% for each of the three Barisal upazilas. As presented in Figure 7 with the
purpled colour column, the Adaptive Capacity has been now slightly to moderately improved in
the three Barisal upazilas. Such scenario analyses could greatly help the policy makers to
understand in which particular indicators should pay attention and invest for a better
vulnerability performance.

RiceClima – Deliverable 3.1
35
Figure 7. Vulnerability Scenarios for Barisal region
The aforementioned methodological and policy relevant concerns were also met in many similar
vulnerability assessment studies. A regional vulnerability assessment in Ethiopia notes the lack
and unreliability of primary data (Deressa, 2008). Another vulnerability analysis of rural
households in Nepal signifies the importance of scenarios for the identification of agro-
economic and infrastructural areas to be improved (Pyia et al., 2012). Other applications of PCA
in national trans-national level were enriched with Geographical Information Systems (GIS) in an
attempt to overcome the static nature of the results (Abson et al. 2012; Borja-Vega and De la
Fuente, 2013).
The farmers’ preferences came to signify the need of both regions to invest on agricultural
market mechanisms, irrigation facilities, seeds and educational seminars for a better adaptation
to climate change. These elements were coincidentally also represented as statistically
significant indicators in PCA analysis. For instance, the Mean Annual Income indicator is highly
related and affected by the market mechanisms which are suggested by farmers. In turn, the
suggested improvements in irrigation facilities are well represented by the Net Irrigation
requirements in the Exposure group of indicators. It is mainly that the farmers pointed out
some broader interventions that could help in better adaptation while the PCA indicators were
focused on specific aspects of these interventions.
It is noted that the suggested improvements by farmers on market conditions seem to mostly
target on the increase of their welfare, an objective which is better viewed through a poverty
analysis.
1.55
9.19
10.56
-6.61
-11.41
-2.38
-1.01
-5.94
-2.01
-9.13 -8.92
-10.69
10.02 9.42
6.69
-7.59
0.27 -0.13
3.41
-1.99
4.31
-12
-7
-2
3
8
13
Godagari Tanore Gomastapur Kalapara Amtoli Patharghata
Adaptive Capacity Adaptive Capacity_Scenario Exposure+Sensitivity Vulnerability

RiceClima – Deliverable 3.1
36
6. Concluding remarks
Climate change impacts are already occurring in Bangladesh and is likely to continue with
greater severity in future. Those who are most vulnerable to the adverse impacts of climate
change are generally the agrarian regions. Therefore, the development of tools to assess socio-
ecological vulnerability, such as in this report, could help to identify measures to create
resilience and mitigate the impacts of climatic vagaries.
The current study attempted to describe in a quantitative manner the vulnerability status of the
drought and saline-flood prone selected upazilas in Bangladesh. Also, some descriptive results
and farmers’ preferences attempted to better clarify and cross-check the vulnerability
assessment.
The findings for the drought prone regions in Rajshshi signified the need to improve the access
to household facilities and moreover the urgency for better groundwater management so as to
meet the current production loss. In particular, as the groundwater availability is gradually
diminishing, HYV boro rice cultivation could be hardly irrigated in the following years. More
efficient irrigation schemes should be developed to meet the current demand while better
water resistant rice varieties should be introduced. Also, cash crops like wheat, maize, mustard,
potato, tomato should be better promoted as a promising response to water scarcity and a
more profitable alternative to rice cultivation.
The introduction of cash crop is also encouraged in Barisal region for the improvement of the
agricultural income. Moreover, the need for better infrastructure and sound water conservation
measures are also prioritized in Barisal region. Also, the education on new technologies in
cultivation through training, demonstration and field days is highly desired.
The current vulnerability assessment is a context-specific approach and the data, methods and
results cannot be transferred without any proper adjustments to other similar studies.

RiceClima – Deliverable 3.1
37
References
Abson D.J., Dougill A.J.and Stringer L.C (2012) Using Principal Component Analysis for
information-rich socio-ecological vulnerability mapping in Southern Africa, Applied Geography
(2012) 1-10
Anderies, J., Janssen, M., Ostrom, E., (2004). A framework to analyse the robustness of socio-
ecological systems from an institutional perspective. Ecology and Society 9, 1–18
Basak, (2011) Climate Change Impacts on Rice Production in Bangladesh: Results from a Model,
Unnayan Onneshan-The Innovators, Bangladesh
Borja-Vega C. and De la Fuente A., (2013) Municipal Vulnerability to Climate Change and
Climate-Related Events in Mexico, Policy Research Working Paper 6417, The World Bank,
Washington
Biswas J.K., Ali M.A., Mahbub, A.M., Ansari, T.H., Raghib Hassan A.W. and Islam, M.S. (2009)
Results and Findings from BRRI-FAO Collaboration under Project Improved Adaptive Capacity to
Climate Change for Sustainable Livelihoods in the Agricultural Sector (LACC II), BRRI, Joydebpur,
Gazipur-1701 & DAE, Khamarbari, Dhaka-1207
Brouwer R., Akter, S., Brander, L. and Haque, E. (2007) Socioeconomic vulnerability and
adaptation to environmental risk: a case study of climate change and flooding in Bangladesh,
Risk Analysis, 27(2):313-26.
Beaumont. R., (2012) An introduction to Principal Component Analysis & Factor Analysis Using
SPSS 19 and R, http://www.floppybunny.org/robin/web/virtualclassroom/stats/statistics2/pca1.pdf
[accessed 2/12/2013]
Cardona, O.D., M.K. van Aalst, J. Birkmann, M. Fordham, G. McGregor, R. Perez, R.S. Pulwarty,
E.L.F. Schipper, and B.T. Sinh, (2012) Determinants of risk: exposure and vulnerability. In:
Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation
[Field, C.B., V. Barros, T.F. Stocker, D. Qin, D.J. Dokken, K.L. Ebi, M.D. Mastrandrea, K.J. Mach,G.-
K. Plattner, S.K. Allen, M. Tignor, and P.M. Midgley (eds.)]. A Special Report of Working Groups I
and II of the Intergovernmental Panel on Climate Change (IPCC). Cambridge University Press,
Cambridge, UK, and New York, NY, USA, pp. 65-108
Center for Environmental and Geographic Information Systems (CEGIS), 2013, Report on
Water Availability Assessment using SWAT Model, www.riceclima.com [accessed 12/09/2013]
Deressa T., Hassan R.M. and Ringler, C., (2008) Measuring Ethiopian Farmers’ Vulnerability to
Climate Change Across Regional States, Environment and Production Technology Division, IFPRI
Discussion Paper, 00806

RiceClima – Deliverable 3.1
38
Devisscher, T. Bharwani, S., Tiani, A.M., Pavageau, C., Kwack, N.E., Taylor R. (2012), Baseline
Assessment of Current Vulnerability and Adaptive Capacity, COBAM Project – 2011,
http://www.cifor.org/cobam/home.html, [accessed 12/11/2013]
Eakin, H. and Luers, A.M. (2006) Assessing the Vulnerability of Social-Environmental Systems,
Annu. Rev. Environ. Resour. 2006. 31:365–94, doi: 10.1146/annurev.energy.30.050504.144352
Eakin, H., and B. Tapia (2008) Insights into the composition of household vulnerability from
multicriteria decision analysis. Global Environmental Change 18(1): 112–127.
Everitt B S and Hothorn T. (2011) An introduction to applied multivariate analysis with R.,
Springer
Fellmann, T., (2012) The assessment of climate change related vulnerability in the agricultural
sector: Reviewing conceptual frameworks in FAO/OECD Workshop on Building Resilience for
Adaptation to Climate Change in the Agriculturel Sector, 23-24 April 2012, RedRoom, FAO
Fisher, S. (2013) Bangladesh: from adaptation to low carbon resilience?,
http://www.iied.org/bangladesh-adaptation-low-carbon-resilience [accessed 07/01/2014]
Fussel, H.S. (2007) Vulnerability: A generally applicable conceptual framework for climate
change research, Global Environmental Change, Volume 17, Issue 2, Pages 155–167
Gallopin, G.C., 2006. Linkages among vulnerability, resilience, and adaptive capacity. Global
Environmental Change 16, 293–303.
Intergovernmental Panel on Climate Change (IPCC) (2001) Annex B. Glossary of Terms,
http://www.ipcc.ch/pdf/glossary/tar-ipcc-terms-en.pdf [accessed 07/02/2013]
Kaly, U., and C. Pratt (2000) Environmental vulnerability index: Development and provisional
indices and profiles for Fiji, Samoa, Tuvalu and Vanuatu. Phase II report for NZODA. SOPAC
Technical Report 306.
Nelson, R., Kokic, P., Crimp S., Meinke, H., Howden, S.M.,(2010a) The vulnerability of Australian
rural communities to climate variability and change: Part I—Conceptualising and measuring
vulnerability, Journal of Environmental Science and Policy 13 (210) 8-17
Nelson, R., Kokic, P., Crimp S., Meinke, H., Howden, S.M.,(2010a) The vulnerability of Australian
rural communities to climate variability and change: Part II—Integrating impacts with
adaptive capacity, Journal of Environmental Science and Policy 13 (210) 18-27
Nienke, M. Beintema and Wais Kabir (2006) Bangladesh, Agricultural Science and Technology
Indicators, ASTI Country Brief No. 34, July 2006

RiceClima – Deliverable 3.1
39
Nguyen, N.V. (2006), Global climate changes and rice food security, International Rice
Commission, FAO, Rome,Italy
O’Brien, K., Eriksen,S., Schjolden, A., Nygaard,L.,(2004) What’s in a word? Conflicting
interpretations of vulnerability in climate change research, CICERO Working Paper 2004:04,
P.B. 1129 Blindern, 0318 Oslo
Piya L., Maharjan, K. L. and N.P. Joshi (2012) Vulnerability of rural households to climate
change and extremes: Analysis of Chepang households in the Mid-Hills of Nepal, Selected Paper
prepared for presentation at the International Association of Agricultural Economists (IAAE)
Triennial Conference, Foz do Iguaçu, Brazil, 18-24 August, 2012.
Ramamasy, S. and Bass S. (2007) Climate variability and change: adaptation to drought in
Bangladesh, A resource book and training guide, Food and Agricultural Organization of the
United Nations, Rome, ISBN 978-92-5-105782-7
Smit, B. and Wandel, J., (2006) Adaptation, adaptive capacity and vulnerability, Global
Environmental Change 16 (2006) 282–292
Vogel, C.H. and Karen O'Brien, 2004. Vulnerability and global environmental change: Rhetoric
and reality. AVISO, 13 (March)
Winston, Y. , Mozaharul, A., Ahmadul, H., Abu Saleh, K., Ruane, A., Rosenzweig, C., Major,
D., Thurlow, J. (2010) Climate Change Risks and Food Security in Bangladesh, ISBN
9781849711302, 176pp. EarthScan, London
World Bank (2010) The Economics of Adaptation to Climate Change : Synthesis Report,
World Bank, Washington DC 20433, United States

RiceClima – Deliverable 3.1
40
Annex 1. Excluded indicators in vulnerability assessment
Code Indicator Unit Reason for exclusion
1
Crop variety replacement within the last
15 years
Nos. No differentiation
between upazillas
2 Irrigation coverage under study sites %
3
Device-wise irrigation cost
Tk
Poor information
4
Crop-wise irrigation number, times and
depth of water applied for crop
production
Nos.
5
Rice disease incidences
No differentiation
between upazillas
6
Rice insect incidences
7
Perceptions on climate changes and its
impacts on agriculture
8
Problems encountered in agriculture due
to climate change
9 Mean Precipitation for Oct.-Nov.
10
Climatic and environmental variation in
last 15 years
11 Environmental concerns in the surveyed
area
12 Effects of major agricultural and societal
issues on livelihood in surveyed area
13
Mean Temperature for Dec.-Jan-Feb.
( C )
High Correlation
14 Mean Temperature for March-Apr-May
15
Mean Temperature for June-July-Aug-Sep.
16
Mean Temperature for Oct.-Nov.
17
Mean Precipitation for Dec.-Jan-Feb.
(mm)
18
Mean Precipitation for March-Apr-May
19
Mean Precipitation for June-July-Aug- Sep.
20
Mean Precipitation for October – Non.

RiceClima – Deliverable 3.1
41
Annex 2. Factor Analysis in PCA
Table 1. Factor Analysis for Adaptive Capacity Indicators
Factor
Factor
Age
-0.60673
-0.493516
Schooling years 0.82118 0.499806
Farm Exp.
-0.57051
-0.170434
FamilySize -0.86289 0.173527
Own Farm
0.44348
0.876551
Farm Size -0.08718 -0.761755
Crop Intens. 0.93086 -0.187183
BCR All -0.35366 -0.466762
Livestock Score
-0.68483
0.565911
Inc. Av. -0.02209 -0.901198
Access Tubewell_Lartrine
0.96757
-0.225128
Access Pacca_Electricity 0.96757 -0.225128
Infr.post
0.35122
0.101852
infr.health -0.76437 -0.293278
infr.vet
0.89280
-0.423456
Infr.coop -0.91303 0.027316
Infr.agr.ext.
0.80956
-0.066638
Infr.finan -0.07238 -0.943837
infr.school
0.59102
-0.208836
infr.college -0.30124 -0.823146
in-migrat
-0.81679
-0.174353
Local m. 0.50491 -0.429069
bigger m.
0.47677
0.142531
hosp. Km 0.27963 -0.459243
town km
0.17976
0.315068
Expl.Var 11.28531 5.876386
Prp.Totl
0.43405
0.226015
Note: The statistically significant variables are presented with red font color while the variables attaining highest
values are framed with greyish shade.

RiceClima – Deliverable 3.1
42
Table 2. Factor Analysis for Sensitivity and Exposure Indicators
Factor
Factor
Crop Pattern 2
-0.88827
-0.348879
Crop Pattern 3 0.68340 0.290611
Small Farm Size -0.43364 0.780922
Tenant Farmer 0.34344 -0.638966
HYV Boro yield -0.58322 -0.692499
T_annual 0.98831 0.135129
P_annual
0.98271
0.178196
Yloss_Aus -0.99655 -0.079465
Yloss_aman
-0.72608
0.569433
YL_slight_aus 0.99064 0.100983
YL_mod_aus
-0.42021
0.390647
YL_severe_aus -0.98791 -0.147803
YL_slight_aman
0.91794
-0.237036
YL_mod_aman 0.76440 -0.000306
YL_severe_aman -0.96482 0.198452
NIR_Aus -0.99517 -0.084552
NIR_Aman -0.91925 0.384395
NIR_Boro -0.92324 -0.290180
Expl.Var
12.54443
2.603757
Prp.Totl 0.69691 0.144653
Note: The statistically significant variables are presented with red font color while the variables attaining highest
values are framed with greyish shade.

RiceClima – Deliverable 3.1
43
Annex 3. Standardizes Values
Table 1. Adaptive Capacity
Age
Schooling
years
Farm
Exp.
FamilySize
Own Farm
Farm
Size
Crop
Intens.
BCR All
Livestock
Score
Inc.
Av.
Acc.Tub.
_Latr.
Acc.B.
H._Elect
.
Access
health
0.81
-0.42
-0.11
-0.18
-1.00
1.00
1.00
0.03
-1.00
1.00
1.00
1.00
-1.00
-1.00
1.00
-1.00
-0.50
1.00
-0.29
0.48
-0.06
-0.92
-0.25
1.00
1.00
-1.00
0.14
0.51
0.92
-1.00
0.36
-0.67
0.45
0.17
-0.25
-1.00
1.00
1.00
-1.00
1.00
-0.87
0.89
0.81
-0.27
0.71
-0.92
1.00
0.22
-0.25
-1.00
-1.00
1.00
0.71
-1.00
1.00
0.46
-0.99
0.00
-1.00
0.81
0.02
0.25
-1.00
-1.00
1.00
0.34
-0.06
0.11
1.00
0.50
-1.00
-0.22
-1.00
1.00
-1.00
-1.00
-1.00
1.00
Infr.post
infr.health
infr.vet
Infr.coop
Infr.agr.ext.
Infr.finan
infr.school
infr.college
in-migrat
Local
m.
bigger
m.
hosp.
Km
town
km
1.00
-0.11
1.00
-1.00
-0.17
1.00
1.00
0.27
0.18
1.00
-1.00
0.14
0.33
0.63
0.06
0.09
-1.00
-0.33
-1.00
0.18
-1.00
-1.00
-0.60
-1.00
-0.71
-0.20
-0.19
-1.00
0.62
-0.83
1.00
-0.77
0.01
-0.66
-0.09
-0.20
1.00
1.00
0.47
0.90
0.40
-1.00
-0.20
-1.00
-1.00
0.32
-1.00
1.00
-1.00
-1.00
-1.00
1.00
-1.00
1.00
-1.00
1.00
-1.00
0.58
-1.00
1.00
0.62
-0.80
-1.00
0.43
-1.00
0.48
-0.14
-1.00
0.25
-1.00
-1.00
-0.11
-1.00
0.50
0.00
-1.00
-0.71
0.47

RiceClima – Deliverable 3.1
44
Table 2. Exposure – Sensitivity
Crop Pattern
2
Crop
Pattern 3
Small Farm
Size
Tenant Farmer
HYV Boro
yield
T_annual P_annual Yloss_Aus Yloss_aman
0.90
-0.41
-0.89
1.00
0.97
-1.00
-0.99
0.74
-0.43
0.75
-1.00
0.25
-1.00
0.85
-1.00
-1.00
0.81
0.14
1.00
-0.35
1.00
-0.36
1.00
-1.00
-0.99
1.00
1.00
0.00
-0.41
-1.00
0.27
0.56
1.00
0.85
-1.00
-1.00
-1.00
1.00
-0.13
0.99
0.58
1.00
0.85
-0.94
-0.43
-1.00
0.47
0.25
-0.50
-1.00
1.00
1.00
-0.94
-0.14
YL_slight_aus
YL_mod_aus
YL_severe_aus
YL_slight_aman
YL_mod_aman
YL_severe_aman
NIR_Aus
NIR_Aman
NIR_Boro
-1.00
1.00
0.88
-1.00
0.00
0.71
0.71
0.17
1.00
-1.00
0.00
1.00
-1.00
0.00
0.71
0.78
0.52
0.80
-1.00
0.00
1.00
-1.00
-1.00
1.00
1.00
1.00
0.39
1.00
-1.00
-0.88
1.00
1.00
-1.00
-1.00
-1.00
-0.91
1.00
-1.00
-0.88
1.00
0.00
-0.71
-0.87
-0.68
-0.67
0.88
1.00
-1.00
-0.20
1.00
-0.14
-0.96
-0.11
-1.00

RiceClima – Deliverable 3.1
45