Bangladesh ranked seventh among the ten countries in the world most vulnerable to climate-induced natural catastrophes, affected from 1999 to 2018 (Global Climate Risk Index, 2020). The GCRI (Global Climate Risk Index) also represents total losses of the 1686.33 million US dollars and the number of events is 191. A tropical cyclone named Amphan affected Bangladesh on 20 May 2020. As of 22 May, the UN estimates that 10 million people in 19 districts were affected, with an official death toll of 17 (OCHA, 2020). Bangladesh has a dominant and the harmful patriarchal societal structure with norms and practices that place females of any age at higher risks of violence (UN Women, 2020). Men in Bangladesh tend to control income distribution, property, and access to credit, decision-making processes, and sources of food. Even in normal times, women lack social power and their condition is being aggravated by the increased demands on them (IDS, 2008). Poverty and womens vulnerability are structural issues requiring the transformation of practices perpetuating them (UNISDR, 2013). Gender inequalities manifest themselves in differing roles, resources, rights, knowledge, & time to adapt to climate change (Babagura, 2010; Petrie, 2010). 

Around two-thirds of the total land of Bangladesh is classified as wetlands (Nanda et al., 2016). Which has been projected around seven to eight million hectares (Rahman et al., 2001). Sunamganj the wetland area of Bangladesh affected by climate change adversely, especially because of its unique geographic location (Rahman et al., 2016). The regions often experience disruptive flash floods (Abedin et al., 2020). Flash floods arise for a short duration with slight or some-times no warnings. Haors unique hydroecological characteristics are shaped in Sunamganj, Sylhet, Moulvibazar, Kishoreganj, Habiganj, Brahmanbaria, and Netrokona Districts (CEGIS, 2012). In the haor area, the socioeconomic and livelihood conditions are much different from the plain and hilly regions (Hossain et al., 2017). The flooding of wetlands negatively affects the economic condition of farmers who are primarily dependent on farming practices (Rahman, 2014). 

Widespread problems the region experiences of food insecurity due to a combination of factors which includes: crop losses due to early or flash floods and erosion; poor access to markets, especially during the flood season; and traders and services isolation (Caldwell, 2011). The wetland region has overall notably low literacy rates among the total population, compared to the national average. Moreover, the region has a high incidence of water-borne diseases and very poor maternal and child nutrition and health indicators (BBS, 2010). Under the prevailing social and economic circumstances, women are lagging far behind their male counterparts. Because of gender-based inequalities, all classes of women are typically at higher risk than men (UN, 2004; Enarson, 2002). Due to separation, divorce, or widowhood, women have less access to rights & resources (Antonopoulos and Floro, 2005). The lack of utility services in informal settlements disproportionately affects poor women, exacerbating inequities based on legal status, race, and other intersecting social circumstances and increasing their vulnerability to climate shocks (Jagriti, 2015; Khosla, 2010). 

To cope with seasonal and episodic weather & natural disasters women have fewer resources. Consequently, traditional roles are reinforced, and womens ability to diversify their livelihoods is diminished (Masika, 2002; Denton and Parikh, 2003). Increasing the number of women in leadership positions would elevate the quality of their contribution to the economy and help Bangladesh achieve middle-income status (Marcus Noland, 2016). There are several researches in the haor (wetland) areas of Bangladesh but very few researches are based on the differentiated impacts analysis. The study area of Sunamganj District is unexplored in terms of identifying the differential impacts of climate change on women.

Site selection for the research

Sunamganj District is located in the north-central region of Bangladesh below in (Fig. 1). The area is 3,669.58 sq. km, located between 24°34 and 25°12 north latitudes & between 90°56 & 91°49 east longitudes. Sunamganj District constitutes the bowl-shaped depression basin of the country, locally called haor. There are 95 haors located in this area. Sunam-ganj District comprises of 11 Upazilas (BBS, 2011). 15 villages of 6 unions under Dowarabazar and Jamalganj Upazilas of Sumanganj District were selected for the designed research. The area of Dowarabazar Upazila is 324.19 sq. km. There are 294 villages, and 9 unions in Dowarabazar Upazila (BBS, 2011). The area of Jamalganj Upazila is 338.74 sq. km. and comprises 5 unions and 165 villages (BBS, 2011). The household survey location map is shown in below Fig. 2.

Research approach and method

Community Disaster Resilience Index (CDRI) was used to measure differentiated climate change impact on women and to enhance social equity in the study area.  

Fig. 1: Study area map of Sunamganj District.

Fig. 2: Household location map in the study area.

The surveys were conducted at individual house-holds of all types of women and men middle-aged, young girls, elderlies, disabled, ethnic, religious minorities, widowed, destitute women, and all income classes of the pro-poor, poor, middle class, or even rich. The Community Disaster Resilience Index (CDRI) method is considered for quantitative and qualitative analysis that consists of five dimensions or capitals. These are directly or indirectly affected by differentiated climate change vulnerabilities in women. The five dimensions are: human, physical, social, economic, and natural. Each of these parameters is highlighted in Table 1.

Table 1: Components and sub-components of the CDRI framework that have been adopted and contextualized in the study.

CDRI calculation

The Community Disaster Resilience Index (CDRI) method encompasses of five dimensions which are equally divided into five parameters. Considering the weighting scheme with five variables their ranked is W1, W2, …. W5 based on their importance. In this ranking “0” indicates no importance, lower importance “1” and most importance “5” in shaping the highest score of a particular parameter and impact dimension. Because this simple structured questionnaire with uniform numbers for each variable and parameter ranging between 1 and 5, it allows a transparent adop-tion of the formula (Eq. 1) named weighted mean to calculate the CDRI scores for each variable, parameter, & dimension in a uniform and synchronized approach (Joerin and Shaw, 2011). The following equation is proposed for integrating the five major domains of capital to generate the Community Disaster Resilience Index (CDRI) from the research perspectives.

Data collection

Both quantitative and qualitative data were collected through primary and secondary methods for this study.

Primary data collection

The accuracy of information depends greatly on the survey methods used. An inspection survey is conduc-ted to understand the areas environment, social capital, and peoples capacity development perception.

Household questionnaire survey

To gather the quantitative data, individual face-to-face interviews were conducted. A semi-structured ques-tionnaire is used to collect data from the household survey. Required information for research is collected from the area through this questionnaire survey. A total of 300 household surveys were directed to fulfill the research purposes. The survey involved 150 male and 150 female interviews.

Focus group discussion (FGD)

Focus group discussions were conducted with both males and females to find out female real conditions. It helps to gather a wide range of information in a short time. The participants of the FGD were 8 to 10 women engaging in different occupations and different ages. Based on the objectives of the research, 10 Focus group discussions (FGD) of both females (5 No.) and males (5 No.) were conducted in both Upazilas.

Key informant interviews (KII)

KII interviews with 10 key informants were performed for this research. The survey engaged the respective Upazilas government, and NGO offices, the Chairman of the union parishad, union Parishad Members of each union, teachers, and other experienced society in the respective area for the interview.

Secondary data collection

To the establish a foundation for the study, various secondary sources of data have been utilized, including study books, journals, articles, newspapers, abstracts, internet documents, government office records, NGO office reports, union and the Upazila Parishad office records, and published reports.

Data analysis

For the estimation of indicators by the respondents multiple and single perceptions considered as high, medium, low, very low, and no values are made for each criterion. These categories are weighted accor-ding to the responses simply by assigning most severe or out of control = 5, very high = 4, high = 3, medium = 2, and low = 1 point and weighing the totals by multiplying with the respondents value and then dividing it by value of sampling shown in annex-1. Thus, a total value for the particular differentiated impacts is derived. All data have been analyzed using ArcGIS 10.7 (Study area and household survey map), SPSS, & Microsoft Excel (different analytical charts).

Socio economic status

During the questionnaire interview, almost half of the respondents opined they were poor, and the value is about 51.21% (Fig. 3). Poor have fewer opportunities for income generation during any disastrous events. A poor household especially a women-headed household severely affected due to lack of earnings because of social & cultural barriers of conservatism, early marri-age, lack of awareness, less technical support from outside assistance, etc. A large group of people (about 38.75%) is in middle-income categories and middle-income families face the most unexposed problems due to floods, flash floods, and other natural stresses.

The intermediate-level population cant receive any kind of help from others because of social crises. Even they dont share any crisis with others. However rich people in the study area are less affected by climate anxieties. Only 10.03% shared they were affected but most of the cases can survive with their own resources.

Fig. 3: Respondents socioeconomic status of the household.

Income generation

Fig. 4 indicates the climate change impact on eco-nomic sectors is very high in the study area. The monthly income is very low which is represented in Fig. 4. It is evaluated that income from 3000 to 10,000 taka per month is poor, from 10,000 to 30,000 are middle class, and more than 30,000 Bangladeshi takas are considered rich. About 62.56% of respondents have no contribution to income generation. Especially during calamities, there are no opportunities for the women to income generation and they are completely dependent on others. Besides, a large group earning money per month from less than 3000 to a maximum of 15,000 taka is not enough to support the family. 

Fig. 4: Income generation per month.

Only a few people are in the middle and rich class and earn money from more than 10,000 to 50,000 taka per month. Although in most cases, poor and middle-class people fight with climate vulnerabilities rich people somewhat can cope with their resources.

Fig. 5: Last 20 years climate change scenarios at Sunamganj.

Climate change at Sunamganj in the last 20 years

In the wetland areas, the climate drastically changes & about 93% of respondents agreed with this statement below Fig. 5. The Haor areas geographical location is an important reason for climate change impacts. Every year different natural events of thunderstorms, flash floods, & floods hurt food production, fish farming, income generation, health, education, etc. Only 2% of respondents disagreed & they think climate change isnt an abnormal issue. Its the rule of nature. 0.667% didnt know and they didnt notice the change. How-ever, 4.33% of respondents verbally experienced & agreed that climate change is noticeable and the impact on women is very high compared to men.

Changes in pattern for the last 20 years

From the analysis (Fig. 6), 93% agreed that climate change  has  been  happening  for  over 20 years. In the last 20 years temperature fluctuation, the river bank erosion, drought, flood, flash flood, cold wave, heavy rainfall, hailstorm, thunderstorm, seasonal variation, and windstorms have increased noticeably in the study area. Raising temperatures, drought in the summer season, seasonal floods, flash floods, heavy rainfall, hailstorms, thunderstorms, and seasonal variation rela-tively more increased than others. Increasing phenol-mena of climate cause large stresses on the human being. Women are particularly immensely affected by the aforementioned catastrophes. At present, thunder-storms, floods, and flash floods in the study area have created an alarming situation for livelihood. Recent statistics indicate that Sunamganj wetland area is more vulnerable to thunderstorms. 

Fig. 6: Climatic events at Sunamganj in the past 20 years.

Dimensional analysis

Human capital

Human capital (Fig. 7a) is subdivided into four indi-cators education, health, skill, and agency. All the indicators are affected by climate change in the haor region. Among these, the most affected is womens agency and the score is 3.48. Women have barriers to deciding on a disaster and it is found high rate from other human indicators. Political power, patriarchal norms, gender roles, lack of knowledge, and religious conservatism create barriers to making the decision. Womens livelihood alternative skills are impacted by the climate stresses and the calculated value is 3.41. Due to a lack of the awareness, money, and other resources, absence from the empowering environment is the root cause of achieving skills. Thirdly, girls education was hampered due to the changes with a value of 3.12. Because they face difficulty in attending school and inability to pay for education. The above graph specifies that there is a lower impact on womens health than the other indicators and the value is 2.29. Waterborne diseases, emotional stresses, menstrual problems, and skin diseases are common health issues for women.

Impacts on womens social capital

Social capital (Fig. 7b) is hampered due to climate change. Trust, networking systems, and social norms are all of the indicators greatly impacted due to climate stresses. The networking system of women is highly affected and the impact score is 4.09. Women face various problems in joining basic service programs. The family didnt permit women to engage in institu-tions to build a network. Sometimes women are not interested although they have enough time. In most cases, networking programs are arranged by male people, so womens decisions are not given much priority, & subjected to physical & mental harassment when attending such sessions. Besides, society has a lack of trust in women during the disaster risk reduction processes. Social trust is the second impacted area according to the calculation and the score is 3.81. Masculinity, lack of awareness, less attention to womens needs, physical states of women, etc. are the sources of the lack of social trust. During the distribution of relief and resources, women faced more difficulties than men when disasters struck. Also, the social norms and values of women are affected by disaster and the score is about 2.74. Climatic effects are responsible for to rise in domestic violence, sexual assault, and social insecurity.

Impacts on womens economic capital

Climate change hurts womens income, savings, and investments. From (Fig. 7c), the most impacted area is savings and the impact score is 3.11. Religious conser-vatism, lack of the savings knowledge, less access to income generation, disapproval from family, and lack of savings institutions developed womens savings-related problems. Secondly, it reduced womens in-come generation and the value is 2.86. The impact on livelihood & income generation due to climate change causes migration and challenges to the migrated people. The challenges change the nature of womens income characteristics. Women also faced different kinds of barriers in the job sectors. Patriarchal norms, conservatism, & illiteracy make women more vulner-able to taking and repaying loans. Women are discour-aged from participating in economic activities due to the fear of falling into a cycle of debt. Womens invest-ment is hindered by poor income sources and social barriers, resulting in a score of 1.99 for economic impact.

Climate change impacts on women physical capital

Community physical resources (houses, businesses, and facilities) are significantly hampered due to hazards (Fig. 7d). In the study area, housing structures affected by climate change resulted in a score of 2.13 and that is the lowest score from other physical structures. Community kaccha houses are frequently impacted by disasters. The house gets inundated, washed away, and sometimes can be tilted. Due to this reason, women cant actively join other services to build their capacity. The community business capacity impacted score is 2.55. People actively linked with different primary and secondary businesses. Climate change directly or indirectly hampered their business. Most of the respondents directly or indirectly con-nected with agri-business. Lack of agricultural inputs and agricultural storage hampered agricultural-related business. Besides, Women face challenges in accessing agricultural and non-agricultural assets for business. Lack of skill, lack of the capital to afford the assets, conservatism, and lack of technical support are the principal barriers to women doing business. Lack of facilities women cant continue their daily activities. Most of the impact is on the facilities of women (score is 2.8). WASH facilities were seriously affected due to the disaster. Toilets were inundated and washed away in times of frequent events. Sometimes toilet materials are disrupted during the flood. Women face difficulties collecting water from far distances and most of the time they use polluted water. Bathing, cooking, and cleaning houses and clothes are extremely hampered due to climatic events.

Fig. 7: (a) Climate change impacts on womens human capital; (b) impacts on social capital; (c) on economic capital; (d) on physical capital and (e) impacts on womens natural capital.

Climate change impacts on women natural capital

The natural resources in the wetland zones have been severely impacted by climate change. Fig. 7e repre-sents women harshly affected by managing forest resources (Impact score is 3.06). Lack of firewood source and storage capacity and gender roles make them more vulnerable. Generally, women collect water for both domestic use and drinking purposes and water resources are the next impacted natural capital, and the impact score is 2.88. Due to the climatic impacts, water sources are the inundated and sometimes washed away. Women feel physical and mental difficulties in collecting water from long-distance muddy roads and queue service to serve their families. Women also faced pregnancy-related complications, diarrhea, cho-lera, and other waterborne diseases. Local people use different purifying methods to drink water as filtration, chlorination, and heating method. Using ecosystem services related to women in the area largely impacted by climatic events. The calculated impact score of ecosystems is 2.83. The distribution of the natural resources is largely influenced by political biases. In the haor areas, improper monitoring of the government body affects to access natural resources for women. Fisher-women are not freely accessed in the whole-haor areas for easy livelihood and their income is reduced due to the unfair behavior of the local leaders. Besides, the lack of management and monitoring of natural resources impacted the womens natural capital highly.

Overall climate change impacts on women

To fulfill the research objectives the CDRI method is used to analyze quantitative impact values through the household data calculation. CDRI method is divided into five capitals human, social, economic, physical, and natural (Fig. 8). Women face social discrimination and the condition is being aggravated by the increased differentiated climate change impacts on them. Social capital indicates that women in the study area are mostly affected due to social barriers and the impacted score is 3.54 which is the highest impact score in this research. From the analyses, climate change impacts the social sector within the weight between 3 to 4 and is categorized as more than high but less than a very high impact. Social trust in women, networks, and norms are greatly affected by climate change. Women are characteristically at higher risk than men in receiving education, health care, and nutrition intake. Study reveals womens human capital is affected as the second highest score is 3.07. Education, health, skill, & agency or decision-making are severely interrupted due to climate stresses. Although women can access it sometimes but cannot control natural resources and other property rights (Ribeiro et al., 2010; Shahjalal M., 2021). The third impacted area in this study on women is natural capital & the score is 2.92 considered as medium type impacts. The natural capital includes water resources, forest resources, and the ecosystem services, & women are hindered moderately due to climate change. Wetland flood reduces the farmers economic condition, mainly dependent on agriculture & other secondary occupations. Economic capital has the fourth highest score of 2.62 and causes medium-level climate change impacts on womens livelihood than males. Economic wealth is related to income, saving, and investment and women are not usually familiar in most cases. Climate-induced flash flood in wetland areas affects settlement, agricultural produc-tion, road, & other communications every year (Islam et al., 2010). The moderate and fifth-rank impact is the physical capital of women and the impact score is 2.49. In the research area, women frequently face housing problems, agricultural and non-agricultural business barriers, and also change to be disrupted in water, sanitation, and hygiene services.

Fig. 8: Dimensional analysis according to the CDRI model.

Women in hoar areas (wetlands) in Bangladesh are highly susceptible to the influences of climate change. Usually, people living in haor areas are more vulner-able to floods, flash floods, riverbank erosion, heavy rainfall, temperature variation, and thunder-storms. Women face social discrimination and the condition is being aggravated by the increased differentiated cli-mate change impacts on them. The frequent ire-gular natural events have severe impacts, especially on the womens income, livelihood, education, health, infra-structure, assets, and skills. Besides, different climates caused health complications, the vulnerable sanitation, poor house structure, illiteracy of women, the water crisis in the dry season, and water source management in the rainy season, directly and indirectly, affected womens lives & livelihood. The analyzed Community Disaster Resilience Index (CDRI) shows the climatic impact score of social capital (3.54), human capital (3.07), natural capital (2.92), economic capital (2.62), and physical capital (2.49). People exist in poverty and to migrate other places due to the loss of income and property. More than 85% are involved directly or indirectly in agricultural work. Paying female roles in agricultural work is very difficult. Respondents of the study area have minor access to savings from their daily earnings. Sometimes they managed the economic losses that arose from damage and inundation of the cultivated lands by taking a loan from microcredit organizations, other banks, and societies while women have fewer opportunities for such recovery. The avail-ability of the financial resources largely constrained recovery from disasters. Migrating to other places often poses an additional problem for the improvised people, especially in securing livelihood options due to their lack of technical and communication skills. They need motivation, counseling, training, & all institu-tional provisions from the government and the NGOs for their sustainable livelihood. Investing in basic social services and infrastructure, particularly in health care, water, sanitation, childcare, and the labor-saving technologies reduces womens workloads and builds resilience without further limiting their time and self-determination. Intersecting inequalities of the women increases womens profits, resources, opportunities, and adaptive dimensions to reduce the differentiated climate change vulnerabilities.

This research was based on a Master of Science thesis by the first author. The authors would like to express their gratitude to the Department of Disaster Resilience and Engineering, the faculty of the Postgraduate Studies, PSTU for the facilitating this research. The authors would also like to express their gratitude to the International Centre for Climate Change & Develop-ment (ICCCAD), Dhaka 1212, Bangladesh, & Bangla-desh Center for Advanced Studies (BCAS) for the MS grant & technical support respectively. Acknowledge-ment also goes to the anonymous reviewers for their careful review and valuable suggestions, which helped us to improve the manuscript.

The authors declare that they have no known compe-ting financial interests or personal relationships that could have influenced the work reported in this paper