M. A. Rakib
Graduate Program in Sustainability Science - Global Leadership Initiative, Graduate School of Frontier Sciences, The University of Tokyo, 5-1-5 Kashiwanoha, Kashiwa, Chiba, 277-8563, Japan and Department of Disaster Management, Begum Rokeya University, Rangpur, 5400, Bangladesh
Jun Sasaki
Department of Socio-Cultural Environmental Studies, Graduate School of Frontier Sciences, The University of Tokyo, 5-1-5 Kashiwanoha, Kashiwa, Chiba, 277-8563, Japan
Sosimohan Pal
Department of Disaster Management, Begum Rokeya University, Rangpur, 5400, Bangladesh
Md. Asif Newaz
Environmental Science Discipline, Khulna University, Khulna, 9208, Bangladesh
Md. Bodrud-Doza
Department of Environmental Sciences, Jahangirnagar University, Savar, Dhaka, 1342, Bangladesh
Mohammad A. H. Bhuiyan
Department of Environmental Sciences, Jahangirnagar University, Savar, Dhaka, 1342, Bangladesh
Coastal hazards, Livelihood crisis, Social vulnerability, Traditional knowledge uncertainty, Multivariate statistic
The Gabura union in the Shymnagar Upazila and Satkhira districts located in the southwestern coastal region of Bangladesh
Risk Management in Agriculture
Climate change
Study area: This study was conducted at the Gabura union in the Shymnagar Upazila and Satkhira districts located in the southwestern coastal region of Bangladesh. It is a site that is highly vulnerable to coastal hazards like cyclonic storm surges and salinity problems, which mostly originate from the southern part of the Bay of Bengal. In Bangladesh, cyclones (e.g., Aila-2009, Nargis-2008 and Sidr-2007) have caused significant damage among the coastal communities, such as loss of life, property (Dube, 2009), and livestock. A large number of people are involved in agriculture, aquaculture, fishing, and mangrove forestrelated livelihood activities. Major portion of the agricultural land has been converted into aquaculture farming. The path of two of the major rivers, Kholpetua and Kobadak, encloses different sites of the union. Geographically, this area is considered to be a highly vulnerable and disaster-prone location, very close to mangrove forests and the Bay of Bengal. Data collection: Methods of data collection consisted of a questionnaire survey, interviews and field observations. These techniques were intensively used to investigate the real scenarios, and conceptualize data collection in the field. A semi-structured questionnaire was formulated using open and closed-ended questions to explore the situational crisis, social crisis, and traditional strategies for disaster risk reduction. The questionnaire was segmented into four sections: (1) general information about the local people, (2) coastal threats and impacts, (3) livelihood vulnerabilities depending on regional natural events and socio- and agro-environmental factors, and (4) traditional techniques for disaster risk reduction and perspectives on disaster management challenges. Likert scale (five scales) type questions were used to code the questionnaire data. A total of 149 questionnaires were collected among respondents (both male and female) who agreed to share their views. All questions were read to respondents in the local language. All information was collected using feedback forms, and some additional information was recorded by hand and later formatted in a data collection sheet. A total number of 32 principal variables were selected. The variables reflected social issues, regional environmental changes, impacts, severe salinity hazards, and local initiatives to take the measures under adverse climatic conditions. These variables were used to evaluate the social vulnerability in terms of the changing status with time using multivariate statistics. Almost all the variables were selected from discussions with the local people. Interviews were conducted with local businessmen, floating farmers, fishermen, honey collectors, day laborers, travelers, and some other unspecified people. Field observation was performed to understand the social problems in relation to research questions. These findings were helpful to verify the questionnaire results. The entire approach to data collection was implemented rigorously to maintain the importance of the study's findings and to fulfill the research goals. The database was organized according to the analysis and interpretative design while similar characteristics of the data were kept together so as to ensure the existing relationships among the variables were apparent. In addition, to conceptualize the relationship between social vulnerability, development crisis, and local knowledge uncertainty, each of the factors corresponding to the three components was gathered through interviews and field observations. Multivariate analysis: Multivariate analysis, including principal component analysis (PCA) and cluster analysis (CA), were performed to examine the relationships, causes, and impacts of socio- and agro-environmental factors. PCA plays a significant role in identifying the associations between two variables by reducing database dimensionality (Helena et al., 2000). The associated variance among the variables expressed by the PCA eigenvalues and the total number of original variables' inclusion in the PCA are specified by the loadings, while the loading scores refer to the transformation of individual observations (Helena et al., 2000). CA identifies the similarities in regard to variable characteristics and classifications into several groups (Bhuiyan et al., 2010). Each of the groups contained a number of variables that exhibited strong internal homogeneity while two groups demonstrated high external heterogeneity (Shrestha and Kazama, 2007). A hierarchical agglomerative approach is the most common technique for expressing the relationship between variables or an entire set while Euclidean distance usually represents the distance between samples, typically explained by a dendrogram or tree diagram (McKenna, 2003; Otto, 1998).
Journal of Environmental Management, Volume 231, 1 February 2019, Pages 419-428
Journal