Vulnerability and livelihood outcomes in the context of salinity intrusion in the Mekong Delta

cts, it reduces crop productivity, household income, and spending. Thirdly, regarding the effects of adaptive capacity on livelihood outcomes, they were proven by many scholars. Scholars often analyzed adaptive capacity in terms of capital resources (natural, human, social, financial, and physical resources ) and they found that all these resources play a positive role in improving livelihood outcomes of households. Fourthly, in terms of the regulatory role of adaptive capacity, although previous studies confirmed the important role of adaptive capacity in reducing negative impacts of LVI component, they mainly approached from a macro perspective and did not thoroughly consider its regulatory role, only analyzed small components in adaptive capacity in the context of vulnerability in general and salinity intrusion in particular. Therefore, it is essential to carry out studies on the regulatory role of adaptive capacity in the relationship of salinity intrusion and livelihood outcomes. 1.5 Research gaps Therefore, there are still some notable research gaps in analyzing the vulnerability and livelihood outcomes, specifically: (1) Most studies applying LVI used weighted balance, but this method was criticized as same weights are applied to different components of LVI. Therefore, it is necessary to develop a more suitable vulnerability calculation method to produce more convincing results. (2) Regarding livelihood vulnerability, domestic and international research often focused on climate change scenarios (temperature change, precipitation)to calculate LVI to indicate which regions and who are affected but without thoroughly studied the context of salinity intrusion in the Mekong Delta - where is confirmed as one of those regions in the 7 world being the most vulnerable to climate change, especially salinity intrusion. (3) Most studied only measured LVI without analyzing the overall role of LVI components in livelihood outcomes. (4) Previous studies only explored the regulatory role of several small components in adaptive capacity, and did not analyze specifically the regulatory role of adaptive capacity in the context of salinity intrusion. Therefore, the main objective of the dissertation is to assess livelihood vulnerability in the context of salinity intrusion in the Mekong Delta with the application of unequal weight proposed by Iyengar and Sudarshan (1982), to concurrently evaluate the impacts of LVI components (exposure, sensitivity, adaptive capacity) on livelihood outcomes, to specifically indicate the regulatory role of adaptive capacity in reducing influence of salinity intrusion on livelihood outcomes in the Mekong Delta. With the above objectives, the author hopes to resolve all research gaps and offer some recommendations to minimize livelihood vulnerability in the context of salinity intrusion in the Mekong Delta. 8 CHAPTER 2 LITERATURE REVIEW 2.1 Livelihood and sustainable livelihood framework 2.1.1 Livelihood and sustainable livelihood The definition of livelihood that is frequently used and cited in later studies is based on the concept of livelihood by Chambers and Conway (1992), in which livelihood is, in the simplest sense, the means to make a living. In a more detailed description, Chambers and Conway defined that livelihood includes the capabilities, resources, and activities required to provide the daily life. Similarly, Scoones (1998) stated that livelihood includes necessary abilities, resources (including physical and social resources) and activities as a means of life for human beings. Taken into account the above definitions, DFID (2001) also argued that livelihood includes the capabilities, resources and activities necessary to serve as a means of life for human beings. In summary, livelihood is the use of resources necessary to carry out activities to achieve desired results (Vu Thi Hoai Thu, 2013). 2.1.2 Sustainable Livelihood Framework of DFID (2001) Sustainable livelihood framework of Department for International Development - DFID (2001) is a tool designed to analyze various factors that influence livelihood outcomes, especially those creating opportunities or posing challenges for livelihood(Twigg, 2001). Accordingly, all households have a means of making a living (livelihood) based on available livelihood resources (five types of resources) in the context of certain policies and institution in the local areas. These factors are also influenced by external factors such as storms, floods, droughts, salinity intrusion... and seasonal elements. The household's selection of livelihood based on current livelihood resources is the result of the interaction between these groups of factors. Figure 2.1: Sustainable Livelihood Framework of DFID (2001) 9 The purpose of the sustainable livelihood framework is to support policymakers and researchers from many fields in participating in the discussion of factors influencing livelihood outcomes and their importance and interaction. This facilitates the identification of suitable elements for livelihood support based on a clear analysis of each factor. Consequently, many scholars had confirmed that the sustainable livelihood framework of DFID (2001) is the most suitable method to analyze livelihood outcomes and disaster vulnerability (Twigg, 2001; Nguyen Duc Huu, 2016). This is also a theoretical framework that the author used to solve research questions in the dissertation. 2.2 Livelihood vulnerability in the context of salinity intrusion 2.2.1 Vulnerability The research overview has provided three broad viewpoints on vulnerability: biophysical, social, and integrated perspectives. It also confirms that the integrated viewpoint of vulnerability as a modern model in analyzing vulnerability in the context of climate change. Therefore, in this dissertation, vulnerability is approached from an integrated point of view, specifically the definition of vulnerability by Intergovernmental Panel on Climate Change (IPCC, 2001) was frequently cited (Ha Hai Duong, 2014; Abeje et al., 2019; Parry, 2007). Accordingly, vulnerability is the degree to which a natural or social system can be vulnerable or unable to cope with adverse effects due to extreme weather patterns and climate change). The IPCC clearly identified vulnerability as a function of three factors including exposure, sensitivity, and adaptive capacity. 2.2.2 Livelihood vulnerability As stated above, livelihood is considered sustainable if it can cope with and recover from harzards, maintain or enhance its abilities and assets, while not weakening its natural resources. Therefore, Scoones (1998) proposed that if livelihoods are unable to engage in dealing (temporary adjustments) or adapting (long-term change), it is considered vulnerable. Vulnerability is therefore used as an attribute of livelihoods and thus it emphasizes people and how they manage their lives (Murungweni et al., 2011). Hence, in the context of salinity intrusion, livelihood vulnerability refers to the degree to which communities/households can be vulnerable or unable to overcome the effects of salinity intrusion. It is also a function of three factors including exposure, sensitivity and adaptive capacity, specifically: Exposure is the presentation of individuals, households, communities, nations or ecosystems affected by an adverse environmental or socio-political factor characterized by the frequency, intensity, time and space of the event (Turner et al., 2003; Adger, 2006). Sensitivity is the degree to which a system is affected (either positively or negatively) by one or more factors from internal or external environment (IPCC, 2001). . Adaptive capacity (Turner et al., 2003) or reactive capacity (Gallopín, 2006) is the 10 potential to respond and reduce the vulnerability of a particular system. Adaptive capacity is highly dependent on the existence and access to resources such as natural resources, financial capacity, infrastructure, political institutions, human resources and social relationships (Brooks and Adger, 2004). 2.2.3 Measurement of livelihood vulnerability due to salinity intrusion Measurement of the exposure of salinity intrusion: As stated above, exposure is an expression of individuals, households, communities, nations or ecosystems affected by an adverse environmental or socio-political factor characterized by the frequency, intensity, time and space of the event (Turner et al., 2003; Adger, 2006;). Therefore, in the context of salinity intrusion and research of the household unit, the degree of exposure is the presentation of a household affected by salinity intrusion characterized by the frequency, intensity, time and sapce of salinity intrusion. Based on the research overview and data of salinity intrusion in the Mekong Delta, the author selected three indicators to measure the exposure of salinity intrusion, namely: (1) annual average salinity; (2) Salinity being above 4 ‰; (3) Salinity fluctuation in the month with most serious drought (April). Measurement of sensitivity: To measure the sensitivity, most authors used mainly three sub-factors including (1) health; (2) food sources and (3) water sources. Each sub-factor includes one of the component indicators. This dissertation also measured the sensitivity based on the above three components and components of the sensitivity were measured through the main water source for domestic use; untreated water for domestic use; crop diversification index; livestock diversification index; the monthly rate indicating the lack of two meals a day, the rate of injured or sick family members in need of caretaking, number of days for sick leave per person. Measurement of adaptive capacity: Adaptive capacity is measured through five livelihood assets (natural capital, human capital, physical capital, financial capital, social capital) and this method was also used by many authors (Pandey et al., 2015; Gerlitz et al., 2017; Zhang et al., 2019). Based on the research overview and on the availability of data on households' living standard in 2014, 2016, 2018, this dissertation measured natural capital via agricultural and forestry area per capita; land diversification; rice cultivated area per capita; cereal cultivated area per capita. Also, the author calculated human capital through the proportion of household members having jobs; Head of household with professional and technical qualifications; Head of household graduated from primary schools or higher education. In addition, the study measured physical capital via the durable goods diversity index; average net residual value; living area per capita; types of house. The author also calculated financial capital through access to savings; access to loans in cash and goods; livelihood diversity index and measured the social capital through the proportion of household members joining the union; number of support forms; and number of means of media. 11 Measurement of LVI Most of the above-stated studies used equal weight and this method was criticized because the same weights were applied to different components of LVI (Beccari, 2016; Miller et al., 2013; Abeje et al., 2019). Therefore, the dissertation will apply unequal weight as proposed by Iyengar and Sudarshan (1982). Accordingly, LVI is calculated as follows: LVI=E*We + S*Ws + A*Wa (iv) LVI ranges from 0 (least vulnerable) to 1 (most vulnerable). wE, wS, wA are weights of exposure, sensitivity and adaptive capacity respectively, in which wE + wS + wA = 1 are weights of exposure, sensitivity and adaptive capacity respectively, in which The calculation of LVI according to the formula (iv) will be explained in Chapter 3 - Methodology. 2.3 Livelihood Outcomes Previous studies all indicated that livelihood outcomes represent the outcomes of livelihood strategy through the use of resources. This is also the author's approach in carrying out the research. Accordingly, livelihood outcomes are achievements or outputs of livelihood strategies, such as increasing income, increasing welfare, minimizing vulnerability, improving food security and using sustainable natural resources. Scholars often used different indicators to measure economic performance, but income is widely recognized as the most important factor in sustainable livelihood outcomes (Ojong, 2011; Albert et al., 2007). Therefore, this dissertation also used income index to measure livelihood outcomes. 2.4 Impacts of LVI components on livelihood outcomes Impacts of exposure on livelihood outcomes: Studies showed that salinity intrusion is considered to be a serious issue that not only reduces agricultural potential, but also produces adverse effects on livelihood outcomes (Haider and Hossain, 2013). Therefore, the thesis proposed the research hypothesis: H1: The exposure of salinity intrusion negatively impacts livelihood outcomes of households in the Mekong Delta. Impacts of sensitivity on livelihood outcomes: As stated above, sensitivity is measured through three sub-components: health, water and food. Studies showed that when households have health problems, and they do not have access to hygienic water and food, it adversely affects their livelihood outcomes. Therefore, the thesis proposes the research hypothesis: H2 (a1-a7): seven indices of sensitivity (the main water source for domestic use; untreated water for domestic use; crop diversification index; livestock diversification index; the monthly rate indicating the lack of two meals a day, the rate of injured or sick family members in need of caretaking, number of days for sick leave per person) have a negative correlation on livelihood outcomes of households in the Mekong Delta. 12 Impacts of adaptive capacity on livelihood outcomes: As stated above, the dissertation used the framework of sustainable livelihoods by DFID (2001) due to the suitable classification of adaptive capacity according to types of capital. Accordingly, adaptive capacity is measured through five livelihood assets (natural capital, human capital, physical capital, financial capital, social capital). The results demonstrated that natural capital, human capital, physical capital, financial capital, social capital have positive impacts on households' livelihood outcomes. When farmers can increase these capital sources while applying them effectively in economic activities, it will increase their livelihood outcomes and vice versa, when the household does not invest in capital, their livelihood outcomes will not be as good as expected. Therefore, the dissertation proposed the research hypothesis: H3 (a1-a17): 17 component indices of adaptive capacity (agricultural and forestry area per capita; land diversification; rice cultivated area per capita; cereal cultivated area per capita, the proportion of household members having jobs; head of household without professional and technical qualifications; head of household not graduated from primary schools or higher education, durable goods diversity index; average net residual value; living area per capita; types of house, access to savings; access to loans in cash and goods; livelihood diversity index and the proportion of household members not joining the union; number of support forms; and number of means of media) have a positive correlation with livelihood outcomes of households in the Mekong Delta. 2.5 The regulatory role of adaptive capacity in minimizing impacts of salinity intrusion on households' livelihood outcomes Similar to the analysis of the influence of adaptive capacity on households' livelihood outcomes, when studying the regulatory role of adaptive capacity in minimizing impacts of salinity intrusion on households' livelihood outcomes, the author also concentrated on analyzing the regulatory role of five sources of capital respectively. Studies illustrated that households can reduce damage from natural disasters by increasing the resistance of assets exposed to impacts of disasters. This implies that in order to mitigate the negative influence of disasters, households can focus on enhancing their livelihood assets. Therefore, the dissertation proposed the research hypothesis: H4 (a1-a17): 17 component indices of adaptive capacity (agricultural and forestry area per capita; land diversification; rice cultivated area per capita; cereal cultivated area per capita, the proportion of household members having jobs; head of household without professional and technical qualifications; head of household not graduated from primary schools or higher education, durable goods diversity index; average net residual value; living area per capita; types of house, access to savings; access to loans in cash and goods; livelihood diversity index and the proportion of household members not joining the union; number of support forms; and number of means 13 of media) have the role of regulating to reduce impacts of salinity intrusion on the livelihood outcomes of households in the Mekong Delta. 2.6 Research model and hypotheses Based on the relationships in the theoretical basis and research overview, the author proposed the following model: Figure 2.3: Research model Source: DFID (2001), research by Hahn et al. (2009). Research hypotheses H1: Exposure of salinity intrusion has a negative correlation with livelihood outcomes of households in the Mekong Delta. H2 (a1-a7): seven components of sensitivity (the main water source for domestic use; untreated water for domestic use; crop diversification index; livestock diversification index; the monthly rate indicating the lack of two meals a day, the rate of injured or sick family members in need of caretaking, number of days for sick leave per person) have a negative correlation with livelihood outcomes of households in the Mekong Delta. H3 (a1-a17):17 component indices of adaptive capacity (agricultural and forestry area per capita; land diversification; rice cultivated area per capita; cereal cultivated area per capita, the proportion of household members having jobs; head of household without professional and technical qualifications; head of household not graduating from primary schools or higher education, durable goods diversity index; average net residual value; living area per capita; types of house, access to savings; access to loans in cash and goods; livelihood diversity index and the proportion of household members not joining the union; number of support forms; and number of means of media) have a positive correlation with livelihood outcomes of households in the Mekong Delta. 14 H4 (a1-a17): 17 component indices of adaptive capacity (agricultural and forestry area per capita; land diversification; rice cultivated area per capita; cereal cultivated area per capita, the proportion of household members having jobs; head of household without professional and technical qualifications; head of household not graduating from primary schools or higher education, durable goods diversity index; average net residual value; living area per capita; types of house, access to savings; access to loans in cash and goods; livelihood diversity index and the proportion of household members not joining the union; number of support forms; and number of means of media) have the role of regulating to reduce impacts of salinity intrusion on the livelihood outcomes of households in the Mekong Delta. 15 CHAPTER 3 METHODOLOGY 3.1. Approaches In order to fulfill research objectives, the sustainable livelihood framework by DFID (2001) (Figure 2.1) approach was used to discover factors facilitating or hindering farmers from developing and implementing livelihood strategies to achieve expected livelihood outcomes. On the basis of the sustainable livelihood framework by DFID (2001) and the theoretical framework for assessing livelihood vulnerability developed by Hahn et al. (2009), indices including exposure (E), sensitivity (S) and adaptive capacity (A) are chosen according to the function: V = f (E, S, A). The determination of these components is based on a review of previous studies. 3.2. Data collection method The study mainly used two sets of data: (1) salinity measurement data of 32 stations provided by the Southern Hydrometeorology Station to calculate the exposure and (2) data set of households' living standards in 2014, 2016 and 2018 by the General Statistics Office of Vietnam to review households' characteristics as well as calculate the sensitivity and adaptive capacity of households. In addition, the study also used other data related to population, labor and employment, economic development, poverty, income, expenditure, climate change, salinity intrusion... of the country in general and the Mekong Delta in particular. The data was collected from previous research and organizations such as Ministry of Natural Resources and Environment, Southern Institute of Irrigation Science, Institute of Meteorology, Hydrology and Climate Change, General Statistics Office, Provincial Statistics Office... 3.3 Data processing method 3.3.1 Calculation method of livelihood vulnerability index The process of identifying and calculating the livelihood vulnerability index can be summarized as shown in Figure 3.2. 16 Figure 3.2: The process of identifying and calculating livelihood vulnerability index 3.3.2 Estimation in Regression Model Estimation in regression model used in this study is econometric regression. The general research model is presented as follows. ln(Yijt) = βo + Ejtβ1 + Sijtβ2 + Aijtβ3 + AijtEjtβ4 + Gtβ5 +εijt (7) 17 In which: Yijt: household's average income i in commune j in year t; Ejt: exposure in commune j in year t; Sijt: household's sensitivity components i in commune j in year t; Aijt: household's adaptive capacity components i in commune j in year t; Gt: dummy variable of the year; εijt: unobserved variables. The author used the Hausman testing and it showed that the use of FEM model is more appropriate. Concurrently, the author utilized the commune-level fixed-impact regression to remove commune-level variables that do not change over time and cannot be observed (such as variables related to policies, infrastructure... in local areas). 18 CHAPTER 4 FINDINGS 4.1 Overview of the research area The Mekong Delta is located in the south of Vietnam, and it consists of 13 provinces, in which, eight provinces bordering the sea including Long An, Tien Giang, Ben Tre, Tra Vinh, Soc Trang, Bac Lieu, Ca Mau and Kien Giang. The region is characterized by fertile alluvium with a humid monsoon tropical climate. The irrigation resources of the Mekong Delta are complicated by the intertwined canal/river network as well as impacts of the Mekong's flow and two tidal regimes: the tides of the East Sea and the semi-diurnal tide of the Gulf of Thailand. Therefore, during the dry season, when the upstream discharge decreases, the tides strongly affect the upstream and inland canals, which results in severe salinity intrusion both in the river and in the land. Salinity intrusion remarkably affects the life, economy, society and livelihood of people in the Mekong Delta. 4.2 Measurement results of livelihood vulnerability in the Mekong Delta due to salinity intrusion The author evaluated the livelihood vulnerability in the Mekong Delta due to salinity intrusion through LVI index with five levels: from 0 - <= 0.20 being a very low vulnerability; > 0.2 - 0.4 - <= 0.6 being medium vulnerability, from> 0.6 - 0.8 - <= 1 is being very high vulnerability. In the research period, in 2014 and 2016, the LVI index reached 0.472 points (in the range of 0.4 - 0.6) within the threshold of average livelihood vulnerability, in 2018 this index reached the level of low vulnerability due to LVI = 0.390 (range 0.2 to 0.4). During the period of 2014-2018, vulnerability was higher in coastal provinces, especially Ben Tre, Kien Giang, Bac Lieu, and Ca Mau had the highest LVI value (ranging from 0.445-0.612). The results showed that there is a difference in vulnerability in the coastal area with the upstream and middle regions with a significance level of 10%, but the same degree of significance did not prove the difference in the LVI between the middle and upstream subregions. The results also demonstrated that the poorer households are, the higher their vulnerability iss, as these groups