Households’ vulnerability to climate change in Thua Thien Hue province

JOURNAL OF SCIENCE, Hue University, Vol. 70, No 1 (2012) pp. 227-236<br /> <br /> HOUSEHOLDS’ VULNERABILITY TO CLIMATE CHANGE<br /> IN THUA THIEN HUE PROVINCE<br /> Bui Dung The, Bui Duc Tinh<br /> College of Economics, Hue University<br /> <br /> Abstract. This paper measures the relative vulnerability of households living in<br /> Thua Thien – Hue province using the indicator approach. Information was<br /> collected via key informant interview, focus group discussion and a questionnaire<br /> survey of 597 households in the coastal, delta and upland areas of Thua Thien Hue<br /> province. It is established in the present study that households in the province are<br /> highly exposed to climatic hazards, particularly aquaculture and fishing households<br /> in the coastal and lowland areas. There is significant difference in adaptive capacity<br /> across different household groups. Household with aquaculture, cropping and<br /> capture fishery as the main livelihoods are highly sensitive to climactic hazards.<br /> Given the situation, agriculture and aquaculture should be given priority in<br /> interventions to enhance local adaptive capacity. High levels of exposure and low<br /> level of adaptive capacity are the main contributors to the vulnerability of<br /> households in the province.<br /> Keywords: climate change, vulnerability, households.<br /> <br /> 1. Introduction<br /> The IPCC Assessment Report (2010) defines vulnerability as: “The degree to<br /> which a system is susceptible to, or unable to cope with, adverse effects of climate<br /> change, including climate variability and extremes. Vulnerability is a function of the<br /> character, magnitude, and rate of climate variation to which a system is exposed, its<br /> sensitivity, and its adaptive capacity” (McCarthy et al., 2001). Vulnerability includes an<br /> external dimension that is represented by the exposure of a system to climate variations,<br /> as well as a more complex internal dimension comprising its sensitivity and adaptive<br /> capacity to these stressors. A highly vulnerable system would be one that is very<br /> sensitive to modest changes in climate, where the sensitivity includes the potential for<br /> substantial harmful effects, and for which the ability to adapt is severely constrained.<br /> Thus, vulnerability is understood as a function of three components: exposure,<br /> sensitivity and adaptive capacity, which are influenced by a range of biophysical and<br /> socio-economic factors. Exposure can be interpreted as the direct danger (the stressor)<br /> together with the nature and extent of changes in a region’s climate variables<br /> (temperature, precipitation, and extreme weather events). Sensitivity describes the<br /> 227<br /> <br /> 228<br /> <br /> Households’ vulnerability to climate change in…<br /> <br /> human–environmental conditions that exacerbate or ameliorate the hazard or trigger an<br /> impact. Exposure and sensitivity are intrinsically linked and mutually influence<br /> potential impacts. Adaptive capacity represents the potential to implement adaptation<br /> measures in efforts to avert potential impacts (Füssel and Klein 2006, Yusuf and<br /> Francisco 2010).<br /> Thua Thien Hue is located in the Central Viet Nam, bordered by the South China<br /> Sea to the east and by Laos to the west. The province has an area of 5,053 square<br /> kilometres 49,107 hectars of which are used as agricultural land. Another 180,412<br /> hectars are occupied by forests. Except Nam Dong and Aluoi districts that are located in<br /> the mountainous area, other districts are in the plain and strongly affected by inundation.<br /> Thua Thien Hue comprises of basins of four main rivers: O Lau, Bo, Huong, and Truoi<br /> rivers. The topography slopes downwards from the western Truong Son mountain range<br /> to the coast and is divided into three areas, i.e., higher mountain area, low-lying area,<br /> and coastal plain.<br /> The province of Thua Thien Hue is considered amongst the most disaster prone<br /> areas of Vietnam. In the past few decades, the frequency and severity of disasters<br /> increased significantly in Thua Thien Hue. Climate changes, especially extreme<br /> disasters killed many people and destroyed livelihoods of, and push many local<br /> communities dropped back poverty (PCFSC 2008 and 2009).<br /> Responses to reduce impacts by climate induced events such as floods and storm<br /> are not only the responsibility of the community itself but also a mandate of government<br /> agencies. The government must have adequate capacity to carry out tasks for climate<br /> change adaptation because successful implementations of adaptation strategy will be<br /> dependent on government’s performance. In fact, capacity for planning and action on<br /> climate change adaptation by local governments is lacking. An insight into how<br /> different household are vulnerable to climate change is of great importance to LGUs.<br /> Therefore, this study was to measure and explain households’ relative vulnerability to<br /> climatic hazards in Thua Thien – Hue province in order to suggest policy implications<br /> to local governments and for adaptation interventions at household level.<br /> 2. Research methods<br /> 2.1. Method to calculate vulnerability index<br /> This study used the indicator approach to measure the vulnerability of<br /> households in Thua Thien Hue province. The indicator approach indentifies indicators<br /> that reflect vulnerability and measures vulnerability by computing indices, averages or<br /> weighted averages for those selected variables or indicators. This approach can be<br /> applied at different levels (household, county/district, province and national). The<br /> indicator approach is valuable for monitoring trends and exploring conceptual<br /> frameworks. According to Leichenko and O’Brien (2002), composite indices capture the<br /> <br /> BUI DUNG THE, BUI DUC TINH<br /> <br /> 229<br /> <br /> multi-dimensionality of vulnerability in a comprehensible form. The indicator approach<br /> is the most common method adopted for quantifying vulnerability in the global change<br /> community. It is used to develop a better understanding of the socio-economic and<br /> biophysical factors contributing to vulnerability (Hebb and Mortsch 2007).<br /> Vulnerability indices of households were constructed based on the interrogation<br /> of a wide range of data sources following the notion that vulnerability is a function of<br /> exposure to climate change and variability, sensitivity to the impacts of that exposure,<br /> and the ability to adapt to ongoing and future changes (Hahn, Riederer, and Foster,<br /> 2009). The measurement of relative vulnerability using the indicator approach includes<br /> a number of important steps, such as indicator identification, assigning weight and<br /> calculating vulnerability index.<br /> Identification of indicators: The selection of indicators was done through an<br /> extensive review of previous reports; in particular, we draw from Gbetibouo and Ringler<br /> (2008), Smith et al (2006), Jusuf and Francisco (2010). These indicators were then<br /> pragmatically assessed through a workshop with the participation of LGUs and social<br /> scientists. This is to ensure that each indicator is practical, specific, measurable and<br /> time-bond. The study identified exposure indictors for five dominant hazards, namely<br /> storm, flood, drought, flashflood and extreme cold. Sensitivity indicators describe the<br /> natural, human, infrastructure and livelihood conditions that can either worsen the<br /> hazard or trigger an impact. Adaptive capacity indicators covered the types of assets that<br /> the local households have.<br /> Assigning weights: The issue of weightings is highly controversial largely due to<br /> the subjectivity inherent in assigning weightings. While the application of weights<br /> facilitates an indication of importance of the different variables, it also leaves the results<br /> open to manipulation. To take the local context and situation into account, weighting for<br /> each indicator, parameter and dimension should be used. Our review of literature<br /> indicates that there are several prevailing methods to assign weights to indicators. They<br /> are: (1) arbitrary choice of equal weight, (2) expert judgment, (3) statistical methods<br /> such as principal component analysis, and (4) consensus among policy makers and<br /> stakeholders. Each method has its own pros and cons. In the present study we do not<br /> assign equal weights because this strategy is too subjective, and the literature shows that<br /> indicators do not equally affect the vulnerability (Hebb and Mortsch 2007). The<br /> development of weights via expert judgment is often constrained by the availability of<br /> expert knowledge in smaller communities and difficulties in reaching a consensus on<br /> the weights among expert panel members. The use of statistical methods appears<br /> complicated and it is hard to involve the stakeholders in the exercise. Therefore we<br /> herein use the method to assign weights to indicators/dimensions through consensus<br /> among policy makers and stakeholders. Policy makers, local government units (LGUs)<br /> and stakeholders discussed and agreed on weights for each indicators/dimensions.<br /> <br /> 230<br /> <br /> Households’ vulnerability to climate change in…<br /> <br /> Calculating vulnerability indices: As discussed earlier, the vulnerability of a<br /> given system largely depends on its exposure, sensitivity, and adaptive capacity. The<br /> climate change vulnerability index was derived through the following steps:<br /> - We assessed the exposure using information from historical data of climaterelated hazards. We considered the past exposure to climate risks as the best available<br /> proxy for future climate risks.<br /> - We calculate hazard index for the climate hazard that households face, such as<br /> storms, floods, droughts, and extreme cold.<br /> - We analyzed socio-economic data of surveyed households and calculated the<br /> sensitivity indices.<br /> - We calculated the adaptive capacity indices for all surveyed households.<br /> - To obtain the overall index of climate change vulnerability, we get the<br /> weighted average of exposure (multiple hazard risk exposure), sensitivity, and the<br /> reverse of adaptive capacity indices.<br /> It should be noted that to make the indicator values are comparable across<br /> households we normalize indicator values using the following formula:<br /> Zij = (Xij – Xi min )/ (Xi max – Xi min)<br /> Where Zij is the normalized value of indicator i of commune j;<br /> Xij is the original value of indicator i of commune j;<br /> Xi max is the highest value of all communes; and<br /> Xi min is the lowest value of all communes.<br /> 2.2. Data collection<br /> Data and information necessary for the study are collected using several<br /> methods including focus group discussions, key informant interview, secondary data<br /> collection, and household survey. The sample for the household survey is 600<br /> households. They were chosen using stratification and random sampling methods. At<br /> first stage, the study stratified all communes into three groups based on their<br /> topographical feature: upland, delta and coastal. In consultation with LGU staff two<br /> communes were selected from each group. They are the upland communes of Huong<br /> Giang, Thuong Quang in Nam Dong upland district; the delta communes of Quang<br /> Thanh commune in Quang Dien and Phong Binh commune in Huong Tra district; and<br /> the coastal communes of Vinh Hai Phu Loc district and Hai Duong in Huong Tra district.<br /> Using the lists of households available at the communes, 100 households were selected<br /> from each commune. In-person interviews were undertaken for the sampled households.<br /> The number of interviews completed and used in the present study is 597.<br /> <br /> BUI DUNG THE, BUI DUC TINH<br /> <br /> 231<br /> <br /> 3. Results and discussion<br /> 3.1. Exposure to climatic hazards<br /> Hazard exposure is the main play in disaster risks to local communities. Local<br /> communities in Thua Thien – Hue are affected by various types of climate hazards.<br /> Table1 shows that household groups with different livelihoods are exposed to different<br /> hazards at different levels. The households who live largely on aquaculture and fishing<br /> activities and forestry and cropping are of more exposure to climatic hazards. Nonfarming household group is considered as least exposed to hazard. Over 46% of total<br /> households have exposure index of over 0.41 to 0.6 scales and about one fourth of<br /> households have exposure index of 0.61 – 1.0. There is a significant difference in hazard<br /> exposure of households who live in different topographical areas of the province.<br /> Households in coastal and delta area have higher level of exposure, as compared with<br /> households in the uplands.<br /> Table 1. Households hazard exposure by types of hazards and livelihoods<br /> <br /> Types of<br /> households<br /> <br /> Storm Floods Drought Landslide<br /> <br /> Flash Extreme Weighted<br /> flood<br /> colds<br /> Means<br /> <br /> Cropping<br /> <br /> 0.69<br /> <br /> 0.19<br /> <br /> 0.11<br /> <br /> 0.04<br /> <br /> 0.06<br /> <br /> 0.24<br /> <br /> 0.46<br /> <br /> Livestock<br /> husbandry<br /> <br /> 0.63<br /> <br /> 0.13<br /> <br /> 0.09<br /> <br /> 0.09<br /> <br /> 0.10<br /> <br /> 0.21<br /> <br /> Aquaculture<br /> & fishing<br /> <br /> 0.78<br /> <br /> 0.15<br /> <br /> 0.06<br /> <br /> 0.05<br /> <br /> 0.02<br /> <br /> 0.28<br /> <br /> Forestry<br /> <br /> 0.81<br /> <br /> 0.03<br /> <br /> 0.09<br /> <br /> 0.05<br /> <br /> 0.25<br /> <br /> 0.24<br /> <br /> 0.48<br /> <br /> Non-farming<br /> <br /> 0.74<br /> <br /> 0.13<br /> <br /> 0.09<br /> <br /> 0.06<br /> <br /> 0.06<br /> <br /> 0.20<br /> <br /> 0.38<br /> <br /> 0.42<br /> 0.49<br /> <br /> (Source: Calculation by authors using the household survey data).<br /> Table 2. Households hazard exposure by types of hazards and livelihoods<br /> <br /> T-Test<br /> Type of region<br /> Coastal communes<br /> <br /> Mean<br /> F<br /> <br /> Sig.<br /> <br /> 13.18389<br /> <br /> 2.5E-06<br /> <br /> 0.46<br /> <br /> Delta communes<br /> <br /> 0.52<br /> <br /> Upland communes<br /> <br /> 0.42<br /> <br /> Total<br /> <br /> 0.47<br /> <br />