Abstract
Vulnerability to water insecurity in Ghana has often been characterised as a biophysical phenomenon that can be objectively solved using technical and rational approaches to manage water resources use and allocation – thus, making vulnerability to water security discourse politically neutral. Contesting this orthodoxy, this article shows the multiple mechanisms and structures that undergird vulnerability of households to water insecurity in semiarid Ghana. Vulnerability to water insecurity is often veiled in political and economic imbalances that constrain the choices of people in their daily water realities. We argue that a rigid fixation on the biophysical conceptualisation of water insecurity separates nature from society while simultaneously ignoring the lived insecurities resulting from the contours of power. Policy formulation based on this conceptualisation without addressing the access rights and differentiated vulnerabilities could be misleading while yielding minimal impact.
Introduction
The struggle for a life of dignity in semiarid Ghana can be characterised as a hydro-social struggle in which water or lack of it shape livelihood outcomes. The undeniable seriousness of the situation is captured in the fact that water underpins life and all its associated activities. Livelihood activities, especially in the semiarid regions of Ghana, are carved around water (Braune and Xu, 2010; Lapworth et al., 2017; Wossen and Berger, 2015), with peasant farming as the dominant activity. Some scholars have characterised the endemic poverty in these regions as a direct consequence of physical water insecurity (Bonsu and Asare, 2004; Cook and Bakker, 2012; Xiao-jun et al., 2014). Water is thus more critical to livelihood outcomes in arid and semiarid regions compared to any other natural resource. Not only does its flow connect people, places and space, it also transforms landscapes, opportunities and economies. Water is a biophysical and social actant with cultural, spiritual and political meanings simultaneously embedded in it. For instance, it is a resource used for economic development, political boundaries and is a source of ancestral connection. It is this socionature of water that engenders contestation over ownership, control and access and users’ rights.
An increasing volume of research has investigated water security in Ghana. Within this inquiry focus, three points of concentration can be identified: The first strand examines sustainable approaches to water-source management and concentrates on physical availability, climate change (floods and droughts) and pollution (Armah et al., 2010; Bonsu and Asare, 2004; Douxchamps et al., 2014; Evans et al., 2012; Kankam-Yeboah et al., 2013); the second explores water delivery systems and is primarily concerned with how infrastructure, policy and water-governance structure results in water insecurity (Addo, 2010; Ainuson, 2010; Beck et al., 2016; Osumanu et al., 2010); and the third strand adopts an institutionalist approach to assess water security patterns in Ghana (Ainuson, 2010; Grönwall, 2016; Yussif, 2012). These studies generally ignore the role-power relations and entitlements in shaping the sociospatial pattern of water security. This study examines vulnerability to water insecurity in North-eastern Ghana using a political ecology approach. This broadly examines how power imbalance is reflected in resource exploitation, access and control. Political ecology offers an opportunity to combine social, political and biophysical factors in analysing water security. It integrates concerns of ecology and political economy to analyse how people interact to manage and contest access to and use of resources (Bridge et al., 2015; Carr, 2015; Wescoat, 2015). It thus provides a nuanced and contextualised understanding of the role of power imbalances in shaping the sociospatial patterns of vulnerability to water insecurity in rural northern Ghana, which is often deemphasised in the literature.
We draw on questions and methods inspired by political ecology to investigate the patterns and politics of water use in semiarid north-eastern Ghana, with emphasis on how power dynamics shape the sociospatial patterns of vulnerability to water insecurity and water resource access and use.
Understanding the lived worlds of vulnerable people and the mechanisms that give rise to such differentiated vulnerabilities to water insecurity are crucial ingredients for theory building. A full understanding of the endemic poverty and low human development in these regions is impossible without understanding the crucial role of water insecurity, and the mechanisms that undergird such vulnerabilities.
Interest in power relations in resource analysis emerged strongly in the mid-1980s when dependency theory dominated environment and development discourse. Neo-Marxist scholars and environmentalists at that time sought to highlight the urgent need to analyse access to and control over resources from a class-differentiating perspective as a direct result of the inherent contradictions of capitalism. Political ecology was born out of an attempt to reconcile power, resource use and environmental sustainability (Forsyth, 2008; Greenberg and Park, 1994; Tetreault, 2017). However, approaches to political ecology have been contested and mutated over the years to reflect broader shifts in methodologies to environment-development debate. Tetreault (2017) identifies two distinct epistemological and ontological strands of political ecology: materialist and post-structuralist perspectives.
The materialist school adopts a structural and class-based angle to explaining environmental degradation and resource insecurity. The premise of this school is that resources tend to move where capital and power is. At the local level, a materialist recognises that access and control over resources (water) are rooted in ‘local histories and social relations’ (Tetreault, 2017: 18), hence why an examination of differential entitlements inherent in society is indispensable in understanding the patterns of access to resources.
We adopt a materialist approach to argue that the sociospatial pattern of water insecurity is often underpinned by a complex interaction of environmental, social and political structures and relations interlinked at various levels and scales. We contend that differentially empowered actors determine vulnerability to water insecurity in North-eastern Ghana.
Political ecology and vulnerability nexus
The concept of vulnerability provokes competing conceptualisations and framings that defy a straightforward and universal characterisation. It basically implies the degree to which individuals, communities and systems are likely to suffer harm or injury to a specific devastating event or process. It invokes images of harm, weakness and incapability in the face of latent or real threat. Different scholars from diverse fields have pronounced different interpretations of vulnerability. The lack of universal consensus on what vulnerability should constitute, reflects the fact that it is contextually experienced. Social scientists’ and climate scientists’ usage of vulnerability often connote different things. The former tends to construe vulnerability as a mixed set of social, political and economic factors acting in part or in unison to undermine people’s ability to cope with or withstand adverse change. The climate scientists, on the other hand, interpret vulnerability in relation to the probability of occurrence, and the impact of weather and climate-related events such as floods, droughts, storms and water stress. Climate scientists tend to highlight the physical impact of the hazards of nature in their definitional shift.
To Chambers (1989), vulnerability to harm inherently has two strata: internal incapacity (the built-in characteristics of the individual); and external influences, which include the wider political economy of risks and exposure. As Chambers summarises, It means not lack or want, but defencelessness, insecurity, and exposure to risk, shocks and stress . . . Vulnerability here refers to exposure to contingencies and stress, and the difficulty in coping with them. Vulnerability thus has two sides: an external side of risks, shocks, and stress to which an individual or household is subject[,] and an internal side which is defencelessness, meaning a lack of means to cope without damaging loss (Chambers, 1989: 1).
Essentially, there are three themes that emerge from Chambers’ definition that need to be filtered out.
● the risk of exposure to crises, stress and shocks;
● the risk of inadequate capacities to cope with stress, crises and shocks; and
● the risk of severe consequences to – and the attendant risks of slow or limited poverty (resiliency) from – crises, risk and shocks.
In sum, the vulnerability of an individual, community or system is a reflection of its characteristic features and circumstances that may put them in harm’s way. This, in turn, renders them unable to cope with and/or recover from the damaging effects of the hazard or perturbation.
Gallopín (2006), in discussing the linkages among the vulnerability, resilience and adaptive capacity, describes perturbation as the major disturbance within a system beyond the normal variability that system is used to. It is often sudden and external to the system, although unstable internal elements may trigger it. For instance, a rainstorm in a nearby community may pollute the only surface drinking water source with debris leading to water insecurity. Unlike perturbations, stresses are slow and of continuous variability, which may be imperceptible at the beginning. Drought and population growth embody good examples of stressors. However, the distinction between stressors and perturbation can sometimes be blurred by geography, scale, time and the field of study. The twin concepts of perturbation and stress are conceptually bounded by the fact that they induce alterations in a community.
The degree of susceptibility of a community to damage is a function of its ability to respond to perturbation and return to its original status prior to the disturbance. Rather than a homogenous entity, a community is composed of a range of different individual and household endowment sets. Vulnerability invariably cuts across lines of differences, often exacerbating existing inequalities within communities. Adger (2006) refers to this ability as ‘sensitivity’. He defines sensitivity as ‘the degree to which a system is modified or affected by perturbations’ (Adger, 2006: 270). This general conceptualisation of sensitivity latently limits it to the adverse effects of perturbation. In climate change literature, the Intergovernmental Panel on Climate Change describes sensitivity as the degree to which a system is affected, either adversely or beneficially, by climate-related stimuli (Bates et al., 2008), such as by a decline in crop yields in response to a decline in mean rainfall in a season, or a poor household losing their last stock of livestock in a major flood. During other times, the effects may be indirect. A case in point is losing seasonal household income in the dry season due to a shortage of water in the village dam for dry-season farming.
Gallopín (2006) clarifies and summarises sensitivity as the degree to which a system, community and livelihoods, for example, are modified by an internal or external disturbance or a set of disturbances. Yaro (2004) argues that any attempt to distinguish between internal and external sensitivity is artificial and arbitrary. There is a constant flux of exchanges and interactions between both internal and external sensitivity. The system approach to vulnerability as proposed by Gallopín (2006) and Holling (1973) relies on creating categorical distinctions that may not exist in reality. Measuring vulnerability using an ‘either-or-approach’ is unnecessarily restrictive and buries the multiplex of interactions that give rise to vulnerability. They propose a more comprehensive approach that captures fully the household inability to avoid a negative outcome such as water insecurity. This view of vulnerability and its measurement enables a deeper understanding of the varying degrees of vulnerability amongst the different layers of society. Individual experiences with water insecurity in structured and hierarchical society filter through the contextualised broader arrangements of access to and use of the communal resource. For certain identifiable population groups such as migrant settlers – women, girls and younger men – access to and use of certain resources at the local level may decline as local power structures tend to determine access rights. Vulnerability is thus materially tied to the political economy of power as social structure and power intentionally or unintentionally exclude certain groups from access to and use of resources, namely water (in this case). These socially constructed and politically enforced inequalities lead to unequal distribution of access and use rights and consequently, opportunities that may come with access to water. Vulnerability to water insecurity is thus not politically neutral; it tends to follow existing contours of power in society.
In contributing to the categorisation for factors that influence vulnerability, Cutter et al. (2003) identified four interlocking and mutually interacting factors of limited political power: lack of access to resources, social and physical capital that influence vulnerability. This dynamic quartet broadly determine vulnerability to water security.
Description of study area and methods
Study area
Four agro-pastoral communities in the Garu-Tempane district of the Upper East Region of Ghana were studied. Worikambo and Kolmasug are to the southeastern part of the district, while Bugwia (Dabila) and Denugu/Danvorga are located to the south of the district capital, Garu. Bugwia (Dabila) was chosen due to the presence of ephemeral streams, which flood the community in the rainy season. Worikambo is particularly suited for this study due to the presence of a dam that continues to shrink in size. Similarly, Denugu/Danvorga was chosen because of the irrigation dam and a river. However, unlike Worikambo, Denugu/Danvorga has a high population density and characteristics of an emerging urban area. Such a comparative study allows spatial variations to emerge for policy discussion and intervention.
Unlike earlier study sites, Kolmasug is located on a rocky upland with limited natural sources of water and is largely dependent on rain-fed agriculture (Figure 1). This case provides a contrast, hence allowing for a deeper analysis of the nature and dimensions of climate-induced water insecurity and determinants of vulnerability. It also gives an opportunity for an in-depth appreciation of variation regarding the water insecurity of households and communities. These areas are typical of the Sudan savannah ecological zone, which is considered to be the most fragile to environmental change since it is at the confluence of the Sahara to the north and the Guinea Savannah to the south. The Sudan savannah is experiencing tremendous environmental change posed by desertification, declining soil fertility and climate change, thus worsening the already precarious water security situation. The Sudan savannah is characterised by pronounced dry and wet seasons. There is a unimodal rainy season from May/June to September/October. The average annual rainfall in the district is 800 mm. The lowest mean temperature is 18°C occurring in December/January, and the highest-mean-monthly temperature is 40°C occurring in March/April. The vegetation is mainly the Sahel savannah type, which consists of scattered drought-resistant trees and grasses that are often burnt during the long dry season. The most common economic trees found within the district are the dawadawa, 1 baobab and shea tree. Agriculture is the predominant occupation of the people in the area; approximately 82.5 percent of the district’s population is engaged in agriculture (Garu-Tempane District Assembly, 2013). The predominant crops are maize, early and late millet, sorghum, rice, soybean, cowpea, groundnuts, sweet potato, ‘frafra potato’ and watermelon. Livestock such as pigs, cattle, sheep, and goats are commonly reared. Cattle especially have sentimental relevance in a traditional sense, because they are seen as a measure of the wealth of households. Again, aside from providing meat and milk, cattle – specifically bulls – are used to draw ploughs. Poultry, especially fowl and guinea fowl production, is quite significant. In the dry season, crops such as onion, tomatoes, pepper, okra, cabbage and leafy vegetables are produced. The rivers in the district are the major source of water for dry-season, vegetable production. According to Garu-Tempane District Agricultural Development Unit (GTDADU, 2010), the district has eight dams and 14 dugouts that serve as a source of water for dry-season vegetable production and drinking water for animals.
Map of the study area.
Data collection and analysis
The study was conducted in two phases in 2016 and early part of 2017. The first phase covered the rainy season that spans from April to November, whilst the second phase was during the dry season from November to March. Such an approach is ideal for this study, as water security is characterised by both intra- and inter-seasonal variability. The nature, scope and dynamics of vulnerability adaptation and coping strategies vary in response to the different climatic seasons. Conducting the study in both seasons allows for a holistic characterisation of water security and deepens our understanding of areas that endure the dual dilemma of droughts and floods.
The research employed a mixed method approach. Mixed methods enable a good explanation of the multifarious socio-biophysical reality (Flowerdew and Martin, 2005; Teye, 2012) while giving the research both solid positivist grounding and an ability to reveal human agency. Questionnaires, survey and in-depth interviews, as well as focus group discussions, were the main tools for data collection.
The 2010 Population and Housing Census District Analytical Report of the Garu-Tempane district estimate the total number of households in the four study sites to be 1147 households. Using a simplified formula (Yamane, 1967) to calculate the sample size at 95% confidence level and assuming maximum variability to be 0.5 (p = 5) with a 5% level of precision, the equation is assumed as follows
where n is the sample size, N is the households in the study communities, and e is the level of precision. The household is the basic unit for data collection of this research. When this formula is applied to the population size of 1260, the result is as follows
Hence the sample household is 297 out of the 1260 households in the study area.
To ensure representativeness of sample size in each study site, the proportion of households in each community relative to the total population of the study communities was calculated. The number of households sampled in each community was then proportionally sampled relative to the sample size of 296 households. Although women and children may bear the burden of collecting water for the household, women’s burdens for collecting water affects the vulnerability of the entire household.
The survey sought to understand the various sources of water in wet and dry seasons; the causes and dynamics of water insecurity; the determinants of vulnerability to climate-induced water insecurity; and the variation in how households respond to changing water availability for agricultural and household uses, as well as the constraints to this adaptation.
Complimentary qualitative interviews and focus group discussions focused on discovering and understanding the experiences, perspectives and thoughts of participants by exploring meaning, purpose or reality towards understanding the lived worlds of people (Baxter and Jack, 2008; Patton and Cochran, 2002; Whitehead, 2002).
Eight in-depth interviews were conducted with farmers, and cases studies were built from these in-depth interviews. Four focus group discussions were held among farmers, both women and men. Women and children were essential in this study as they are primarily responsible for fetching water for household consumption. Again, patriarchal societal structures in the study exacerbate their vulnerability to water-related insecurities and risks. Studying such a group allows for a deeper understanding of vulnerability as they reveal their experiential stories. Themes of the interviews and focus group discussion included access to water resources and determinants of access. The rest were water governance, changing availability of water for household and agricultural use. It also sought to understand their perceptions of vulnerability and coping and adaptation strategies.
What makes households experience water insecurity?
In general, the study observed that the insecurity of a household arises out of a combination of several of factors. Two household case studies (Tiesema’s household in Scenario 1 and Alima’s household in Scenario 2) highlight what makes households experience water insecurity.
Scenario 1. Tiesema is the breadwinner of her household of nine members since her husband migrates to Kumasi seasonally to work in cocoa fields. He barely remits them. She is the only person working with seven other dependents. She brews pito—a type of beer made from fermented millet or sorghum in northern Ghana—and burns charcoal alongside but has no donkey to fetch water for her. Her daughters are too young to assist in fetching water, so after a tedious day of burning charcoal, she fetches water herself.
Scenario 2. Alima became the breadwinner of her household after her husband was diagnosed with arthrosis. She lives with him and her grandson. She brews pito, makes dawadawa and weaves basket. She is also a traditional birth attendant. 2 All her children are in Kumasi. They remit her to take care of her husband. She also has a donkey with a cart that fetches water from the dam for brewing pito. She, however, wished she had an irrigable land by the dam (refer to case study 2 for a detailed description of Alima’s household).
From the scenarios above, the following are key determinants of household vulnerability. First, transport assets are key in determining household water security; longer times and distances to water sources lead to collecting less water even if more is required. Table 1 further illuminates the assertion.
Determinants of the amount of water use and decision to use a water source.
Source: Field survey data for this study.
Compared to Alima, Tiesema is likely to fetch less water because she carries it herself. It has been widely documented in the literature that households with means of transport are water-secure because they are able to travel longer distances and fetch larger quantities from good quality sources (Armah et al., 2010; Bonsu and Asare, 2004; Ndaruzaniye, 2011). Households with no transport assets such as Tiesema’s are likely to override improved sources for proximity and convenience. This underscores the importance of time and efforts in household water security.
The second determinant of household vulnerability that emerges from the above scenarios is remittance and migration. Households with remittance tend to accumulate an asset base that enables them to generate further income, dispersing the impact of water insecurity on crops into other non-farm ventures.
The third determinant that arose from the above narratives is income diversification. Households with multiple livelihoods that are not climate-dependent, such as traditional birth attendant, cope better than those whose only source of livelihood is from agriculture and water-related activities. The following quote from an interview with a retired police officer who is also a farmer highlights the importance of income and wealth as factors that determine the vulnerability of a household to water insecurity.
Wealth is the main determinant of resilience. When I say wealth, I mean livestock [animals such as poultry, donkeys, and cattle] which others haven’t got. About two-thirds of the people haven’t got wealth to cushion them in times of drought. So in times of drought, they may have to starve. The capable household is able to endure. The two-thirds [of] households which are incapable have to get a member to migrate down south and remit them to survive. (Cletus, 58 years, retired police officer, farmer Worikambo)
Under normal circumstances, water collection is associated with women, although men sometimes get involved if the distance increases in the dry season. In all the study communities, women bear the brunt of the labour demand for water collection. This reflects the gender-specific activity pattern characteristic of patriarchal traditional norms. A 55-year-old man succinctly illustrates the gender aspect of water insecurity when he said, In fact, generally, women suffer more than men on the grounds that they care for the family. A man may leave the house for a week without coming home. Women cannot do that. If she is going, she does so with the kids. So she has to deal with her water needs and also satisfy the kids’ water needs too. Therefore the women suffer the most. Women are those who fetch water for the household.
The above finding supports the view that vulnerability to water insecurity reflects the existing structural inequalities in society (Agrawal, 2008; Boelens et al., 2016; Yaro et al., 2015). It also exposes the weakness of the reductionist approach to water security that emphasises aggregation at the individual, meso-level over micro-level security. The qualitative approach adequately presents determinants of inter- and intra-household water insecurity that is not sufficiently captured in quantification. This finding diffuses the tacit assumption of the quantification approach: homogeneity in a heterogeneously complex real world. The above findings consolidate the view that it is necessary to apply partly different strategies and policies to combat water insecurity by being cognisant of the contextual issues.
Broader determinants of household water insecurity
Households may experience different levels of water insecurity even though climate-induced water insecurity is a natural hazard that may be homogenous across the community. The individual characteristics of one household could make it more vulnerable than another, nearby household. The story of Tiesema in Scenario 1 is a portrait of how social conditions make households vulnerable to water insecurity. Tiesema’s household insecurity shows that a household’s water insecurity is not only determined by its idiosyncratic characteristics. Broader structural forces that set the arena within people’s vulnerability are contextualised. Tiesema has a large household with a high dependency ratio. Without her husband to help her out and her aged father and mother-in-law, she can barely cope. This is further exacerbated by customs that enforce gender roles. Her male children prefer tending the livestock than fetching water, leaving her almost exclusively responsible for household water security.
Case Study 1: Idiosyncratic and structural determinants of vulnerability
Case Study 1 presents a series of quotes from an interview with Boasup, a female farmer. She outlines the idiosyncratic and structural determinants of vulnerability. It shows how broader well-established relations can shape the vulnerability outcome for an individual household. Box 1 presents a series of quotes from an interview with Boasup.
The story of Boasup in Box 1 provides a summary of lived experiences that are routine to the many other households in the study area. It shows how multiple factors intersect to give rise to vulnerability. It also highlights how intra-household level power politics shapes vulnerability. Her story shows how male privilege and disguised misogyny makes women more vulnerable at the household level as compared to men. It also thrusts to the fore the nature of livelihoods and amplifies the gendered nature of vulnerability. Households whose income is carved directly around water are more at risk. In her case, alternative income activities such as dawadawa, brewing pito and making shea butter are constrained by lack of water. These activities are major sources of income for women in the study area. This throws light on the gendered and differentiated vulnerability to water insecurity that this thesis unravels. Boasup draws attention to the fact that experiences with insecurity are embedded and filtered through existing inequalities and privileges.
It emerged that the nature of the households and their resource base such as transport assets are critical elements in determining the amount of water to be collected and the usage of water from the different sources. For instance, households with donkeys and bicycles are less likely to reduce water consumption in the dry season compared to households that are not endowed with any means of transport. Women largely shoulder the responsibility of collecting water in addition to being the base of agriculture labour. This suggests that in times of water scarcity, women tend to be more vulnerable than men. This is a reflection of the patriarchal nature of the study communities. Patriarchy imposes male privilege at the expense of female labour and sacrifice. In a focus group discussion in Kolmasug, one woman sums up the everyday struggles of women in the community when the wells run dry.
In times of water scarcity, everyone is affected but women bear most of the brunt. We are responsible for fetching water in this community for household consumption. We are responsible for cooking hence naturally we fetch water for the household unless you have a daughter. Aside from that[,] our independent source of income[,] which is pottery, pito brewing, dawadawa and shea butter processing[,] is also compromised when there is water scarcity. These activities require water so when there is no water[,] poverty eats us. Water is a saviour.
Political ecology and access outcomes
Accessibility to land closest to the dam sites in both Denugu and Worikambo was shaped by social links and connections with the village chief and or royal line. The pattern has a semblance of the von Thunen’s land use model (O’Kelly & Bryan, 1996). The land immediately by the dam was exclusively dominated by households that had direct links to the ruling clan in both villages, with the periphery assigned to other clans. Figure 2 gives a pictorial illustration of this observation.
Spatial distribution of access to land at dam sites.
The above socially determined rules of access exclude the poor and most vulnerable in society. It denies vulnerable households of the opportunity to engage in income-generating activities and to climb the ladder out of poverty. Since the dam uses gravity to pump water into the canal, land that is closest to water, even when the dam is at its lowest, is the most valuable. Farms closest are thus more likely to get sufficient water before the valve is closed each day. Those at the fringes are inadvertently sacrificed.
It emerged from the interviews conducted that physical proximity to water does not automatically translate into access and use. Instead, connections with local political elite played a significant role in determining access outcomes. The story of Alima in Case Study 2 captures this claim. Despite being close to the dam canal, she and husband are yet to get a plot of land near it for dry-season farming.
Case Study 2: Identity politics and access outcomes; excluded by identity?
Alima is about 58 years old. She lives with her husband Anaba who is at least 60 in a fon (a small division) of Worikambo. Her three children, Asibi, Awin and Nyaaba[,] all left for greener pastures in Kumasi. They are alone with their 9-year-old grandson, Ayeebo. Alima is a robust woman with a desire for hard work. She is a woman of many jobs. She brews pito once a week but also prepares dawadawa (a local condiment for soup/stew) for sale in between. Aside from weaving cane baskets and mats, she is also a traditional midwife.
3
But with the new CHIPS
4
compound, she has no guarantee or stability of income from midwifery activity. With the exception of the declining traditional midwifery job, all her other incomes sources are dependent on water or rainfall. The straws for weaving are only available at the beginning of the rainy season each year. Dawadawa processing is tedious and requires more water. She is therefore compelled to carry the soaked and fermented seeds to the dam[,] which is about 1 km away. Anaba[,] though active[,] is a pale shadow of himself. He was diagnosed with arthrosis—a degenerative disease of the joints—probably from his former job as a mason. He helps with menial tasks at home although he does not have an independent source of income anymore. He depends on the irregular remittance from his children in Kumasi[,] although Alima admits it is insufficient. They are however okay since they have a donkey cart that Ayeebo uses to fetch water from the dam every two days before it is her turn to brew pito. Life would have been better if they had access to the land close to the dam to do dry-season farming. Alima relates that she has tried on several occasions to obtain land there from the chief[,] but she is yet to break through. ‘We are migrants. We are not a priority when it comes to land by the canal. Since we are strangers, the chief has not honoured our request’.
Even though Alima’s household is very close to the dam, they do not have access to land to farm in the dry season. Therefore, if there is inadequate rainfall during the main farming season, Alima and her family do not have the opportunity to cope using the dry-season farming as a buffer. Her story highlights the social context vulnerability.
Aside from performing their judo-political responsibilities as custodians of customary law, chiefs are one of the important managers of local communal resources including land and water. Chiefs are therefore critical in determining and protecting access and use rights to communal water resources through granting use rights to irrigated land.
Alima’s story highlights the importance of local institutions and how social structure may constrain access to resources and consequently successful adaptation. This validates Agrawal (2008) claim that climate adaptation is inherently local, thus local institutions such as chieftaincy mediate access and use of resources. These local institutions are important in understanding the portrait of vulnerability and the adaptive options that different social groups choose. Yaro et al. (2015) similarly recognise the importance of local institutions in shaping vulnerability and adaptive options. Amina’s ability to cope with the seasonal variability of rain by using dry-season irrigation is limited by social exclusion.
The satellite image (Figure 3) gives a visualisation to Alima’s story. It contextualises how ownership and association with power systematically exclude people who may be closer to water resources and irrigated lands. Even though Alima’s house is close to the chief’s palace as shown in the satellite image above, that does not guarantee access. She is considered an outsider within the local context despite her Ghanaian citizenship. Conventional spatial configuration literature has examined the connection between place and access to resources. The findings reveal that physical geographical proximity does not automatically or necessarily translate to access; instead, political and economic proximity defines access. The geographical location of water resources may be a blessing at community level, but individual access is filtered through social differentiation and relations of power. Social differentiation rather than spatial (locational) differentiation was critical in denying Alima access to irrigated land. Alima’s fate was decided by the local polity.
A map of the social context of vulnerability in Garu-Tampane.
The foregoing discussion reveals complex relationships of vulnerability, power, and access to water in particular and natural resources in general. In the context of the literature, the results highlight the limitations of the underlying assumption of the ‘Falkenmark Water Stress Indicator’ (Falkenmark, 1989), which inadvertently equates water availability to water security. Undue emphasis on aggregate water availability diverts attention from the more fundamental issue of how particular individuals and groups of people gain access to and use of water resources. In the context of Amina’s case, bonds and linkages to power are enabling or constraining factors for access to irrigation, as Figure 3 reveals. Agrawal (2008), however, does warn of the limiting nature of bonds and linkages, as they have the tendency of restricting close groups from participating. Again, vulnerability reflects the existing contours of inequalities dictated by local institutions. The finding also confirms the assertion of Yaro et al. (2015) that individual household vulnerability filters through the fabric of broader society laid by local institutions. Household vulnerability outcomes are as a result shaped by the contours of power relations.
Meso level determinants of vulnerability
Beyond the local political ecology, inter-community vulnerability was largely shaped by physical characteristics. Such a spatial differentiation of vulnerability according to Hartshorne (Hartshorne, 1959) is the call of the geographer. Table 2 shows a cross-tabulation and availability of water source throughout the year.
Water sources, cross tabulation study community, and perennial availability of water.
A χ2 test at 95% confidence level shows that there is a significant relationship between location and vulnerability to water insecurity (Pearson χ2 = 20.285; p-value = 0.000). See Table 3.
χ2 tests of the relationship between location and vulnerability.
Note: 0 cells (0.0%) have an expected count less than 5. The minimum expected count is 22.13.
df: degrees of freedom; Asymp. Sig. (2-sided): p-value.
Nearly three out of four households in Kolmasug had to find a farther alternative source of water in the course of the year due to their water source drying up. A similar portion of households in Bugwia also had to do same.
In response a to a question on water availability and access, one woman said, Access to water is a significant challenge in this community. There is no surface water from which our animals drink. They roam a lot before they get water. We depend on two boreholes of which one has very low yield. This community was named after the river but there is no water flowing there. It is a dry river bed. It has been so for about ten years. We also have wells but they are seasonal. (A woman in her forties, in a focus group discussion with women in Kolmasug)
This observation was corroborated by a man in a separate focus group discussion in the same community.
We get water from mainly boreholes and wells both for household and animal consumption. But these sources have low yields by February. In such times we go to Worikambo (6 km away). Animals either move to the dam in worikambo or the ‘Muare’ [a tributary of the White Volta, which serves as the borderline with the Republic of Togo]. Muare
5
is further afield. It is really far from here consequently our cattle either get lost or cattle rustlers steal them.
To understand the macro level vulnerability of both Kolmasug and Bugwia, an examination of the resource base and their physical characteristics are imperative. Kolmasug is located on a rugged yet gentle slope that becomes gentler as it unfolds towards Worikambo to the southwest. The landscape is underlain by clay with stony to rocky surface especially to the north and northeast. The clay serves as a good source of raw material for the dwindling pottery craft industry in this community. It has a dry river valley that cuts through northwest to the southeast and comes to life in the rainy season. It is this dry river valley that gives the community its name Kolmasug, which translates literally as cold river. The clayey underlay blocks percolation – the lifeblood for groundwater recharge. Wells and boreholes have low yields due to the declining recharge. The vulnerability of this community is linked to its location.
Bugwia is located on a sandy landscape with rocky outcrops. It is generally low with an imperceptible slope toward the south. Unlike Kolmasug, it has no surface water (neither a river nor a dam) and is dependent on rain and groundwater. However, the impervious rocky underlay poses a significant challenge to drilling for water. The few wells and boreholes available barely survive the dry season because they are shallow. Climate change and environmental degradation generally have reduced water availability in the rivers; these have become drier, but still serve as avenues for floods in a changing weather regime.
Worikambo and Denugu are relatively less vulnerable, in that they are in low-lying locations with relative pervious underlay that allows for recharge of the groundwater.
Inasmuch as a spatial display of vulnerability is relevant, it masks household level vulnerability. Although the physical characteristics of a community are important determinants of vulnerability, local norms and idiosyncratic household characteristics tend to be important determinants of vulnerability, as shown by the case studies. Critical factors such as wealth and asset base, social connection and livelihood ventures have a compelling impact on determining vulnerability. Again, vulnerability varies among and within a group, in line with the structure of society defining social vulnerability.
Conclusions
This study investigated the circumstances that predispose households and communities to vulnerability to water insecurity. Vulnerability to water insecurity for household and agricultural use are seldom separate. A rural household may experience both at the same time. This is often a continuum undergirded by the contours of power. Three factors determined vulnerability to water insecurity; the first is power relations as defined by the social structures.
In this first factor, powerful individuals and institutional coalitions often determine access, albeit there may exist open and universal usufruct rights to water guaranteed by Ghana’s 1992 constitution. In the case of Alima, although the dam is a public property sanctioned by the State, it is local arrangements and rules that determine access. Having legitimate claims to a property sanctioned by the state does not automatically translate to benefiting from same. Sikor and Lund (2009) emphasise that property [enforceable claim(s) to some use or benefit of something] is distinct from access which entails the ability to benefit from and use the property in question. So, people may have property rights to a resource but are unable to derive any material or nonmaterial benefit from the resource. In effect, they lack the ‘real’ (Cousins, 1997) or ‘effective’ (Verdery, 2003) rights to water promised by law but deprived by practice. Thus, access and control over resources (water) is rooted in ‘local histories and social relations’. That is, access outcomes filters through the existing contours of power – the political ecology of interaction. On a policy level, the yet-to-be implemented Government of Ghana policy of ‘one village – one dam’ assumes that water insecurity in northern Ghana is a biophysical problem of availability and that constructing more dams is the solution. We argue that a rigid fixation on the biophysical conceptualisation of water insecurity separates nature from society while simultaneously ignoring the lived insecurities resulting from the contours of power. Policy formulation based on this conceptualisation without addressing access rights and differentiated vulnerabilities could be misleading while yielding minimal impact.
The second factor that determined vulnerability is idiosyncratic characteristics of a household. The individual characteristics of a household could make it more vulnerable than surrounding households. Tiesema’s case study is a portrait of how prevailing household characterises predisposes a household to water insecurity. The poorer and larger the household, the more susceptible they are to water insecurity. Although large household size may provide labour for fetching water, such gain is eroded by the number of users of water. The nature of the source of household income is also critical. Households who earn their income from water-related activities such as farming, pito brewing and pottery are most susceptible to insecurity. Having non-water–dependent income sources such as government work and remittances reduces livelihood vulnerability to water insecurity.
The third factor, which is largely beyond the control of the households, is physical characteristics of a community involving the nature and configuration of the land. In communities such as Kolmasug and Bugwia where there is neither a river nor dam, households were primarily disadvantaged by non-availability of water points.
A household vulnerability is related to economic, political and social structures and processes, at both macro and micro levels that continually construct and reconstruct vulnerability outcomes. Seldom do these factors act alone. These forces often act in combination at different levels to determine vulnerability outcomes. Mindful of the roles that existing structures of inequality play to perpetuate unequal access to irrigated land, it is imperative that reforms are initiated to protect the poor. This reform could be in a form of bylaws that grant special access to vulnerable households. Such a reform requires that the district assembly work directly with traditional authorities in the study communities.
Footnotes
Acknowledgements
This article was completed while on a research stay at the University of Rwanda under the Africa Regional International Staff/Student Exchange: Food Security and Human Wellbeing Programme. We wish to thank the anonymous reviewers for their constructive comments on reshaping this paper.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.