Identication of driving factors of land, water, energy and food (LWEF) nexus resources: A perspective of environmental degradation in Gidabo Watershed, Southern Ethiopia

: In Ethiopia, land, water, energy, and food (LWEF) nexus resources are under pressure, due to population growth, urbanization and economic growth. However, the impact of these factors are less explored at local scale level. As a result, securing nexus resource is becoming a serious challenge. This necessitates identification of driving factors for sustainable utilization of scarce LWEF nexus resources. Our study provides a systemic look at the driving factor indicator that induces nexus resource degradation. We use Analytical Hierarchical Process (AHP) to develop indicator weight and Path Analysis Model(PAM) to quantitatively estimates the effect of driving factor indicators on LWEF nexus resources. The result indicates that, social (48%), economic (19%) and policy and institutional changes (14%) are the major nexus resource driving factor indicators. The path analysis result indicates that from social driving factor indicators population growth, and consumption patterns have significant direct effect on LWEF nexus with the path coefficient of 0.15 and 0.089, respectively. Similarly, the potential of LWEF nexus resources is also influenced by the institutional and policy change drivers such as outdated legislation and poor institutional structure with the path coefficient of 0.46 and 0.39, respectively. This implies that population growth and consumption pattern are the leading social drivers, while outdated legislation and poor institutional structure are institutional and policies change drivers which have potential impact on LWEF nexus resources degradation. Similarly, other driving factor indicators (environmental, economic and technological) also affect nexus resources with varying amount. The finding of our study show the benefits of managing identified driving factors for the protection of LWEF nexus resources, which have close link with human health and the environment. In order to alleviate the adverse effects of driving factors, all stake holders need to show permanent individual and collective commitment. Furthermore, we underline the necessity of applying LWEF nexus approaches to manage these drivers, and optimize environmental and social outcomes.

distribution) that substantially contribute for degradation of nexus resources (Imasiku and Ntagwirumugara, 2020). This diminishes the capacity of nexus resources that perform developed the nexus resources driver indicators from local level perspective, for which our study area is not exceptional. In the study landscape the existence of LWEF nexus and the prevalence of their drivers appear to be high and vary throughout the area as a result of multifaceted of land-uses, water, energy access, and food insecurity that had been causing the nexus resource degradation (Dibaba et al., 2020). Therefore, this study attempts to select nexus resource drivers in relation to the prevailing conditions in the study area and finally use path coefficient analysis to model the causal relationship between LWEF nexus and drivers. With this, the study intended to (i) identify LWEF nexus resources driving factor indicators that affect nexus resource (ii) explore the extent and trends of a driver on four nexus resources (iii) identify the direct and indirect impact of driving factors on LWEF nexus resources.  This watershed is bordered by the catchment of Lake Hawassa to the North, River Bilate to the West, river Galana to the South and Genale-Dawa River basin to the East, which considered as a potential livelihoods sources for Southern Ethiopia. The populations are settled more towards the eastern highland and the population density decreases towards the eastern low lands. There is more population around the eastern highlands and it reduces as one goes down to eastern lowlands. However, the population of the watershed has been growing alarmingly in the last three decades and currently it is more than 1.5 million. The upper part is more populated (>500 inhabitants per square kilometers), and has immense impact on potential nexus resources. The livelihoods of the local community in the study area primarily depend on mixed farming and livestock rearing , which are sensitive to land, water, energy and food. Therefore, due to the current nexus resource degradation by various driving factors, most farmers in the watershed live with lowest access to electricity, clean water and modern infrastructure and they are vulnerable to frequent food insecurity.

Data sources and Techniques
This study focuses on driving factors indicator of LWEF nexus, which were identified from varied literature sources and deep survey in the history of nexus resources status. Additionally, data were collected from key informants and local community using combination of 5 structured interviews. Secondary data sources were also used for validation of LWEF nexus indicators. Identification of driving factor indicators helps to have common understanding and to easily categorize which LWEF nexus resources directly or indirectly affected by driver.
The survey was conducted within four-month period from July 2019 to October 2019, following two approaches. First, expert interviews and focus group discussion (N=50) were conducted with respondents from natural resources, agriculture, water and energy sectors in order to characterize driving factors (Table 1). Second, questionnaires were distributed to a total of 434 households, in order to find out how local community perceives identified driving factor indicators. To perform this, the respondents were asked to rank each predefined indicator impact on LWEF nexus from 0 to 4 ranking (4= extreme, 3= high, 2=medium, 1= low, 0= negligible).

Selection of LWEF nexus resources drivers
According to Taherzadeh and Caro (2019), driving factors are defined as driving forces consisting of different components that can affect societal change or natural system.
Understanding driving factors helps in identifying status, development and management of LWEF nexus resource to ensure equity and sustainability(Terrapon-Pfaff et al., 2018). The driving factor indicator is a variable that describes the status of the nexus resources and its impact on the availability, and distribution trends of nexus resources. In nexus resources management, a single indicator cannot efficiently describe a complex process that affects land, water, energy and food degradation (Spiegelberg et al., 2017). Therefore, indicators are combined to create a composite index to monitor the state of the nexus resources. Sola et al.
(2017) stated that environmental indicators may be considered as a simplified form to present information to a certain region. Therefore, in our current study, we consider nexus resources drivers' indicators to clearly understand area specific driving factors ( Table 1).
The nexus resources driving factor variables were selected in relation to the objectives and the prevailing conditions in the case study area from relevant literature and discussed further with expert's. Finally, the researchers and expertise from the four sectors were organized in a group, to provide feedback on interim results of the factors affecting nexus resources. Table   1 provides five main and twenty-five sub driving factors indicators and their explanation.
During identification and grouping of indicators, particularly sub-indicators, we tried to reduce overlap and mix up of sub-indicators to reduce multicollinearity of sub-indicators.
During our field investigation, the local community choses 47 different driving factor indicators. Following this, a discussion was held among expert from agriculture, water and energy, natural resource and environmental management sectors to score and rank sub-6 indicator and categorize into the main indicators. After careful evaluation, the experts bring identified 25 sub-driving factor indicators.
To verify this result we computed multicollinearity for the entire identified indicators using Variance Inflation Factor (VIF) analysis. The finding of the analysis revealed that 22 indicators had strong correlation with each other and removed from the analysis the final result of the analysis shows that the computed mean VIF is 3.4 which is satisfactory to show there is no multicollinearity. Therefore, finally 25 sub-indicators were identified and categorized into five main driving factor indicators. Besides, we also defined each subindicator in order to reduce data-based multicollinearity, which can be removed using subject-area knowledge and factor in the goal, as reported by (Graham, 2003;Shieh, 2010)

Analytical Hierarchical Process and Pair wise comparison matrix
The Analytic Hierarchy Process (AHP) and the Pairwise Comparison Matrix (PCM) were used to normalize indicators and establish the indicators weight from expert's judgment    Table 1). an endogenous variable is a response; however, the difference is that the endogenous variable can be both a predictor and response in a system of equation. An important assumption is that exogenous variables are measured without error, On the other hand, each endogenous variable is assumed to have error.
In path analysis, the correlation co-efficient is partitioned in to direct and indirect effect of independent variable (i.e. driving factors) on the dependent variable (i.e. LWEF nexus resources). In order to estimate direct and indirect effects of the correlated characters, let say 1, , 2 and 3 as driving factor effects on n nexus resources, a set of simultaneous equations is required to be formulated as shown blow; Where, denotes path coefficients, denotes simple correlation co-efficient. The total correlation between 1 and n is thus partitioned as follows ; 1 = the direct effect of 1 on n, 2 1 2 = the indirect effects of 1 via 2 on n, 3 1 2 = the indirect effects of 1 via 3 . Data processing was done by MS excel 2016 and Stata14 statistical package were used for analysis.

Analysis of land, water, energy and food (LWEF) nexus driver factor indicators
This section introduces main and sub-driving factors that affect one or more of the land, water, energy, and food nexus system. We identified twenty-five sub driving factors, which were categorized under five main driving factors. The results of the analysis indicated that from the five main driving factor indicators, social (48%), economic (19%), as well as institutional and policy changes (14%) are the major driving factors that affect nexus resources (Figure 3). Similarly, the average weight of the main nexus resource driving factor is also computed shows the same trends ( Figure 2). The direct and indirect impacts of these driving factors are explained in the subsequent section. Social life changes, economic sources and institutional and policy change are major driving factors, critical to ensuring land, water, energy and food nexus security. Nowadays, limited understanding towards those drivers results in poor LWEF nexus resource management (van Gevelt, 2020).

Social drivers
Six social driving factor indicators were identified based on the related literature (Table 1).  The demographic development and characteristics in the study area has implication for high demand on the existing patterns of land, water, energy and food availability and strongly driven nexus resource potential. The rural and urban population in the study area has been increasing from 2005 to 2020, and also expected to increase up to 2035 with increasing demand of nexus resource ( Figure 5). This population growth may also increase levels of poverty which results in unwise land use, water, energy and food. The relationship between population growth and nexus resource demand is changing and becoming complex due to increased consumption ( Figure 5). It also shows the impact of urban and rural population growth on land, water, energy, and food nexus alongside socioeconomic and political factors. Such kind of population increase needs adopting an integrated method of monitoring the LWEF nexus for the studied landscape.
The current institutional setup and implementation of policies in land, water, energy and food nexus are still facing challenges in Ethiopia. However, it is the only option to restore degraded nexus resources. The country becomes the second most populated nation in Africa which needs additional food production, which could be obtained by intensive management of LWEF nexus resources. Karlberg et al (Karlberg et al., 2015) showed that while on-going agricultural intensification require more use of energy, the energy required by the majority of smallholder farmers comes from biomass sources rather than centralized energy infrastructure.

Economic drivers
An increase in nexus resource scarcity is linked with economic well-being. Low economic potential declines the quality and quantity of nexus resources which enforces human beings to deplete, convert, pollute or degrade nexus resources (Agboraw and Jones, 2017). Figure 6 depicts five economic nexus resource driving factor indicator.   (Figure 6).
Regarding low capital (EC2), it has both direct (p=0.12) and negative indirect (p=-0.003) impacts on LWEF nexus resources ( Figure 6). This necessitates governments to set capitals to stimulate economic activity to meet particular sectorial development goal. The low capital, which is the characteristics of low economic potential have taken the brunt of the blame by causing unsustainable use of nexus resources which is caused by low investment.
Therefore, the identification of pathway of economic driver impacts on LWEF nexus resource degradation, which requires an understanding of direct economic driver and solving it through prioritization. To incorporate this understanding the pathway between LWEF nexus degradation and economic driving factor indicator need to be mapped ( Figure 6).
Increasing land value (EC4) indirectly drives (p=0.015) LWEF nexus resources which are mediated by increasing income variability ( Figure 6). The improper land use practices affect the availability of water, energy, and food which have environmental effects. This implies that due to economic development water and energy demand for food production is influenced by past land uses which in turn affect future land use. With this, the economic benefits of nexus resources by the local community have recently changed dramatically because wider forces such as overexploitation and unsustainable use permit them to do so.

Environmental drivers
A critical challenge for the environment over the coming decades is the demand for food production within the scarce land, water, and energy sources (Cosgrove and Loucks, 2015).
Decline on those nexus resources will be risky for food production in the future than in the past due to complex driving factors.
Environmental driving factors have a complex character and they also exhibit different impacts on LWEF nexus resources. As pressure in the environment grows, the nexus resources might change; consequently, there might be an increase in the degradation process.
According to Rijal et al. (2020), there may be different environmental driving factor indicator which selected in relation to the objectives and the prevailing condition. As depicted in Table 2

Technology as driver
Technological developments necessitate higher levels of productivity in terms of the use of land, water, and energy to increase food production. Technology can be expressed in various indicators, however, the current study identified four indicators based on expert's judgment and literature, such as lack of input supply (TC1), inadequate technology adoption and implementation (TC2), attitude towards technology innovation and development (TC3) and lack of infrastructure (TC4).
Based on the weight of the indicator, lack of input supply (TC1) is ranked as the most important technology driver that imposes LWEF nexus in the study area (Appendix Table 1).
Lack of access to technological input supply results in unproductive use of land, water, and energy resources (Kanianska, 2016). However, the technological processes may also produce unwanted results, such as loss of biodiversity, ecosystem disturbance, increased deforestation which create a trade-off on LWEF nexus. Table 3     In general, one of the major causes of prolonged environmental problems is a technology and how humans use it since it can be both source and remedy of environmental problems. It also plays a critical role as an instrument for observing and monitoring the environment on global and local scales, in agreement with the finding of (Ferreira et al., 2020)

Institutional and policy change as a driver
Aligned institutional structure and policy help to utilize nexus resources effectively, efficiently, and equitably. However, to date, the institutional structure and policy to integrate and manage the LWEF system as one component in Ethiopia entangled with complex problems of relevance, quality, accessibility and equity (Flintan and Tedla, 2010), which results in land, water, energy, and food insecurity.
Even though institutions and policy are working on Ethiopian forests, in the last three decades, the forest cover of the country drops from 40% to 3% (Gebrehiwot et al., 2014). This is linked with unintended institutional and policy change which induce land degradation, results on WEF nexus insecurity across the country.
The causes of failure in nexus resources insecurity were linked with frequent institutional and policy change. These changes became LWEF resource drivers (Table 1)

Socio-economic impact
Land, water, energy, and food resources are critical for the development and survival of societies (Liu et al., 2017). However, access to these resources is complicated by factors driving nexus resource degradation. According to Mohamed (2017), following the complexity of LWEF nexus resource, more than 18.1 million people in Ethiopia are forced for frequent nexus resource insecurity which will affect different socio-economic components.
The result shows that access/availability of food (34.6%), overall electric supply (26%), and poor health conditions (10.57%) are key socio-economic characteristics strongly affected by LWEF nexus degradation in the study area (Table 3). LWEF nexus degradation affects access/availability of food by 34.6% as compared with other socio-economic characteristics (Table 4). This implies that land use change, water and energy crisis challenges food security under changing climatic conditions. Understanding sustainable management of land, water and energy is one of the keystones for establishing livelihood security, maintaining the provision of ecosystem services and adaptive capacity against nexus resource degradation. Socio-economic characteristics such as age (8.11%), poor health condition (10.57%), and level of education (8.85%) affected by LWEF nexus resource degradation (Table 4). Nexus resource degradation affects socio-economic conditions of human wellbeing, characterized by a poor health condition, low level of education, low food supply (Wolde et al., 2020). The current study area has good water potential compared with other nexus resources however, irrigation water potential was reduced by 7.1% due to land use change (Table 4). This causes the population in the study area to continuously depends on food aid, supported by USAID (MOHAMMED, 2016).

Ecological impact
Due to resultant nexus resource degradation in the study area ecological balance was disturbed, which characterized by loss of biodiversity and migration of rural people to search for additional land, water and food.
Land, water, energy and food degradation also interrupts the regulating and provisioning services of the ecosystem, in particular through agro-ecological variations, climate change, forest loss, gully formation, soil erosion, loss of biodiversity, and drying of wetland (Coates et al., 2013). During our field investigation, the farming community reported that land use change, water potential reduction, lack of fuel wood, and declining food production was observed in the last four decades. Following this ecological variation in 2017 frost occurrence was observed in the highland part of the study area that affected the major livelihoods of the local community by declining productivity of crops particularly coffee and Enset (Figure 9). Figure 9. The impact of frost due to climatic variability in the highland part of Gidabo watershed Climate change caused by nexus resource degradation reduces ecological efficiency to supply productive land, sufficient water, safe energy and food ( Figure 10). The study area has been subjected to drastic ecological change attributed to climate change, forest loss, loss of biodiversity and low agricultural productivity which affect the overall environment ( Figure   10).
The threats to ecology arise because of change in the quality and quantity of LWEF nexus which underpin all ecological processes , and lack of stakeholders to invest in nexus resources result on low conservation of biological diversity and ecological integrity.

Figure 10. LWEF nexus resources degradation impact on ecological characteristics
Figure 10 reveals that climate change and low water potential, forest loss, and soil erosion, gully formation and biodiversity loss and low agricultural productivity are all interdependent characteristics of ecological degradation because any action on one of these characters can affect others. Climate change and low water potential ascertained by forest loss and soil erosion, it may also affect land use and loss of biodiversity. Most important is that low agricultural productivity, from the nexus component, food production is highly affected by climate change, and low water potential, forest loss and soil erosion and loss of ecosystem.

Livelihood impact
The livelihoods of rural communities entirely depend on the availability and consumption of land, water, energy and food nexus resources, which directly or indirectly affect the livelihoods activities (Table 5). We used a structured household survey to explore how livelihoods affected by nexus resource degradation. Table 5 summarizes the response of 434 households on how LWEF nexus degradation affects livelihoods.
As shown in Table 5, crop production (38.1%), livestock rearing (16.3%), and agroforestry practices (10.4%) were the major livelihood activities affected by LWEF nexus resource degradation in the study area. We are unable to obtain exact details about the size and extent of impacts during the interview. However, crop production and livestock are potential livelihoods are affected by nexus resource degradation. Spiegelberg et al. (2017) reported that sustainable management of WEF nexus is one of the keystones for establishing livelihood security.
Beekeeping (5.29%), fishing (9.9%) and fruit production (7.5%) are also an important livelihoods activity affected by the degradation of nexus resources (Table 5). Therefore, LWEF nexus resource degradation hampered the livelihoods of rural communities and becoming a major challenge for sustainable development.
Most rural poor are smallholders practicing low-input agricultural production, which need a substantial amount of productive land, water and energy. Lack of these nexus resources could have triggered a reduction in food production, energy, and water access.
Energy insecurity is a significant challenge in the study area which affects the expansion of the school, health center, institutional structure and other. Furthermore, the lack of available land, water and energy affects small-enterprise by 12.5% (table 5), which is a source of income for the rural community. Generally, policymakers in the land, water, energy and food sector need to gauge the influence of the driving factors on LWEF nexus resources in order to identify the problem and carry out effective strategies. In this context, the identification of drivers of LWEF nexus is essential not only for small geographical unity, but also for the sustainable management of basic nexus resource potential in the long run.

Conclusion
Nowadays in accelerated global socio-environmental change the land, water, energy and food nexus has received increasing attention in academia and policymakers to restore nexus We argue that many nexus discourses focus on population growth and urbanization as a major nexus resource driver. However, these drivers were studied from a macro scale and growing megacities for the last decades, but the variability of these drivers from region to region was under looked at and needs policy attention. Therefore, this study focuses on understanding social, economic, environmental, institutional and policy change, and technology as a driver which induce degradation.  Indicators   SC1  SC2  SC3  SC4  SC5  SC6  EC1  EC2  EC3  EC4  EC5  IP1  IP2  IP3  IP4  EN1  EN2  EN3  EN4  EN5  EN6  TC1  TC2  TC3  TC4                      The social driving factor indicator (SC1= population growth, SC2=poverty, SC3= lack of alternative livelihoods, SC4= consumption patterns, SC5= community awareness, and SC6= lack of public involvement) Figure 5 Population projection in the study area (data source: Central Rift Valley document) Figure 6 Economic driver indicator of LWEF nexus resources (EC1= increasing income variability, EC2= local capital, EC3= increasing WEF prices, EC4= increasing land value, and EC5= inadequate nancial resources)

Figure 7
The direct and indirect impact of institutional and policy change driver on LWEF nexus (Outdated legislation (IP1), inadequate nancial capital (IP2), poor institutional structure (IP3) and poor stakeholder network (IP4)) The impact of frost due to climatic variability in the highland part of Gidabo watershed Figure 10 LWEF nexus resources degradation impact on ecological characteristics