Tesfahun Alemayehu Belew

Inland valley wetlands in West Africa are believed to be a conducive environment for rice-based farming systems and have gained increasing attention as a solution to improve food and nutrition security for smallholder farmers. Rice is the most consumed cereal in West Africa. However, domestic production meets only 40-50% of demand, leaving the region heavily reliant on imports. Moreover, the 2008 global surge in cereal prices triggered food shortages and social instability across West African countries. As a result, extensive rice production in the inland valleys has been considered a quick fix to address food security challenges. However, extensive rice production has often degraded ecosystem services in inland valleys without adequately meeting food production demands. This narrow focus on rice production has limited attention to the broader functions and services provided by inland valley systems, creating a knowledge gap that this thesis addresses. Research on the diverse uses and benefits of inland valleys and farmers’ preferences for alternative uses and development attributes in the inland valleys is limited. Therefore, this thesis aims to explore options to realise the potential of inland valleys to contribute to national food security. Four specific research questions guide the thesis: (1) the characteristics of inland valley farming systems and the relevance of ecosystem services for smallholder farmers, (2) the determinants influencing farmers' decisions to adopt inland valley farming and rice production practices, (3) alternative land use and development options in inland valley agriculture that are attractive to farmers, and how do they make monetary trade-offs between specific attributes of these options, and what factors drive potential heterogeneity in farmer preferences, and (4) the contribution of wild foods from the inland valley for farm households’ food security and poverty reduction and what factors influence farm households’ wild food collection, and the relationship between wild food collection and farmers’ awareness of provisioning ecosystem services. This thesis undertakes an in-depth characterisation of the current inland valley farm and farming systems. In addition, it investigates the factors that influence farmers’ decisions to adopt inland valley farming and rice production, and it explores farmers’ preferences for alternative land use and development attributes. Lastly, it examines and quantifies the contribution of wild foods to household food security and poverty reduction. In Chapter 1, I introduce key concepts and the theoretical background of the research, the research questions, and the methodology of the studies. Chapters 2-5 present the main empirical studies of the thesis. Chapter 2 provides a detailed characterisation of the current inland valley farms and farming systems, and farmers’ perceptions of ecosystem services. The study identified four farm types: (1) farmers who rent all the land they cultivate, (2) farmers who own some land and rent extra land, (3) farmers who own and cultivate all their land, and (4) farmers cultivating only a part of the land they own. I show that variations in farm households’ demographic and economic characteristics, and institutional services, such as access to credit and agricultural extension services, are greater between locations than within locations. The majority of farmers attach high relevance to inland valleys’ provisioning of ecosystem services, particularly the ability to provide food, followed by regulating services such as water storage. However, attached lower relevance to cultural ecosystem services, indicating an untapped policy opportunity for ecotourism or cultural preservation. The findings contribute additional insights to the literature on the use and development of inland valleys by highlighting the diversity in farmer resource endowments and underscoring the importance of the multifunctionality of inland valleys. In Chapter 3, the thesis identifies the factors influencing farmers’ adoption decisions regarding inland valley farming and rice production. Results show that access to credit and extension services, and physical capital, significantly influence adoption. In particular, owners of perennial tree crops are less likely to adopt inland valley farming and rice production than non-owners. This provides new insight into the role of ownership of perennial tree crops in agricultural adoption. Moreover, farm size has a different effect on adoption depending on the type of agricultural practices. Particularly, female farmers and older farmers were likely to adopt inland valley farming, reflecting unequal access to productive resources rather than a lack of interest. This highlights that adoption constraints are heterogeneous across household groups. The findings underscore that strengthening local institutions and service providers to target specific groups of farmers can help promote the adoption of inland valley farming and rice production. Chapter 4 builds on the results of Chapter 2 and goes a step further to explore farmers’ preferences for alternative land use and development attributes. I find that farmers demonstrate a strong preference for diversified production systems that balance rice production with biodiversity conservation and other ecosystem services. Moreover, on average, inland valley farmers express a preference for maintaining current biodiversity levels, as they perceive that further biodiversity loss would negatively affect their livelihoods and increase the need for subsidies. Importantly, this preference for the status quo does not reflect opposition to biodiversity conservation; rather, it indicates substantial heterogeneity in biodiversity preferences and suggests that farmers view existing biodiversity levels as broadly optimal for their production systems. Farmers appear willing to invest in biodiversity, but their support is shaped by their own assessments of how much biodiversity is beneficial, which in this context corresponds to current conditions. Taken together, the results in Chapters 2 and 4 suggest the importance of integrating farmers’ priorities and the diverse use of inland valleys to achieve sustainable and socially acceptable inland valley management strategies. Chapter 5 deals with assessing the extent of wild food collection from inland valley wetlands and quantifying its monetary value in supporting household food security and income. Building on the findings of Chapters 1 and 4, this thesis demonstrates that wild foods directly enhance dietary caloric intake, reduce the proportion of food-insecure households, and provide a vital source of income, lifting farmers above the international poverty line. For the poorest income quintile, wild food income alone covered 47% of the annual poverty line threshold and contributed 79% of total annual cash income. Excluding wild foods from food availability assessments would increase measured food insecurity by 7–15%. Building on the findings of Chapter 1, this thesis further establishes a novel, empirical link between wild food collection and farmers' awareness of provisioning ecosystem services. The results show that farmers involved in wild food collection demonstrate greater awareness of provisioning ecosystem services. The imputed monetary value of wild food energy contribution ranged from 149 to 470 USD PPP per adult equivalent annually, depending on food availability class, and the dietary energy contribution ranged from 576 to 2346 kcal MAE−1 day−1. These findings highlight that wild food collection is embedded within broader ecological knowledge systems. The study also finds that structural and situational drivers influence wild food collection. Taken together, the results from Chapters 4 and 5 suggest that farmers’ management of inland valleys is shaped by both ecological considerations and livelihood needs, emphasising the importance of integrating farmers’ priorities, diverse resource use, and socio-economic heterogeneity to achieve sustainable and socially acceptable inland valley management strategies. Thus, I argue that wild foods could be integrated into national food policies in a more tailored way to support both ecosystem management and rural livelihoods. Three main themes emerge from the thesis findings. First, farmers’ values and preferences play an important role in shaping adoption and land-use decisions. Farmers strongly prioritise provisioning ecosystem services, attach lower relevance to cultural ecosystem services, and express a clear preference for maintaining current biodiversity. Second, the results highlight important trade-offs between rice intensification and ecosystem services. Wild foods function as a critical safety net for the poorest households, contributing both income and dietary energy. When intensification is not carefully managed, it can undermine the ecosystem services that rural livelihoods depend upon, creating long-term social and environmental costs. Third, the findings underline the importance of recognising heterogeneity among farmers. Different groups experience distinct constraints and benefits. Female and older farmers, landless households, and farmers who own perennial tree crops are consistently less likely to adopt inland valley practices. Poorer households, meanwhile, depend more heavily on wild foods for both subsistence and income generation. Together, these patterns suggest that one-size-fits-all policies may unintentionally deepen existing inequalities rather than reduce them. Chapter 6 provides a synthesis of the main chapters and discusses the broader policy implications, limitations of the research findings, and ways forward for future research. Overall, this thesis shows that efforts to increase national rice production could benefit from considering the diverse uses and benefits of inland valleys and farmers’ preferences. The findings carry three priority policy implications: (i) managing trade-offs between rice intensification and ecosystem services through spatially differentiated zoning and buffer zones; (ii) supporting diversified farming systems aligned with farmer preferences for biodiversity; and (iii) tailoring adoption policies to heterogeneous household constraints, including gender, age, land access, and perennial crop ownership. Furthermore, this thesis indicates the need to support farmers’ diverse livelihood strategies, strengthen institutional capacities such as extension and credit services, and raise awareness about the broad benefits provided by inland valleys. Methodologically, this thesis combined farm typology analysis, econometric modelling, discrete choice experiments, and replacement cost valuation of wild foods. The findings contribute to the debate on whether extensive rice production poses a threat or an opportunity for inland valley ecosystem services. Evidence presented in this thesis shows that inland valleys directly support household food security and income, implying that poorly balanced extensive rice production degrades ecosystem services with significant consequences. In conclusion, this thesis demonstrates substantial room to realise the contribution of inland valleys to sustainable food security in West Africa.