Carroll, Sarah Louise, authorReid, Robin, advisorBoone, Randy, committee memberSalerno, Jon, committee memberWittemyer, George, committee member2023-08-282025-08-282023https://hdl.handle.net/10217/236965In East Africa, rangelands and savannas are complex social-ecological systems with a history of land sharing among pastoralists, their livestock and wildlife. Today, many are systems of global importance for biodiversity conservation. As in Earth's other biomes, East African rangelands and their inhabitants face growing challenges as the result of global change, namely ecological, climate, political, and socioeconomic changes that are threatening wildlife populations and straining human-wildlife relationships. African savanna elephants (Loxodonta africana) are key actors in these systems and the center of many of these challenges. Thus, sophisticated land-use planning that addresses the resource use and needs of both people and elephants that can be integrated into conservation policy at relevant scales of governance is a key need for elephant conservation. However, studies of African elephant spatial ecology rarely examine both ecological and human-elephant relationships in the ecosystems they inhabit, whereas existing studies from human dimensions disciplines, which focus on social aspects of human-elephant relationships, very rarely include landscape-specific analyses. Additionally, detailed, spatially explicit information describing human-elephant interactions is limited, particularly in regions where people interacting with elephants still herd livestock more than they grow crops. This dissertation addresses this research need by applying novel methods with a holistic approach to examine the spatial dynamics of human-elephant interactions in multi-use landscapes with a focus on elephant interactions with pastoral peoples and their livestock. Most research in this dissertation is specifically focused on the spatial ecology of human-elephant interactions in the Greater Mara Ecosystem of southwest Kenya and combines inference from images captured by remote camera traps, high-resolution GPS tracking data, and social survey data to: 1) Understand how livestock and land management influence spatiotemporal patterns of elephant occurrence, 2) Investigate elephant movement behaviors and understand how people and environmental variation influence elephant movement behaviors, 3) Map core elephant habitat and movement corridors to support conservation planning, 4) Understand how people in mixed-use savannas relate to elephants and how elephants and wildlife conservation impact their lived experiences, and 5) Map and quantify social willingness to coexist with elephants in mixed-use landscapes to support conservation planning that accounts for the needs of both people and elephants. I found that elephants shifted the quantity and timing of their activity in community conservancies where livestock are present relative to the neighboring protected area where livestock were absent. Elephants were also more likely to occur in the protected area than community conservancies even when controlling for habitat variation. Finally, I found that areas in community conservancies used with higher intensity by sheep and goats, and separately cattle, were somewhat less likely to be used by elephants during the daytime. However, this finding was not consistent across years and the estimated effects had low precision and additional alternative analyses may make this relationship clearer. I then apply network theory to analyze combined information on movement path properties, use intensity, and structural properties of movement networks calculated from GPS tracking data to delineate the functional landscape of movement for elephants in the wider Mara-Serengeti ecosystem. After identifying movement behaviors, I then investigate the environmental variables driving different movement behaviors with a focus on delineating the habitats that support high elephant use and elephant movement corridors. Finally, I contrast how movement behaviors and the environmental variables driving movement vary between elephants inhabiting the mesic, wet savannas of the wider Mara-Serengeti ecosystem with those previously published and observed in elephants inhabiting the xeric savannas of the Samburu-Laikipia ecosystem. Results showed that human presence strongly influenced elephant movement behavior in the Mara and specifically influenced the location of core areas, whereas in Samburu, water availability and vegetation productivity and predictability strongly were the most important variables explaining core area use for elephants. Although vegetation productivity also influenced elephant core area use in the Mara, predictability did not, and human presence and canopy cover strongly influenced core area use in the Mara more strongly than water availability. Overall, these findings indicate that elephants in the Mara are likely less constrained by water and forage availability than elephants in Samburu and have more flexibility to access these key resources while minimizing the risks posed by people. I apply cognitive hierarchy theory to understand how elephants impact people in the Greater Mara Ecosystem by investigating values and attitudes associated with African elephants and elephant conservation in communities sharing space with elephants. I use data collected from semi- structured interviewers at 177 households across a mixed-use, agropastoral landscape that also functions as an unprotected elephant corridor and analyzed responses using Bayesian hierarchical models to quantify positive attitudes towards elephants while accounting for self-reporting bias. I interpret quantitative model estimates in the context of qualitative attitude assessments and sociocultural values to gain a deeper understanding of what explains attitudes towards elephants in the region. We found that although a majority of people expressed positive attitudes about elephant conservation in general, most were not also positive about sharing space with elephants on community and private lands at a local scale. Model estimates showed that people who believed that elephants had sociocultural value were the most likely to be positive towards elephant conservation in general, but experiencing conflict with any wildlife lowered the probability of respondents to have a positive attitude towards sharing space with elephants at a local scale. Qualitative data revealed that safety and well-being concerns related to the perceived threats that elephants pose to human life, livestock, and crops, coupled with few social and economic incentives to support elephant conservation in community and private lands contribute to low local positive attitudes. Our results suggest that conservation approaches focused on sustaining existing sociocultural values and relationships with wildlife while also investing in human well-being and safety measures and could improve conservation outcomes in shared landscapes. Overall, this research develops contributions to the understanding of human-elephant interactions in East African savannas and provides practical applications for elephant conservation. Specifically, this dissertation through the creation of several map products can support conservation planning that accounts for both people and elephants in the Greater Mara Ecosystem. Some of the most important takeaways come from co-interpretation of results with agropastoral communities and can thus provide direct guidance to conservation practitioners on how to better address human well-being in community-based conservation efforts. Though this research was produced in collaboration with non-government organizations, community-based conservation leaders, and government wildlife officials in Kenya, I recommend that future work can improve the collaborative research process by more successfully including local communities as stakeholders at all stages of the collaborative research process.born digitaldoctoral dissertationsengCopyright and other restrictions may apply. User is responsible for compliance with all applicable laws. For information about copyright law, please see https://libguides.colostate.edu/copyright.TextEmbargo expires: 08/28/2025.