Interactions between Climate, Vegetation and People in East African Savannas: a Kenyan Case Study through the Post-Colonial Era.

The interactions between biotic and abiotic factors driving savanna vegetation structure are complex that a combination of resource-based and disturbance-based theories are used to explain the coexistence between trees and grasses. Human impact further complicates these interactions and consequently, the structure of wildlife populations. As human development is linked to environmental sustainability, understanding the impact of the interactions between changing climates and land use patterns on savanna ecology requires an interdisciplinary approach that integrates social and natural factors. In this thesis, the importance of rainfall variability in driving woody vegetation biomass, production and turnover across Kenyan savannas is first assessed. It is established that woody biomass and production increases with rainfall while turnover rates decrease with rainfall. Secondly, to explore the history of land use changes, perceptions from community elders in two savanna ecosystems in southern Kenya (Amboseli and Mara) are collated using a semi-structured questionnaire. The elders from Amboseli regarded rainfall variability as key in shaping land use change decisions while those in Mara regarded socio-economic factors and conservation initiatives as important determinants of land use types. Thirdly, to explore the impact of climate and land use change, an agent based model that used grass biomass data, simulated by a dynamic global vegetation model, as input data is developed. Development of the model incorporated natural and social factors by using insights from the vegetation survey and from the community elders. The model showed that provision of conservation subsidies, up to 200 $ yr-1 for 1 km2 grazing land, is key in driving livestock and wildlife densities and further increases in conservation subsidies maintains the density of livestock and wildlife. The interdisciplinary nature of this thesis highlights the value of integrating local community perspectives and science-based interventions to address the sustainability of savannas, particularly sub-arable savannas. It also highlights the value of conservation subsidies in promoting wildlife numbers and pastoral well-being.