Use this URL to cite or link to this record in EThOS:
Title: Large carnivores under threat : investigating human impacts on large carnivores in East Africa
Author: Jacobson, A. P.
Awarding Body: UCL (University College London)
Current Institution: University College London (University of London)
Date of Award: 2017
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Large carnivores are a polarizing group of species that play an outsized role in relation to their number. They structure ecosystems and feature prominently in human culture. Yet, their place in a rapidly changing world is uncertain. The large carnivore guild in the five countries of East Africa, Burundi, Kenya, Rwanda, Tanzania, and Uganda, is largely intact; however, expanding human populations pose a substantial threat. Interventions are necessary to promote coexistence. To accomplish this, more accurate identification of threats, and improved understanding of species’ responses are needed. Primary threats to large carnivores in the region include habitat loss and human-wildlife conflict (HWC). Problematically, identification of human impacted areas from earth observation data can be difficult in heterogeneous savannah habitat, much of East Africa. I create a tool that enables land cover classification using Google Earth’s high-resolution imagery. With this tool I develop a data set of human impacted areas for East Africa. To ascertain carnivore response to human dominated lands, I use correlative species distribution modeling (SDM). Yet, there is no clear consensus on proper methods for generating pseudo-absence (PsA) data in these models. I review some existing methods in the context of their ecological meaning, and propose new PsA selection strategies. I then apply two novel and one existing PsA strategy to assess four carnivores’ (cheetah, wild dog, leopard, and lion) responses to human land cover and human population densities. Results suggest these carnivores are more susceptible to human land cover than human populations. Finally, I consider existing approaches of using SDM with HWC records to generate spatial risk maps with the goal of alleviating conflict. I draw on the SDM literature to highlight and demonstrate how two commonly overlooked issues in spatial risk modeling can hamper generating useful conclusions. In sum, these efforts represent attempts at improving commonly used methods used to study wildlife distribution and threats, and can be widely applied to other species and systems.
Supervisor: Disney, M. ; Durant, S. ; Pettorelli, N. Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available