Balancing ecological integrity and infrastructure development: Optimizing the UK's contribution to Sustainable Development Goals in sub-Sahara Africa

Lead Research Organisation: University of Glasgow
Department Name: College of Medical, Veterinary &Life Sci


Studentship strategic priority area:Sustainable use of natural resources
Keywords:Animal movement, infrastructure development, ecologically informed development


One of the common problems with environmental assessments of large infrastructure projects in Africa is that they often fail to plan for current and future wildlife distributions. The apparent absence of wildlife in an area, however, does not necessarily imply that the area is unsuitable for wildlife. Many wildlife populations are collapsing and/or individuals are only present for short portions of the year (e.g. migratory species). Given the difficulty of directly monitoring all populations in all areas, the only viable alternative is to model distributions as a probabilistic use based on habitat preferences observed intensively elsewhere. This approach forms the basis of this PhD.

Specifically, the methodological approach will:
1) Capitalize on existing collaborations through Tanzania Wildlife Research Institute (TAWIRI) and the Smithsonian Conservation Biology Institute (SCBI) to collate GPS collar data for multiple African species. Data sources include: Glasgow (Serengeti/Tarangire/Loita Plains wildebeest, zebra, eland, oribi, lions hyena), the SCBI (Chadian scimitar-horned oryx, Laikipia giraffe, Amboseli/Athi-Kaputiei wildebeest, East African elephant) and TAWIRI (wild dogs, elephant, rhino). In addition, GPS data may also be available through an upcoming meeting in South Africa (March 2018) to synthesize African-based collaring studies and includes vultures, kudu, buffalo, elephant, zebra, Sudanese kob. This represents a completely unique cross-taxa, multi-trophic movement dataset, and will ensure that the student begins in a strong position to carry out a successful and impactful project from the onset.
2) Movement data will be combined with readily available landscape metrics such as grass greenness (NDVI MODIS), seasonal rainfall and water availability (WorldClim), soil quality (World Harmonized Soil Database) and habitat structure (Sentinel-2) to understand how each species moves in response to environmental cues. The student will apply new classes of state-space movement models (e.g. hidden Markov models) that build on conditional resource selection to capture how environmental covariates account for the variation in the step lengths and turn angles of animal movement trajectories.
3) By examining how movement trajectories change near existing anthropogenic features, the student will quantify how infrastructure modifies functional behaviours (e.g. speed, directedness, foraging state), within different environmental contexts.
4) Secondly, remotely-sensed data products will be used to develop a multispecies " informed probability of connectedness" (i.e. a multivariate GIS surface of species-specific use, defined by landscape attributes) at regional or national scales which can be used for conservation planning and/or restoration efforts. We will identify existing or planned infrastructure projects through our network of collaborators (several projects in Tanzania/Kenya have already been identified, including the "Serengeti Road" (Hopcraft et al. 2015) and fencing in the Masai Mara). The probabilistic maps can be used within an optimization framework to assess which infrastructure designs or mitigation strategies will provide the greatest service to humans at the least ecological cost.
5) Ultimately empirical information will be used to parameterize simulations that will enable managers and policy advisors at WCMC a more informed assessment of the cost-consequence-outcome of alternative infrastructure development scenarios.


10 25 50