Supervisors: Dr David Williams(University of Leeds), Dr Casey Ryan (University of Edinburgh), Prof Carla Staver.

Climate change, human modification, and changes in biodiversity are fundamentally altering ecosystems across the world. In West Africa, these changes are leading to shifts in the distribution of forest and savanna ecosystems, but exactly what changes are occurring and why remains largely unknown.

Odzala-Kokoua National Park (OKNP) in the Republic of Congo is one of Africa’s oldest parks and contains a unique mix of ecosystems and biodiversity, from savanna-forest mosaic through to primary tropical rainforest—part of the second largest expanse of rainforest in the world. It holds thousands of forest elephants, gorillas, chimpanzees, West Africa’s last remaining spotted hyenas, and over 100 other mammal species, as well as over 400 species of birds.

However, a combination of climatic change and human pressures could lead to fundamental shifts in OKNP’s ecological communities, potentially leading to the loss of iconic species and unique ecosystems such as bais—elephant-created clearings in the forest.

Responding to these changes requires an in-depth understanding of how ecosystems are changing, and what the consequences of these changes will be for biodiversity, carbon stocks, and people.

This PhD will combine cutting edge artificial intelligence (AI) and machine learning (ML) with earth observation data to understand how ecosystems in OKNP are responding to climate change and other pressures.

Aims and objectives: this PhD will (1) develop open source AI and ML tools to understand how OKNP’s ecosystems are changing, and what is driving these changes; (2) apply these tools across West Africa to build a comprehensive picture of how the savanna-forest boundary is likely to change in the future; (3) combine these projections with long-term biodiversity and carbon stock data to build an understanding of how these changes will affect people and the environment.

Methodology: the student will use a range of earth observation products to first build an understanding of how the savanna-forest boundary in OKNP is changing. These products will include long-term, coarse resolution data (e.g. LandSat, ALOS, JERS) to detect broad-scale changes (e.g. in the position of the savanna-forest boundary); and higher resolution data (e.g. Planet, Sentinel 1 and 2) to investigate finer-scale shifts (e.g. changes in bais). They will then use a range of AI and ML approaches to automate the detection of changes and use the most promising to develop open-source tools to identify hotspots of change across OKNP and the wider region.

The student will then use Bayesian statistical approaches to link observed changes to long-term data on climate change, agricultural land-use change, and natural resource use, using a range of on-the-ground and earth observation datasets. The project will then focus on projecting how these causal agents will change in the future, using regional climate change scenarios (e.g. from CMIP6) and land-use projections from the supervisory team (Williams et al 2021). Depending on the student’s interest, projected and observed changes can also be linked to biodiversity data from earth-observation and on-the-ground datasets, above- and below-ground carbon stocks, and other ecosystem service data.

Supervisory Team: The student will be supervised by Dr David Williams (University of Leeds, expertise in conservation science; land-use change and land-use policy; biodiversity), Dr Casey Ryan (University of Edinburgh), expertise in remote sensing of deforestation and degradation; African savannas and woodlands) and Prof Carla Staver, Associate Professor of Ecology and Evolutionary Biology; Associate Director of The Yale Institute for Biospheric Studies. The team will work closely with the OKNP research and monitoring team to ensure the project is as relevant and useful as possible.

Candidate description: A strong quantitative background and interest in remote sensing and environmental issues (essential); interests in AI, machine learning, and spatial data analyses (desirable); interest in, and willingness to work with, policy makers and decision-making processes.

Potential Project partner/collaboration with African Parks