Quantifying deforestation-induced climate risks to human wellbeing in Sub-Saharan Africa

Human wellbeing fundamentally depends on a liveable climate and access to safe and nutritious food. However, these two goals may increasing be in conflict: economic and demographic shifts are driving ever greater consumption, leading to deforestation for new agricultural land, particularly in the tropics. Such tropical deforestation leads to local warming and reduced rainfall, accounting for >30% of heat related deaths in deforested tropical areas and threatening agricultural yields—potentially driving further forest loss to maintain production in a positive feedback loop. This deforestation is also a major driver of biodiversity loss and can drive the emergence of new zoonotic diseases. Research by the supervisory team has also highlighted Sub-Saharan Africa as a region where demographic and economic shifts are likely to combine with low agricultural yields and high climate vulnerability to drive major agricultural expansion—with increased risks to people and the environment.

A major challenge for coming decades, therefore, is to understand and quantify how and where these risks will develop, and what societies can do to minimise them while maintaining food security. In particular, understanding the uncertainty around past impacts and projected future risks will be vital to allow decision makers—from policy makers to individual farmers—to make sensible choices in the face of this major societal challenge.

There is considerable uncertainty surrounding many of these issues, allowing you to tailor this project to your interests. A likely first step in your project would be to extend models and approaches previously used in Neotropical moist broadleaf forests (“tropical rainforests”) to understand how the loss of other habitats (e.g. savannas, dry forests) affects local temperatures and precipitation, potentially focusing on the risk of extreme events.

You could then explore a suite of questions focused on the impacts of such changes, applying data science techniques to estimate how such land-use change has affected risks to human wellbeing, biodiversity, agriculture, or zoonotic disease emergence. For example, understanding how the higher temperatures (but potentially lower humidity) induced by deforestation could have affected heat-related mortality and morbidity. Alternatively, you could focus on indirect impacts, for example the probability of climate-induced crop failures from high temperatures and reduced rainfall, or biodiversity loss.

A second suite of questions involves looking forwards: using machine learning approaches to estimate spatially explicit probabilities of future land-use change and linking these to projected probabilities of extreme heat events. You will be able to leverage previous models developed by the supervisory team and existing data on economic, demographic and agricultural trends. A key issue here would be whether uncertainty in local climatic conditions is driven primarily by uncertainty in projections of land-use change, or in the models that link such land-use change to climatic change.

Whatever the focus, you can work with project partners, potentially including RSPB and Rabobank to:

• Identify areas of particular importance for human wellbeing, empowering indigenous and local communities to manage their land to deliver both forest conservation and human well-being. E.g. building on ongoing work with RSPB in the Greater Gola Landscape, Liberia.
• Identify additional plausible public/private initiatives to incentivise decision makers to conserve key areas (collaborating with Rabobank’s on-going research).
• Leverage project partners’ extensive policy, finance, and conservation networks to disseminate results and incentivise greater support for forest conservation and restoration.

Applicant profile

To make the most of this opportunity, we are looking for students with a strong background in environmental science, climate science or other numerical discipline who want to apply their skills to an urgent societal problem.

Importantly, you need to be interested in understanding the uncertainty that is central to this research, as well as just making projections. Building a robust uncertainty of where this uncertainty comes from, and what it means for risk and robust projections is central to this project.

You will need to have strong analytical skills and be comfortable with large datasets. Desirable experience includes:

·      Handling Earth observation data and/or climate model output;

·      Data science or machine learning;

·      Very strong statistical knowledge and experience;

·      R or Python coding.

Further reading

• https://www.rspb.org.uk/whats-happening/news/gola-world-heritage, https://golarainforest.org/
• Reddington, C.L. et al (2025). Tropical deforestation is associated with considerable heat-related mortality. Nat. Clim. Chang., 15, 992–999.
• Smith, C., Baker, J.C.A. & Spracklen, D.V. (2023). Tropical deforestation causes large reductions in observed precipitation. Nature, 615, 270–275.
• Williams, D.R. et al (2021). Proactive Conservation to Prevent Habitat Losses to Agricultural Expansion. Nature Sustainability.