PhD Studentship: Predictive Modelling of Biomass Loss from Forests

Location: Mathematics, Streatham Campus, Exeter 

The University of Exeter’s Department of Mathematics  is inviting applications for a PhD studentship fully-funded by the University of Exeter to commence on 23 September 2024 or as soon as possible thereafter.  

For eligible students the studentship will cover Home or International tuition fees plus an annual tax-free stipend of at least £19,237 for 4 years full-time, or pro rata for part-time study.  The student would be based in the Centre for Doctoral Training in Environmental Intelligence and affiliated to the Department of Mathematics in the Faculty of Environment, Science and Economy at the Streatham Campus in Exeter. 

Project Description:

Forests are pivotal in regulating the Earth's climate, sequestering a significant portion of anthropogenic CO2 emissions and maintaining biodiversity (Artaxo et al., 2022; Bonan 2008). However, the resilience of forests is under global threat due to escalating climate change impacts, leading to increased tree mortality and biomass loss (Allen et al., 2010; Hartmann et al., 2022). Understanding and predicting these changes is crucial for forest management and climate mitigation strategies (Anderegg et al. 2020; Millar et al. 2007). Recent advancements in remote sensing, exemplified by high-resolution biomass maps such as MapBiomas (Souza et al. 2020), provide unprecedented data on forests over extended time periods. Coupled with other remote sensing variables (e.g. NDVI, canopy height), these datasets offer a foundation for in-depth analysis of forest dynamics.

The primary challenge for this PhD project lies in accurately identifying and categorising tree mortality events. This requires the integration of diverse data sources, including local carbon and environmental conditions, as well as remotely sensed information. Developing a robust model that can effectively process and learn from these heterogeneous data sources is crucial. The model must not only understand current patterns of biomass loss but also predict future trends in the context of rapidly changing climate conditions. 

The project will employ a three-fold approach:

Data Integration and Preprocessing: Leverage high-resolution biomass maps and other remote sensing variables, integrating them with local carbon (e.g., GPP, biomass) and environmental conditions (e.g., climate, soils) and insights from the tree mortality network.

Machine Learning Model Development: Develop a sophisticated machine learning model to analyse the integrated data, identify patterns of biomass loss, and categorise these as mortality events. The model will be trained and validated using historical data, ensuring its robustness and accuracy.Model Integration and Prediction: Integrate the developed model into the JULES carbon cycle model (Argles et al. 2020; Clark et al. 2011), enabling the prediction of tree mortality and biomass loss under various climate scenarios. This integration will provide a powerful tool for forecasting and managing forest resilience in the face of climate change 

This PhD studentship is funded by University of Exeter. The student will be hosted by the UKRI Centre for Doctoral Training in Environmental Intelligence: Data Science and AI for Sustainable Futures (https://www.exeter.ac.uk/research/eicdt/). The EI CDT provides PhD students from a range of academic backgrounds with the core skills required for a new generation of interdisciplinary researchers to apply data science and AI techniques to environmental challenges. Training is provided in data science, AI and statistics; in ethics, governance, and responsible innovation principles related to the use of big data and AI; and in a range of environmental challenge domains. After initial training, the student will develop their research proposal around the topic outlined in this advertisement.

Advert information