Background
Agricultural practices such as cover cropping, no-till farming, and agroforestry have been advocated to mitigate against climate change by storing and sequestering carbon into soils. However, we lack an understanding of their efficacy under diverse environmental conditions in a changing climate. Some practices can either sequester carbon or emit carbon into the atmosphere, and not knowing how climate change and climate variability may drive those can lead to promoting, via policy or other incentives, the wrong measure taken by farmers. This project will aim to provide a comprehensive, scalable framework for policymakers and farmers to promote climate-resilient, sustainable agriculture in the 21st century, and contribute to better decision tools in agricultural climate mitigation strategies.
PhD opportunity
This PhD project will explore the integration of machine learning (ML) and process-based models to evaluate alternative farming practices for carbon sequestration and reduction of GHG emissions. The work will enhance the predictive capabilities of process-based models simulating soil carbon dynamics and GHG fluxes coupled to future climate projections. A key aspect of the research will address the growing uncertainty of climate variability, using models to analyze the impact of increasingly variable weather on the efficacy of farming practices over time. For example, ML can be employed to analyze datasets of long-term agricultural experiments to estimate the outcomes of different farming strategies across diverse environmental conditions. Process based models, on the other hand, can couple to climate models and real-time weather data to predict these same using mechanistic approach. By addressing the increasing uncertainty of climate variability, this research will contribute to a more robust understanding of how sustainable agricultural practices can adapt and remain effective in the face of changing weather patterns.
The research will result in decision-support tools that can provide actionable insights for farmers and policymakers, enabling them to implement practices that maximize climate resilience and minimize environmental impact, addressing the increasing unpredictability of climate impacts on agriculture.
The student will be co-supervised by Prof. Guy Ziv (on the modelling/ML side) and Prof. Pippa Chapman (on the soil science aspects), both from School of Geography in the University of Leeds, and will gain expertise in AI/ML and process-based modelling, climate science, and agricultural systems, equipping them with valuable skills to contribute to climate-resilient agriculture in the 21st century.
Other information
Applicant profile: For this PhD project, students with a strong background in data science, machine learning, environmental science, or agricultural science are well-suited. Familiarity with process-based modeling, particularly in soil science or climate systems, would be advantageous. The ideal student should also have a keen interest in sustainable agriculture, carbon sequestration, and GHG reduction, with a capacity for interdisciplinary research and problem-solving. Strong analytical skills and the ability to work with large datasets will be essential.
