top of page

Transparent Carbon Footprint Quantification and Reporting in Agriculture 

Dr Milan Markovic, University of Aberdeen; Dr Paul Williams, Queen’s University Belfast; Dr Matthias Kuhnert, University of Aberdeen; plus an Industry Supervisor

​

Research Aims

​

Agriculture, responsible for 10% of global greenhouse gas emissions, holds significant potential for negative emissions, particularly in soil carbon sequestration. However, existing Monitoring, Reporting, and Verification MRV) systems lack standardization, transparency, and face challenges in data availability. Symbolic AI technologies (e.g., knowledge graphs) can be used to capture additional information such as geospatial context and data provenance of the data analysis which could enhance the transparency and thus support novel MRV systems.

​

This PhD project will explore practical applications of AI technologies to address the challenges in one or a combination of the following topics: 

​

  • AI-based quality and trust assessments of carbon footprint and sequestration reporting (e.g., using provenance knowledge graphs documenting calculation processes)

  • AI-based approaches to automate carbon footprint and sequestration reporting (e.g., based on semantic digital twins representing business assets) 

  • AI-based enhancements to resilience of carbon footprint and sequestration techniques (e.g., accounting for extreme events)

  • Automated validations of local data in federated Machine Learning infrastructures for privacy preserving soil-based carbon capture analysis

​

While the PhD is in the field of computer science, the successful student will be expected to apply an interdisciplinary approach to solving these challenges. This could involve gaining in-depth knowledge related to soil-based carbon capture analysis and greenhouse gas emissions, as well as applying user-centered design methodologies to the developed prototype software and be willing to carry out stakeholder engagement activities (e.g., onsite visits of industry partners). 

In addition to the interdisciplinary skills, the successful candidate should have good programming skills and general understanding of the AI technologies, especially symbolic AI techniques. The successful candidate will further develop their skills in designing, applying, and testing semantic applications in the context of real-world challenges. The successful candidate will be based in the Computer Science Department, University of Aberdeen.

​

​

bottom of page