AI-enabled carbon farming for climate-resilient smallholder agriculture

Carbon farming has significant potential to support climate change mitigation while improving the resilience and livelihoods of smallholder farmers. However, reliable measurement, reporting and verification (MRV) of carbon outcomes across fragmented and heterogeneous farms remains costly and challenging to scale. Current approaches often depend on extensive field measurements, creating a need for more efficient methods that maintain confidence and credibility of the carbon estimates while reducing monitoring costs.

This project will investigate how artificial intelligence and integrated data sources can optimise carbon MRV across smallholder farming systems. The research will assess existing MRV practices and investigate how ground observations can be combined with drone, satellite, environmental and farm-level data. AI-based models will be developed to estimate carbon outcomes and quantify uncertainty. An adaptive sampling and optimisation framework will then determine where, when and how much additional ground monitoring is required to achieve specified levels of accuracy at minimum cost.

The research will provide training in machine learning, remote sensing, spatial and environmental data analysis, economic modelling, uncertainty assessment, adaptive sampling and optimisation, alongside economic evaluation of monitoring strategies. The student will gain experience working with an industry partner and applying AI to a real-world sustainability challenge. The resulting framework aims to support more cost-effective, reliable and scalable carbon MRV, helping enable the expansion of carbon farming among smallholder producers.

Please note that this project is not open to researchers from Afghanistan, Cameroon, Myanmar, Sudan or Iran.