Drained peatlands in the UK are leaking carbon dioxide into the atmosphere, and a new digital twin aims to help farmers stop it. Agriculture and land use account for 12% of UK greenhouse gas emissions, with 4% coming from cultivated and drained peatlands where drainage releases stored organic carbon as CO₂. A previous research project developed a self-learning digital twin—an AI-driven model that improves its own equations using field data from carbon flux monitors and satellite images. This follow-on funding targets a critical gap: getting that technology into the hands of farmers who can actually use it. The project partners with three leading farm-scale carbon calculator companies, which already serve thousands of UK farmers. By embedding the self-learning digital twin into these calculators, the team aims to validate and demonstrate the technology at NASA readiness levels 5 and 6—moving it from proof-of-concept to an operational tool. The goal is a 20% emission reduction from cultivated peatland by 2030, with the digital twin made accessible to at least 500 farmers. If successful, this could transform how agricultural policy and practice manage one of the UK’s most stubborn carbon sources, without requiring farmers to adopt entirely new systems.
View original technical description
Agriculture and other land uses contribute 12% of net greenhouse gas emissions in the UK. In the 4% coming from cultivated and drained peatlands, organic carbon is released to the atmosphere as carbon dioxide after drainage. In a previous research project, we have developed a self-learning digital twin. It uses Artificial intelligence (AI) and machine learning models in order to inform farmers and policy-makers how they can improve making decisions to reduce greenhouse gas emissions from drained peatland. The digital twin is self-learning because it is able to discover new model equations to improve its mathematical knowledge of greenhouse gas emissions based on observational data. It uses data from monitoring stations in the field that directly measure carbon fluxes together with satellite images. The current grant focuses on farmers and policy stakeholders as potential end users of the digital twin. The follow-on funding aims to broaden that engagement and build new partnerships with three leading companies offering farm-scale carbon calculators in the UK. The three companies have existing networks of thousands of farmers who are already using their carbon calculators for assessing their carbon footprints. The new engagement in this project will deliver greater benefits from the previous research project by making the digital twin information accessible to thousands of farmers. We plan to exchange knowledge with the three market-leading carbon calculator companies, farmers and policy-makers to embed the self-learning digital twin in the carbon calculators. This will accelerate the commercialisation of the technology. We also aim to build on the current proof-of-concept of the self-learning digital twin (NASA technology readiness level [TRL] 3) and integrate it with a farm-scale carbon calculator. This will enable us to validate (TRL 5) and demonstrate (TRL 6) the digital twin technology in an operational environment. In dialogue with the carbon calculator companies, we want to explore the market potential of this exploitation pathway, as well as policy opportunities and barriers and growth opportunities. Our objectives are to enable a 20% emission reduction from cultivated peatland by 2030; exchange knowledge with market-leading carbon calculator companies; and embed the self-learning digital twin technology into the carbon calculators to make it available to at least 500 farmers.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know