Digital twin for pest population monitoring

What if farmers or consultants could use a digital tool to see real-time predictions about when agricultural pests might become a problem in their farms or the wider landscape? What if they could query this digital tool for pest forecasts for the coming season? Or longer term, in their changing climate?

Perhaps the same digital tool could help them run scenario testing, using an accessible interface, about the impact of management interventions, or who they might need to work together with to most effectively suppress pests at a broader scale.

As part of our ongoing work to develop a Digital Twin (DT) of an agricultural landscape, we are developing a proof-of-concept DT of pest population monitoring that prioritises easy access to current information and provides tools to dynamically obtain future population predictions in real-time, while integrating climate, land use and pest management information.

This proof-of-concept DT demonstrates both the viability and suitability of the DT framework for agricultural pest monitoring and management; it also showcases the potential of DTs as valuable decision-support tools in the agricultural landscape.

A brief discussion of this DT was presented at the recently concluded 2024 IEEE Conference on Digital Twin (Fiji). Additionally, we’re working on making these and other scientific tools more easily accessible to all stakeholders, regardless of their technical expertise. A communication outlining our ideas about using Large Language Models and popular communication channels was recently published in Nature.

References

• Dhakal, S., and Parry, H. (2024). A Digital Twin for Pest Population Monitoring. In The 2024 IEEE Internation Conference on Digital Twin (Digital Twin 2024). December 2-7, 2024. Fiji.
• Dhakal, S., & Parry, H. (2024). Large language models can help to translate science into real-world impact. Nature, 636(8042), 299.