Agricultural landscapes are complex systems where choices about land use, crop management, and livestock production interact to influence environmental sustainability, productivity, and greenhouse gas emissions.  This PhD project investigates how digital twins can be applied for multi-objective optimization of these systems.

The project finds the spatial configurations of land use and management that balance between environmental and economic goals by combining process-based crop and livestock models, remote sensing data, and optimisation algorithms. The framework will be developed and tested on pilot farms in Denmark, using model simulations and detailed field data to evaluate trade-offs between productivity, nutrient losses, GHG emissions, and land-use change options.

The digital twin will eventually be expanded to the national level and used as an interactive tool for practitioners and students to help in decision-making and education. Through scenario exploration, the system allows users to observe how different land management strategies including low-productivity area reforestation and enhanced manure handling and crop rotation alternatives impact both agricultural output and environmental results and greenhouse gas emissions. By integrating model outputs, spatial data, and optimization results, the system aims to support resource-efficient and climate-smart farming practices across Denmark.