GreenTwin: A Digital Twin Solution for Emission Sustainability in Agriculture
Coach Name
Jordi Bosch i Garcia
EU Organization
Libelium LAB
Members
- MetaWorldX & The Pennsylvania State University
US-CA Organization
Libelium LAB SLU
Members
- MetaWorldX Inc., The Pennsylvania State University
Project Overview
GREENTWIN tackles a critical gap in the agricultural sector: while farming is a major contributor to greenhouse gas emissions, its true environmental impact is difficult to measure at local level.
The project introduces a Digital Twin of agricultural systems, combining IoT sensors, advanced modelling, and data spaces to estimate emissions such as methane (CH₄), ammonia (NH₃), particulates, and nitrogen compounds.
Validated in a real-world pilot in a mango orchard in Spain, GREENTWIN enables farmers, cooperatives, and policymakers to monitor emissions at field level, evaluate the impact of specific practices, and simulate mitigation strategies.
By transforming complex environmental data into accessible insights, the project supports more sustainable, data-driven agriculture and policy-making.
Methods and approaches
Hybrid Emissions Modelling Combining IoT and Scientific Models
GREENTWIN integrates real-time IoT sensor data with advanced atmospheric and emissions models to produce accurate, field-level estimates.
This includes:
- In-situ measurements of soil, air quality, and meteorological conditions
- Integration with the CHIMERE chemical transport model for background data
- A hybrid approach combining top-down (Gaussian inversion) and bottom-up (emission factors) methodologies
This multi-layered modelling framework allows precise estimation of emissions such as PM₂.₅, PM₁₀, VOCs, CO₂, NH₃, and N₂O at parcel scale.
Digital Twin Interface and Stakeholder-Centric Design
The project translates complex environmental data into interactive dashboards and 3D visualizations, enabling non-experts to understand and act on emissions data.
Through stakeholder engagement (farmers, cooperatives, policymakers), the system was designed to:
- Provide actionable insights for decision-making
- Support evaluation of farming practices (e.g., fertiliser use, fuel consumption)
- Enable scenario simulation and future planning
This ensures that the solution is not only technically advanced but also usable, accessible, and aligned with real-world needs.
Key Achievements
Development and validation of a hybrid emissions-assessment pipeline combining IoT data and advanced modelling.
Successful deployment of IoT sensors for soil, air quality, and weather monitoring in a real agricultural field.
Generation of a detailed emissions baseline for the pilot plot, covering multiple pollutants.
Creation of a functional Digital Twin interface with dashboards and 3D visualizations.
Definition of a scalable business model and replication strategy, including SaaS offerings.
Delivery of best practices and policy-oriented guidelines to support adoption and scaling.
Impact & Results
Scientific Impact
GREENTWIN demonstrates a fully operational hybrid emissions modelling framework at field scale, combining IoT sensing, atmospheric modelling, and uncertainty analysis. It provides a replicable methodology that advances research in environmental monitoring and digital twins for agriculture.
Societal Impact
The project increases transparency around agricultural emissions, enabling stakeholders to understand the environmental impact of farming practices and make informed decisions. This supports more sustainable agriculture, improved communication with consumers, and evidence-based policy development.
Economic Impact
GREENTWIN establishes a clear pathway toward commercialization through a SaaS-based model, offering digital twin services, consulting, and data-driven insights. It opens new opportunities in AgTech and EnviroTech markets, particularly for precision agriculture and sustainability monitoring.
Publications and Open-Source Contributions
- Green Twin Emissions Report (detailed methodology and results)
- Best Practices and Business Strategies report
- Business Model Canvas and replication guidelines
- Social media dissemination (LinkedIn, Facebook, X)
- Planned release of open-source components and project repository
Future directions