Eco-schemes under the CAP
- Case studies
Feature image: Crop type occurrence in the demo field in Czechia captured by the PEOPLE4CAP project.
The Common Agricultural Policy (CAP) has evolved to prioritise performance-driven environmental and climate goals. Eco-schemes are at the heart of this transition, incentivising sustainable farming practices that go beyond compliance with mandatory requirements. These include crop rotation, soil conservation, grassland management, and more.
Monitoring compliance with these eco-schemes poses challenges for authorities. Traditional on-site inspections are labour-intensive, costly, and limited in scalability. As the transition to area-based monitoring systems advances, reliable and frequent data on agricultural practices is essential to ensure accuracy and accountability in subsidy allocation.
High-resolution satellite data from Sentinel-1 (radar) and Sentinel-2 (optical) satellites provide insights into crop health, vegetation patterns, and soil conditions. By analysing temporal vegetation indices such as NDVI, EO data enables:
- Tracking crop rotation, fallow lands, and soil cover.
- Monitoring grassland management activities such as mowing and grazing.
- Identifying tillage practices and assessing compliance with eco-scheme requirements.
Artificial Intelligence (AI) and machine learning models enhance EO data by detecting patterns, validating ground truth, and automating monitoring processes. This reduces the need for manual inspections and ensures a transparent decision-making framework for subsidy allocation.
Key examples
Monitoring Eco-Schemes with Area Monitoring Systems (AMS)
The "People for CAP" project, funded by ESA, developed innovative EO-based tools for CAP monitoring. Key achievements include:
- Crop Rotation Analysis: Detecting compliance with rotation rules and erosion-prone area regulations using crop classification data.
- Grassland Management: Identifying managed vs unmanaged parcels by analysing NDVI temporal profiles.
- Winter Soil Cover Monitoring: Evaluating the presence and extent of catch crops for eco-scheme compliance.
AI-Enhanced Tillage Detection
The "AI-Based Tillage Detection" project demonstrated the potential of using radar and optical data to identify conservation and conventional tillage practices. Ground truth data from farmers and paying agencies validated the models, enabling accurate classification of tillage types.
Farmer Engagement via Geotagged Photos
To complement satellite data, farmers can submit geotagged photos to resolve ambiguous cases. AI-driven analysis ensures photo quality and categorises field activities such as mowing or grazing, streamlining the decision-making process.