Objectives
Soil Health RI Lab aims to perform research and validate innovative concepts that allow the development of EO-based products on soil health and quality assessment. Our RI Lab will experiment with sensor data and edge computing concepts that enable decision support in management applications related to soil fertility.
Technology / Methodology
- Use Federated AI to topsoil Soil Organic Carbon (SOC) model building at a regional level.
- Apply innovative standardization processes in developing soil health quality products that rely on satellite data to increase interoperability, data sharing and reuse.
- Mount hyperspectral sensors to increase the mapping ability on different soil parameters.
- Combine sensor data with satellite images to produce an optimal estimate of soil parameters.
- Application of edge computing to reduce bandwidth, enabling real-time feedback
Expected outcome
- Recommendations on using federated AI for developing EO-based Soil Health products at the EU level.
- Recommendations on using EO and edge computing for providing soil services related to soil fertilization.
- Methodologies that support further the standardization in the development of soil health products that rely on satellite data as a way, and increase the interoperability, data sharing and reuse in different vertical domains.
- Hyperspectral sensors in ScaleAgData will be aligned with the ESA Copernicus Hyperspectral Imaging Mission for the Environment programme by providing additional ground truth information required for calibration of the space-borne data.
Partners
Lab partners:
- EV ILVO - Flanders Research Institute for Agricultural, Fisheries and Food
- AUTH-Aristotle University of Thessaloniki
Technology providers:
- EV ILVO - Flanders Research Institute for Agricultural, Fisheries and Food
- AUTH-Aristotle University of Thessaloniki
- VTT
- EGM
- KUVA
Application area
Target crops: potatoes and barley
Target areas: Flanders and Central Macedonia