SciTransfer
ThinkingEarth · Project

AI-Powered Earth Observation Models for Energy, Food Security and Climate Risk Management

environmentTestedTRL 4

Imagine a giant digital brain that has memorized every detail of Earth's satellite imagery. Instead of building a new AI for every single task, this project creates a general-purpose base that understands how the planet works. It's like teaching a computer to read and write before asking it to write a specific legal contract or a medical report.

By the numbers
15TB
Data volume of SSL4EO-S dataset
10TB
Volume of FoMo-Bench forest monitoring benchmark
43
Global flood events in Kuro Siwo dataset
13
Number of consortium partners
The business problem

What needed solving

Companies struggle to analyze satellite data because AI models are usually built for one specific task and fail when applied to new regions or different problems. This creates high costs and slow deployment for climate and energy monitoring.

The solution

What was built

Task-agnostic Copernicus Foundation Models and a graph representation model of the Earth. They also built five specialized large-scale datasets for land-cover, floods, forests, and vision-language understanding.

Audience

Who needs this

Agricultural insurance companiesRenewable energy developersEnvironmental monitoring agenciesClimate risk analystsForestry management firms
Business applications

Who can put this to work

Renewable Energy
enterprise
Target: Clean energy grid operators

If you are a grid operator dealing with volatile energy sources — this project developed foundation models that help accelerate Europe's clean energy transition and independence from fossil fuels.

Agriculture
mid-size
Target: Agri-tech insurance providers

If you are an insurer dealing with unpredictable crop failures — this project developed tools for assessing and modeling the impact of climate emergencies on food security.

Environmental Consulting
SME
Target: Biodiversity monitoring firms

If you are a consultant dealing with fragmented ecological data — this project developed a graph representation model of the Earth to better understand biodiversity conservation.

Frequently asked

Quick answers

What is the cost or pricing for using these models?

Based on available project data, there is no specific pricing model mentioned; the project is funded by a EUR 2,999,875 EU contribution.

Can this be scaled to an industrial level?

Yes, the project uses large-scale use cases and datasets, including the SSL4EO-S dataset which integrates approximately 15TB of data from Copernicus Sentinel missions.

Who owns the IP and how is it licensed?

Based on available project data, the licensing terms are not specified, but the project involves a consortium of 13 partners including 7 industry members.

How does this integrate with existing non-satellite data?

The project integrates distributed industrial and user non-EO datasets into large-scale use cases to solve socio-environmental problems.

What is the timeline for deployment?

The project period runs from 2024-01-01 to 2026-12-31.

Consortium

Who built it

The consortium is heavily weighted toward commercial application, with a 54% industry ratio (7 industry partners, including 4 SMEs). This suggests a strong focus on market viability rather than pure academic research, supported by 3 universities and 2 research centers across 7 European countries.

How to reach the team

Contact ETHNIKO ASTEROSKOPEIO ATHINON in Greece

Next steps

Talk to the team behind this work.

Contact us to explore how to integrate Copernicus Foundation Models into your environmental monitoring pipeline.

More in Environment & Climate
See all Environment & Climate projects