Use cases

GenAI4Earth validates its services through three scientific pilots acting as vertical demonstrators within EOSC. These are domain-specific yet transferable use cases, ensuring project methods apply across fields. Each pilot starts with clear scientific and societal goals, uses multimodal FAIR datasets, develops LLMs for data curation and foundation AI Models (FMs) for data exploitation, and validates results in real-world conditions. This consistent design makes the pilots comparable and highlights their complementarities.

• Urban Environments: Developing tailored foundation models to detect environmental changes, monitor climate stress, and assess risks in urban areas.
• Agro-Forestry Environments: Developing GenAI methods for trait-based phenotyping of crops and forest species to improve food security, sustainable agriculture, and forest resilience.
• Seismology: Integrating conventional seismology with emerging Fiber-Optics Distributed Acoustic Sensing (FO-DAS) data to detect and forecast natural hazards in complex geophysical settings.