Destination Earth Machine-Learning Demonstrators
The European Centre for Medium-Range Weather Forecasts (ECMWF) contributes to the European Commission-funded Destination Earth (DestinE) initiative by developing advanced digital twin systems that integrate Earth System Modelling, observations, and high-performance computing. DestinE aims to enable data-informed climate adaptation, risk mitigation, and sustainable development, thereby supporting the European Union’s Green Deal and Digital Europe objectives. Through these efforts, DestinE delivers actionable, high-resolution climate and weather data to policy-makers, industry, and society at large.
Within the scope of this procurement, ECMWFEuropean Centre for Medium-Range Weather Forecasts More invites proposals to demonstrate the added value of Artificial IntelligenceArtificial Intelligence is the capacity of an algorithm to a More and Machine LearningField of statistical research focused on training computatio More (AI/ML) methods in leveraging the DestinE Climate Digital Twin. These demonstrators will support the expansion of AI/ML-driven approaches in the DestinE ecosystem, providing targeted, sector-relevant applications that align with DestinE’s co-design methodology. The main objectives of this tender include:
- Developing Impact-Sector Applications: The awarded contracts shall implement AI/ML approaches in three domains: water resilience, food security, and the integration of DestinE Climate Digital Twin data with other heterogeneous climate projections. The solutions should produce high-value information tailored to sector-specific needs and demonstrate clear potential for supporting climate-related decision-making.
- Integrating User-Driven Co-Design: The project teams shall ensure the active engagement of end-users and stakeholders throughout the development cycle. This includes applying co-design principles to ensure that AI/ML implementations address real-world challenges and provide relevant, scalable outputs.
- Advancing Simulation and Decision Support: The demonstrators shall highlight the capability of DestinE’s ecosystem—particularly its structured data and EuroHPC-enabled computing infrastructure—to support AI/ML innovation. Solutions should showcase how the fusion of physics-based models and data-driven methods can improve predictive insights and decision-support tools.
The initial contract period will support the development and deployment of these demonstrators, with the goal of expanding DestinE’s machine learning portfolio beyond existing pilot activities in energy system management, flood risk assessment, and data assimilation.
Depending on performance and evolving requirements, follow-on opportunities may include:
- Scaling AI/ML Capabilities Across Sectors: Extend the demonstrators’ methodologies to additional domains where detailed weather and climate information can offer operational and strategic value, such as biodiversity, urban planning, and transport.
- Strengthening Integration with DestinE Services: Enhance the interoperability and reusability of developed AI/ML components within the broader DestinE ecosystem, ensuring alignment with existing digital twin architectures and data services.
Tender responses must remain valid for at least nine months following the Closing Date. ECMWFEuropean Centre for Medium-Range Weather Forecasts More reserves the right to modify timelines as necessary.