DIVE

Digital twin for innovation in oceanic visualization and exploration

Context

Digital twins for oceans respond to the requirements to protect and restore ocean ecosystems, as well as the infrastructure assisting collaborations and generating the information and knowledge required to define strategies and policies. In a digital twin for oceans, the processing chain starts with the data acquisition from sensors, as in situ imaging, physical, chemical, biological measurements or airborne and satellite observations. Presently, data processing focuses on data-to-data transformations, while the information and knowledge extraction are still hindered by the data’s volume, diversity and complexity. An advanced convolutional neural network applications for ocean remote sensing was presented in 2024 by Haoyu Wang and Xiaofeng Li in IEEE Geoscience and Remote Sensing Magazine.

However, even the most recent advances in artificial intelligence, as Foundation Models or Causal Discovery, which achieve extraordinary adaptability to a diverse range of downstream tasks, are inadequate for ocean data. In that context, DIVE seeks to bridge the gap between ocean science and artificial intelligence, enabling actionable insights and fostering a climate-neutral, resilient blue economy.

Objectives

To facilitate the integration, visualization, and scenario prediction of data derived from biophysical ocean observations, simulation models, and remote sensing
To support decision-making for the management of sensitive marine communities and ecosystems in the face of climate change
To improve the accuracy of the decision-support tools for offshore wind farm operations
To develop tools for advanced space-based remote sensing observations

Scientific work planned

Definition and management of datasets

• Data management and acquisition
• Data cube and analysis ready data generation
• Benchmark procedures and tools

Definition of AI4Ocean methodology and development of the different modules

• ”What if” explorer digital twin component
• ”Causal” explorer digital twin component
• Ocean ”visualizer”

Tools demonstration through case studies

• Use case 1: Currents and sediment transports within the Western Black Sea coastal areas
• Use case 2: Lofoten-Vesterålen (LoVe) blue economy digital twin
• Use case 3: Rain radars and satellite observations applications to offshore wind
• Establishment of a 3 basins living lab
• Elaboration of a long-term strategic activity plan

Contribution of France Energies Marines

Contribution to the development of a digital representation of the ocean, Ie. digital twin, based on artificial intelligence tools that enable multimodal data to be taken into account

Expected results

Innovative decision support tools based on the “AI4Ocean” paradigm which consists of implementing 3 complementary functions:
- ”what if” explorer to improve or supplement existing coastal and ocean models
- ”causal” explorer to guide decision-making, explore, and validate cause-and-effect relationships, identify incomplete data sets, and provide measures of uncertainty
- ocean ”visualizer” to interactively explore and analyse raw or simulated data in order to facilitate and guide experts in their decision-making
Establishment of a joint living lab for 3 European basins, to demonstrate and scale-up the innovations and businesses based on the developed components in real-life scenarios to create sustainable impact via a strong iterative interaction with stakeholders

Fiche du projet

Pitch vidéo

Webinaire

Rapports publics

Conclusion

Lorem ipsum dolor sit amet consectetur.

Sagittis maecenas aliquam integer lobortis eu dignissim blandit tellus metus. Amet et tristique malesuada mattis pharetra bibendum purus id. Risus in elit ullamcorper ut. Imperdiet amet id feugiat arcu malesuada at ut vulputate condimentum.

Partners

This project is led by University POLITEHNICA of Bucharest.

Funding

This project receives funding from the Sustainable Blue Economy Partnership’s Second Joint Transnational Call (SBEP2024-31).

Labellisation

Lorem ipsum dolor sit amet consectetur. Sagittis maecenas aliquam integer lobortis eu dignissim blandit tellus metus. Amet et tristique malesuada mattis pharetra bibendum purus id. Risus in elit ullamcorper ut. Imperdiet amet id feugiat arcu malesuada at ut vulputate condimentum.

DIVE

Context

Objectives