Marine digital twins: A new era for transparent, smart oceans

The ocean covers 71% of Earth’s surface and contains vast mineral, biological, and energy resources, making it central to economic growth and scientific exploration. However, marine engineering must operate under harsh conditions such as corrosion, typhoons, waves, and structural fatigue, while traditional analytical and simulation methods struggle to meet the growing complexity and scale of deep-sea operations. Additionally, marine data are fragmented in structure, format, time scale, and precision, limiting information interoperability, prediction accuracy, and decision efficiency. Machine learning applications are expanding, yet constrained by insufficient high-quality data and slow simulation computation. Due to these challenges, there is a need to develop intelligent technologies such as marine digital twins for real-time modeling and analysis.

Researchers from Tsinghua University proposed a comprehensive framework and development roadmap for marine digital twin systems in a review published (DOI: 10.26599/OCEAN.2025.9470001) in Ocean in 2025. The review analyzes technical foundations, application cases, and future trends of digital twins in ocean engineering, providing a technical route for monitoring, simulation, prediction, and real-time decision-making in marine systems. Their work suggests that DT could significantly improve safety, efficiency, and sustainability in ocean exploitation and marine equipment operation.

The review refines the definition of Marine Digital Twin (MDT) and constructs a five-layer application architecture: perception layer, data layer, model layer, fusion layer, and application layer. The perception layer integrates diverse marine sensors for real-time monitoring, while the data layer enables high-capacity storage and database management. The model layer builds virtual structural, hydrodynamic, electrical, or environmental models using the finite element method (FEM), computational fluid dynamics (CFD), multi-physics coupling, and data-driven algorithms. The fusion layer synchronizes digital models with sensor feedback for state inversion and updating. Finally, the application layer enables visualization, fault diagnosis, lifespan prediction, optimization control, and risk assessment.

The paper summarizes applications across offshore wind turbines, ships, pipelines, subsea structures, autonomous underwater vehicles, and environmental monitoring systems. Case studies demonstrate condition monitoring, model updating using sensor feedback, fatigue life estimation, and strategy optimization for maintenance and energy efficiency. Researchers highlight the role of cloud storage, internet of things (IoT), artificial intelligence (AI), virtual reality / augmented reality (VR/AR), geographic information system (GIS), and edge computing in supporting MDT scalability and decision automation. Future challenges include data heterogeneity, multi-scale coupling, real-time computing cost, and model fidelity.

“The convergence of sensing, simulation, and intelligent computing is reshaping how we understand and operate the ocean,” the authors note. They emphasize that digital twins are not just virtual replicas but evolving systems capable of perceiving, predicting, and optimizing marine engineering processes. With real-time synchronization between physical and digital spaces, MDT provides engineers with a powerful tool for early warning, optimization, and decision-making, enabling safer large-scale offshore construction and deep-sea exploration.

Marine digital twins are expected to accelerate offshore wind deployment, reduce operation and maintenance costs, support autonomous vessel navigation, and strengthen disaster monitoring and environmental protection. Their ability to fuse heterogeneous marine data and simulate dynamic ocean systems could enable transparent oceans, enhance carbon-neutral energy development, and safeguard deep-sea operations. As advanced sensors, AI modeling, and cloud computing mature, MDT may become a core engine for digital ocean governance and the smart marine industry. Future efforts will focus on data standards, multi-scale integration, and high-performance simulation to advance large-scale deployment across global oceans.

Funding information

This work is granted by Guangdong Basic and Applied Basic Research Foundation (Grant No. 2022B1515130006) and the National Natural Science Foundation of China (Grant No. L2424326).

About Ocean

Ocean is an international, peer-reviewed, open-access journal that provides a multidisciplinary platform for cutting-edge research and practical applications in the fields of ocean science, marine technology, and marine engineering. The journal publishes articles, reviews, and perspectives aimed at advancing theoretical, numerical, site-based, and experimental developments to promote global sustainability.