The ocean is one of Earth’s most dynamic yet least accessible systems—a vast and ever-changing regulator of climate, ecosystems, and resources. Capturing its complexity requires observation frameworks that are precise, interpretable, and scalable across diverse spatial and temporal scales—from fine-scale turbulence and sediment transport to basin-wide circulation and air–sea exchange.

However, underwater environments remain particularly challenging for observation due to light attenuation, scattering, turbidity, and complex nonlinear interactions among physical, chemical, and biological processes. These limitations obscure optical clarity and restrict the integration of imaging data with quantitative ocean models. Recent breakthroughs in computational imaging, artificial intelligence (AI), and physics-informed modeling are transforming this landscape. Deep learning–based restoration, differentiable 3D reconstruction, and physics-constrained inversion techniques have made it possible to retrieve structural and radiometric information from degraded underwater data. Emerging paradigms such as Neural Radiance Fields (NeRFs) and 3D Gaussian Splatting (3DGS) further enable physically consistent, high-fidelity representations of underwater environments.

At the same time, the expansion of IoT-enabled sensor networks, autonomous observation platforms, and edge–cloud computing has enabled continuous and real-time ocean monitoring. When coupled with data-assimilative models of ocean circulation, biogeochemical cycling, and climate–ecosystem interactions, these systems form the foundation for a connected digital ocean framework linking observation, interpretation, and prediction.

In parallel, marine system science has advanced toward integrated modeling of physical, biological, and chemical processes, offering new insight into climate variability, air–sea feedbacks, and ecosystem dynamics. Progress in coupled atmosphere–ocean modeling has deepened understanding of regional and global climate feedbacks, while studies of coastal, lagoonal, and open-ocean ecosystems have illuminated how climate change, anthropogenic forcing, and natural variability influence biodiversity, productivity, and system resilience.

Bridging computational imaging and AI-based observation with mechanistic models of ecosystem and climate processes provides a powerful opportunity to transform ocean observation from passive data collection to active interpretation and prediction—fostering a more sustainable, climate-resilient, and knowledge-driven ocean science.

This Research Topic aims to advance interdisciplinary approaches that unify computational imaging, artificial intelligence, and marine system modeling to enhance the observation, interpretation, and forecasting of ocean processes. Specifically, it seeks to:

• Develop novel algorithms and systems for underwater imaging, sensor fusion, and 3D reconstruction under realistic oceanic and coastal conditions.

• Bridge physics-based optical modeling with AI-driven learning to produce interpretable, generalizable, and physically consistent imaging frameworks.

• Integrate ecosystem, biogeochemical, and climate–ocean coupling models with AI-enhanced observation to achieve multi-domain understanding of physical–biological feedbacks.

• Advance quantitative modeling of air–sea coupling, climate variability, and biogeochemical fluxes through coupled observation–simulation frameworks.

• Design IoT- and edge-enabled architectures for distributed, energy-efficient, and real-time ocean monitoring.

• Strengthen data reliability, security, and interoperability through decentralized and blockchain-based communication frameworks for marine IoT systems.

• Foster interdisciplinary collaboration among oceanographers, climate scientists, imaging researchers, and AI experts to promote unified approaches to marine system analysis.

• Encourage reproducible, open-science, and sustainable practices supporting long-term ocean observation, ecosystem assessment, and climate prediction.

This Research Topic welcomes manuscripts that contribute to the integration of imaging, computation, and environmental modeling for ocean science. Areas of interest include, but are not limited to:

• Underwater Imaging and Reconstruction: Image restoration, enhancement, and structure-preserving 3D reconstruction in optically complex and turbid environments.

• Physics-Informed and AI-Based Modeling: Hybrid approaches combining optical physics, radiative transfer, and machine learning for physically grounded image interpretation.

• 3D Scene Representation and Visualization: Development of volumetric modeling approaches (e.g., NeRFs, 3DGS, differentiable rendering) for underwater habitats and geomorphology.

• Multi-Sensor and Multi-Modal Observation: Integration of optical, acoustic, LiDAR, hyperspectral, and chemical sensing for multi-scale ocean monitoring.

• Marine System and Climate Modeling: Innovations in ocean circulation, biogeochemical, and climate–ecosystem models; data assimilation linking imaging and predictive simulations.

• Air–Sea Interaction and Climate Variability: Quantitative analyses of coupled feedback mechanisms and their influence on regional and global ocean–atmosphere dynamics.

• IoT and Networked Ocean Sensing: Distributed sensor systems, autonomous vehicles, and cloud–edge collaboration for adaptive and scalable observation.

• Cybersecurity and Data Governance: Blockchain-based verification and secure communication protocols for transparent and trustworthy data sharing.

• Applications and Validation: Case studies in habitat mapping, biodiversity assessment, pollution tracking, ecosystem health monitoring, and predictive modeling of ocean–climate interactions.

• Open Science and Data Resources: Development of benchmark datasets, reproducibility frameworks, and open-source platforms for computational marine research.

This Research Topic envisions a next-generation digital ocean framework that unites AI-enhanced imaging, multi-sensor observation, and predictive marine and climate modeling into a single coherent system. By merging the descriptive capabilities of computational imaging with the predictive strength of ecosystem and climate models, it seeks to establish a seamless continuum between observation, understanding, and forecasting.

Such integration will empower scientists to capture the interconnected dynamics of physical, biological, and climatic systems, from local biogeochemical exchanges to large-scale coupled variability. The ultimate goal is to propel marine science beyond descriptive observation toward predictive, climate-resilient, and sustainability-driven ocean intelligence, supporting both research and informed environmental stewardship.

Are you interested in contributing a manuscript to this Research Topic?

Advanced Imaging and Computational Methods for Ocean Observation