Lessons Learned in Establishing an Environmental Digital Twin

Phytoplankton blooms are an important component of marine ecosystems and present both positive and negative impacts on ecosystem services and the blue economy, yet their triggers and effects throughout the water column remain poorly understood and difficult to predict. The SyncED-Ocean project deployed a digital twin for a two-month period during summer 2024. This digital twin integrated Earth Observation (EO) data with in-situ measurements from autonomous, crewed, and fixed ocean platforms. A modular data analysis toolbox was developed to cross-calibrate EO and in-situ data, ensuring quality-controlled assimilation into a virtual twin model at daily intervals. Forecasts and nowcasts generated by the virtual twin model were analysed by AI-driven applications to optimise future data collection by generating and transmitting mission plans to autonomous underwater vehicles in real-time. This closed-loop digital twin system provided optimised high spatial and temporal resolution data suitable for detecting phytoplankton blooms and related deoxygenation events as they emerge. Key lessons from the deployment highlight the need for rigorous engineering in data integration and management, ensuring resilience against erroneous data. Trade-offs between data timeliness and quality necessitated redesigning workflows to meet strict operational timeframes while maintaining data integrity. Additionally, despite automation, human-on-the-loop monitoring remained essential for overseeing data pipelines and machine learning decision-making. These insights provide valuable guidance for future environmental digital twins, enhancing scientific understanding and supporting sustainable marine management. The SyncED-Ocean experience underscores the transformative potential of AI, marine autonomy, and digital twins in improving marine ecosystem monitoring, management and economic resilience.