@article{chattopadhyay_gray_wu_lowe_he_2024, title={OceanNet: a principled neural operator-based digital twin for regional oceans}, volume={14}, ISSN={["2045-2322"]}, DOI={10.1038/s41598-024-72145-0}, abstractNote={Abstract While data-driven approaches demonstrate great potential in atmospheric modeling and weather forecasting, ocean modeling poses distinct challenges due to complex bathymetry, land, vertical structure, and flow non-linearity. This study introduces OceanNet, a principled neural operator-based digital twin for regional sea-suface height emulation. OceanNet uses a Fourier neural operator and predictor-evaluate-corrector integration scheme to mitigate autoregressive error growth and enhance stability over extended time scales. A spectral regularizer counteracts spectral bias at smaller scales. OceanNet is applied to the northwest Atlantic Ocean western boundary current (the Gulf Stream), focusing on the task of seasonal prediction for Loop Current eddies and the Gulf Stream meander. Trained using historical sea surface height (SSH) data, OceanNet demonstrates competitive forecast skill compared to a state-of-the-art dynamical ocean model forecast, reducing computation by 500,000 times. These accomplishments demonstrate initial steps for physics-inspired deep neural operators as cost-effective alternatives to high-resolution numerical ocean models.}, number={1}, journal={SCIENTIFIC REPORTS}, author={Chattopadhyay, Ashesh and Gray, Michael and Wu, Tianning and Lowe, Anna B. and He, Ruoying}, year={2024}, month={Sep} }
 @article{cocker_bert_torres_shreve_kalb_lee_poimboeuf_fautley_adams_lee_et al._2022, title={Low-Cost, Intelligent Drifter Fleet for Large-Scale, Distributed Ocean Observation}, ISBN={["978-1-6654-6809-1"]}, ISSN={["0197-7385"]}, DOI={10.1109/OCEANS47191.2022.9977209}, abstractNote={We have developed Persistent Environmental Awareness Reporting and Location (PEARL) ocean drifters. PEARL drifters are small, rugged, low-cost, autonomous, environmentally friendly ocean drifters that represent a significant opportunity for high-impact applications in both national security and environmental ecosystem monitoring. Drifters record and report data which is processed by advanced edge analytics before being compressed for satellite transmission to populate a large data repository with sensor data that is combined and analyzed to discover signals of interest in the ocean environment with the goal of increasing distributed maritime awareness. Each drifter is entirely self-contained, powered by solar panels and backup batteries, with a large array of sensors, compute module for onboard data processing, and satellite modem for data reporting. The drifter architecture is flexible and can be customized for a specific purpose. The complete data record is stored locally and processed by the onboard compute module, which runs anomaly detection algorithms that detect nearby activity. Anomalous events, as well as baseline environmental data, are reported to a cloud database using satellite short burst data transmission. Though each independent drifter is a powerful sensing tool, the low unit cost permits large scale deployment. To date thousands of drifters have been deployed over vast areas of the ocean and are reporting data to a remote database where cloud-based analytics algorithms develop global situational awareness and update local edge algorithms on the drifters based on learnings across the full network.}, journal={2022 OCEANS HAMPTON ROADS}, author={Cocker, Eric and Bert, Julie A. and Torres, Francisco and Shreve, Matthew and Kalb, Jamie and Lee, Joseph and Poimboeuf, Michael and Fautley, Paloma and Adams, Samuel and Lee, Joanne and et al.}, year={2022} }