ORIGINAL RESEARCH article
Front. Mar. Sci.
Sec. Physical Oceanography
A data-driven approach to mesoscale ocean forecasting
Provisionally accepted
Peter R. Oke*
Tatiana Rykova- CSIRO Environment, Hobart, Australia
Select one of your emails
You have multiple emails registered with Frontiers:
Notify me on publication
Please enter your email address:
If you already have an account, please login
You don't have a Frontiers account ? You can register here
Accurate ocean forecasting is essential for many marine industries, including oil and gas, search and rescue, and Defence. Traditional forecasting systems typically produce analyses that are not dynamically consistent - leading to initialisation shock that degrades forecasts. These systems are computationally intensive and generate vast amounts of data, making it difficult for end users to interpret and exploit. Here, we develop a data-driven alternative using analog forecasting. We use along-track sea-level anomaly observations to identify past ocean states that most closely match present conditions in a large archive of model simulations. These historical cases serve as analogs to the present state. The subsequent evolution of each analog is then assembled into an ensemble forecast. We generate 15-day sea-level anomaly forecasts for twelve 5x5 degrees regions around Australia and demonstrate that our system outperforms traditional operational forecasts in 40-60% of cases, performs equally well (no statistical difference) in about 30% of cases, and is outperformed in about 10-25% of cases. By offering a computationally efficient approach to predicting mesoscale ocean circulation, analog forecasting presents a viable and practical alternative or compliment for ocean prediction.
Keywords: Ocean forecasting, eddies, Observations, Analog forecasting, Nearest neighbours, data assimilation
Received: 20 Oct 2025; Accepted: 25 Nov 2025.
Copyright: © 2025 Oke and Rykova. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence: Peter R. Oke
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.