Crafting the Future: Machine learning for  ocean forecasting

Heimbach, Patrick; O'Donncha, Fearghal; Smith, Timothy A.; Garcia-Valdecasas, Jose Maria; Arnaud, Alain; Wan, Liying

doi:https://doi.org/10.5194/sp-5-opsr-22-2025

Articles | Volume 5-opsr

https://doi.org/10.5194/sp-5-opsr-22-2025

© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.

Report:

Ocean prediction: present status and state of the art (OPSR)

https://doi.org/10.5194/sp-5-opsr-22-2025

© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 5-opsr

02 Jun 2025

| OPSR | Chapter 8.3

| 02 Jun 2025 | OPSR | Chapter 8.3

Crafting the Future: Machine learning for ocean forecasting

Patrick Heimbach, Fearghal O'Donncha, Timothy A. Smith, Jose Maria Garcia-Valdecasas, Alain Arnaud, and Liying Wan

Download

Final revised paper (published on 02 Jun 2025)
Preprint (discussion started on 20 Sep 2024)

Interactive discussion

Status: closed

RC1:
'Comment on sp-2024-18', Anonymous Referee #1, 22 Oct 2024

The comment was uploaded in the form of a supplement: https://sp.copernicus.org/preprints/sp-2024-18/sp-2024-18-RC1-supplement.pdf

Citation: https://doi.org/10.5194/sp-2024-18-RC1
- AC1: 'Reply on RC1', Patrick Heimbach, 01 Dec 2024
  
  Please see supplementary document.
  
  Citation: https://doi.org/10.5194/sp-2024-18-AC1
- AC2: 'Reply on RC2', Patrick Heimbach, 01 Dec 2024
  
  Please see supplementary file.
  
  Citation: https://doi.org/10.5194/sp-2024-18-AC2
RC2:
'Comment on sp-2024-18', Anonymous Referee #2, 23 Oct 2024

This is an interesting review of the current status of Machine Learning for Ocean Forecasting, especially for people from outside the subject domain.
The main concern I have is that level of detail of the discussion is rather uneven. For example, the discussion of Sec. 3 "Enhancing data assimilation with ML algorithms" seems just a placeholder for further development. The alternative is that not a lot of activity has been going in the field, in which case this should be stated.
Other comments are posted in the attached annotated version of the manuscript.
Thanks

Citation: https://doi.org/10.5194/sp-2024-18-RC2
- AC2: 'Reply on RC2', Patrick Heimbach, 01 Dec 2024
  
  Please see supplementary file.
  
  Citation: https://doi.org/10.5194/sp-2024-18-AC2
- AC1: 'Reply on RC1', Patrick Heimbach, 01 Dec 2024
  
  Please see supplementary document.
  
  Citation: https://doi.org/10.5194/sp-2024-18-AC1

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision | EF: Editorial file upload

ED: Publish subject to minor revisions (review by editor) (06 Dec 2024) by Swadhin Behera

AR by Patrick Heimbach on behalf of the Authors (09 Dec 2024) Author's response Author's tracked changes Manuscript

ED: Publish as is (18 Dec 2024) by Swadhin Behera

AR by Patrick Heimbach on behalf of the Authors (18 Dec 2024) Author's response Manuscript

Short summary

Operational ocean prediction relies on computationally expensive numerical models and complex workflows, known as data assimilation, in which models are fit to observations to produce optimal initial conditions for prediction. Machine learning has the potential to vastly accelerate ocean prediction, improve uncertainty quantification through massive surrogate model-based ensembles, and render simulations more accurate through better model calibration. We review the developments and challenges.