Articles | Volume 22, issue 19
https://doi.org/10.5194/bg-22-5349-2025
https://doi.org/10.5194/bg-22-5349-2025
Research article
 | 
08 Oct 2025
Research article |  | 08 Oct 2025

Optimization of the World Ocean Model of Biogeochemistry and Trophic dynamics (WOMBAT) using surrogate machine learning methods

Pearse J. Buchanan, P. Jyoteeshkumar Reddy, Richard J. Matear, Matthew A. Chamberlain, Tyler Rohr, Dougal Squire, and Elizabeth H. Shadwick

Download

Interactive discussion

Status: closed

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on egusphere-2024-4026', Anonymous Referee #1, 17 Mar 2025
    • AC1: 'Reply on RC1', Pearse Buchanan, 28 Apr 2025
    • AC3: 'Reply on RC1', Pearse Buchanan, 30 Apr 2025
  • RC2: 'Comment on egusphere-2024-4026', Anonymous Referee #2, 25 Mar 2025
    • AC2: 'Reply on RC2', Pearse Buchanan, 28 Apr 2025
    • AC4: 'Reply on RC2', Pearse Buchanan, 30 Apr 2025

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision | EF: Editorial file upload
ED: Reconsider after major revisions (02 May 2025) by Liuqian Yu
AR by Pearse Buchanan on behalf of the Authors (19 May 2025)  Author's response   Author's tracked changes   Manuscript 
ED: Referee Nomination & Report Request started (20 May 2025) by Liuqian Yu
RR by Damien Couespel (19 Jun 2025)
RR by Joost de Vries (22 Jun 2025)
ED: Publish subject to minor revisions (review by editor) (30 Jun 2025) by Liuqian Yu
AR by Pearse Buchanan on behalf of the Authors (16 Jul 2025)  Author's response   Author's tracked changes   Manuscript 
ED: Publish as is (18 Jul 2025) by Liuqian Yu
AR by Pearse Buchanan on behalf of the Authors (22 Jul 2025)  Author's response   Manuscript 
Download
Short summary
We calibrate a new version of the World Ocean Model of Biogeochemistry and Trophic dynamics (WOMBAT-lite) using a surrogate machine learning approach. A Gaussian process surrogate trained on 512 simulations emulated tens of thousands, enabling global sensitivity analysis and Bayesian optimization of 26 parameters. We constrain 13 key parameters, improving fit to 8 datasets (chlorophyll a, air–sea CO₂ fluxes, nutrient limitation), and provide an optimal set for community use.
Share
Altmetrics
Final-revised paper
Preprint