A machine learning‐based dissolved organic carbon climatology

Panaiotis, Thelma ORCID: https://orcid.org/0000-0001-5615-6766; Wilson, Jamie ORCID: https://orcid.org/0000-0001-7509-4791; Cael, BB ORCID: https://orcid.org/0000-0003-1317-5718. 2025 A machine learning‐based dissolved organic carbon climatology. Geophysical Research Letters, 52 (7). 10.1029/2024GL112792

Before downloading, please read NORA policies.

[thumbnail of Geophysical Research Letters - 2025 - Panaïotis - A Machine Learning‐Based Dissolved Organic Carbon Climatology.pdf]

Preview

Text
© 2025. The Author(s).This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
Geophysical Research Letters - 2025 - Panaïotis - A Machine Learning‐Based Dissolved Organic Carbon Climatology.pdf - Published Version
Available under License Creative Commons Attribution 4.0.
Download (1MB) | Preview

Official URL: https://doi.org/10.1029/2024GL112792

Abstract/Summary

Marine dissolved organic carbon (DOC) is a major carbon reservoir influencing climate, but is poorly quantified. The lack of a comprehensive DOC climatology hinders model validation, estimation of the modern DOC inventory, and understanding of DOC's role in the carbon cycle and climate. To address this problem, we used boosted regression trees to relate a compilation of DOC observations to different environmental climatologies, and extrapolated these inferred relationships to the entire ocean to compute annual layer-wise DOC climatologies with uncertainties. Prediction performance was satisfactory, with R2 values within 0.6–0.8 for all layers and prediction error comparable to within-pixel measurement variability. DOC was mainly predicted by dissolved oxygen in the bathypelagic layer, and by nutrients in other layers. We estimate the total oceanic DOC inventory to be around 690 Pg C. Our results exemplify that machine learning is a powerful tool for constructing climatologies from limited observations.

Item Type:	Publication - Article
Digital Object Identifier (DOI):	10.1029/2024GL112792
ISSN:	0094-8276
Additional Keywords:	dissolved organic carbon, climatology, machine learning
Date made live:	28 Apr 2025 12:25 +0 (UTC)
URI:	https://nora.nerc.ac.uk/id/eprint/539338

Item Type:

Publication - Article

Digital Object Identifier (DOI):

10.1029/2024GL112792

ISSN:

0094-8276

Additional Keywords:

dissolved organic carbon, climatology, machine learning

Date made live:

28 Apr 2025 12:25 +0 (UTC)

URI:

https://nora.nerc.ac.uk/id/eprint/539338