Global monthly ocean dissolved oxygen (1960–2023) reconstructed to 5902&thinsp;m with BLENDR, a Bayesian-optimized ensemble learning framework

Han, Mingyu; Xing, Xiaogang; Zhou, Yuntao

doi:10.5194/essd-18-3757-2026

Articles | Volume 18, issue 6

https://doi.org/10.5194/essd-18-3757-2026

© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/essd-18-3757-2026

© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 18, issue 6

Data description article

|

03 Jun 2026

Data description article |

| 03 Jun 2026

Global monthly ocean dissolved oxygen (1960–2023) reconstructed to 5902 m with BLENDR, a Bayesian-optimized ensemble learning framework

Mingyu Han, Xiaogang Xing, and Yuntao Zhou

Supplement

https://doi.org/10.5194/essd-18-3757-2026-supplement

Data sets

Global Monthly Dissolved Oxygen Reconstruction via Bayesian Ensemble Machine Learning Mingyu Han and Yuntao Zhou https://doi.org/10.5281/zenodo.19705526

Download

Article (9442 KB)
Full-text XML

Short summary

We combined ship and float measurements with machine learning to reconstruct monthly dissolved oxygen in the global ocean from 1960 to 2023, from the surface to 5902 m. The results reveal a persistent loss of oxygen, strongest below the surface and in major low-oxygen zones, with recent acceleration in several ocean regions. This open dataset supports climate research and assessments of risks for marine ecosystems.