the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 28 Jul 2025
Linear trends of temperature, salinity, and oxygen in the North Pacific based on GOBAI-O2 over 2 decades and their controlling factors
Abstract. Oxygen concentrations in the ocean are believed to be declining due to global warming. However, our understanding of its variability remains limited compared to physical parameters such as temperature and salinity, because oxygen is difficult to observe with high spatial and temporal resolution. In this study, we analyzed linear trends in potential temperature, salinity, and dissolved oxygen in the North Pacific over the past two decades (2004–2023), using the GOBAI-O2 data. We then compared oxygen trends with physical parameters to investigate the spatial pattern of linear changes across the region. The oxygen trends derived from GOBAI-O2 were consistent with those observed along hydrographic lines that have been relatively frequently and continuously surveyed by ship-based observations. Although an overall declining trend in dissolved oxygen was evident, localized increases were observed in certain density layers. By examining the associated physical conditions, we found that the spatial heterogeneity of the oxygen trends can be attributed to known oceanographic processes, such as the southward retreat of the oxygen minimum layer and the northward migration of a recently identified front separating the subtropical and subarctic gyres. Our findings highlight the utility of GOBAI-O2 data in linking physical changes to previously unrecognized biological and biogeochemical patterns in the ocean.
- Preprint
(4970 KB) - Metadata XML
-
Supplement
(904 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2025-2746', Anonymous Referee #1, 06 Aug 2025
-
AC1: 'Reply on RC1', Miho Ishizu, 05 Sep 2025
Thank you for your comments. We will consider your comments and reflect them in the revised manuscript.
Citation: https://doi.org/10.5194/egusphere-2025-2746-AC1
-
AC1: 'Reply on RC1', Miho Ishizu, 05 Sep 2025
-
RC2: 'Comment on egusphere-2025-2746', Anonymous Referee #2, 29 Sep 2025
The manuscript examines two decades of linear trends in temperature, salinity, and dissolved oxygen in the North Pacific using the GOBAI-O2 dataset. This timely topic addresses the significant issue of ocean deoxygenation, and the emergence of biogeochemical Argo products offers new avenues for basin-scale analysis. The results align with previous ship-based observations, while also highlighting spatial heterogeneity in oxygen trends attributed to circulation changes like the retreat of the oxygen minimum layer and the northward migration of the subtropical–subarctic front.
The study is potentially valuable, but the manuscript in its present form does not sufficiently establish what is new compared to earlier syntheses. The discussion of previous work, notably Ito et al. (2017, 2024), Sasano et al. (2014), and Kolodziejczyk et al. (2024), is too brief. The paper sometimes reads as a confirmation exercise, whereas it should highlight more clearly what GOBAI-O2 reveals that was inaccessible with earlier datasets.
A more serious concern is the treatment of uncertainty. While fields of uncertainty are presented, their implications for the robustness of the detected trends are not discussed. In several regions the uncertainty is of the same order as the trends themselves. Furthermore, recent work has identified a systematic negative bias in oxygen from air-calibrated BGC-Argo floats relative to shipboard reference profiles, with offsets of roughly –2.7 µmol kg-1 at depth, which propagate into carbonate system estimates such as pCO2 (+3.2 µatm; Bushinsky et al., 2025). Since GOBAI-O2 is trained on these float observations, such biases must be considered explicitly. Without a careful discussion of these uncertainties, the attribution of observed patterns to physical processes remains premature.
On the methodological side, the decision to interpolate GOBAI-O2 fields to 1 m vertical resolution requires justification, as it may artificially smooth variability. The decomposition of oxygen change following equation (2) is interesting but not easy to follow; a clearer derivation or schematic would help. Some of the proposed mechanisms, for example the role of NEC/NECC migration, are plausible but speculative unless supported by independent evidence such as altimetry or reanalyses. Finally, the exclusive focus on linear trends risks obscuring the role of low-frequency climate variability such as the PDO or NPGO.
The language and presentation require significant improvement. Numerous grammatical errors hinder readability, and several figures, especially the large multi-panel maps, are overly dense. A careful edit and improved figure design would considerably enhance clarity.
In conclusion, the dataset is of great interest and the paper has merit, but substantial revision is needed. The authors should better situate their findings relative to previous work, integrate uncertainties and known biases into their interpretation, clarify methodological choices, and moderate or substantiate their physical explanations. With these improvements, the manuscript could become a valuable contribution to our understanding of oxygen variability in the North Pacific.
Citation: https://doi.org/10.5194/egusphere-2025-2746-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 748 | 44 | 19 | 811 | 27 | 15 | 13 |
- HTML: 748
- PDF: 44
- XML: 19
- Total: 811
- Supplement: 27
- BibTeX: 15
- EndNote: 13
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
This paper investigated the linear trend of oxygen concentration in the North Pacific and their relation to the circulation change. Deoxidization is an important issue not only for our scientific community but also for the broader society. Thus, many papers tackled in the issue (ex. Ito et al. 2017, 2024; Kolodziejczyk et al. 2024). However, in the manuscript, there is insufficient discussion to clarify the relationship with those previous studies. Furthermore, uncertainty is also inadequately discussed. Spatial distributions of uncertainty in mapped oxygen and that in linear regression are given, but there is no description how these uncertainties affect the results. Since oxygen concentration in GOBAI-O2 are estimated by using temperature and salinity information, uncertainties need to be fully investigated before discussing physical implications.
Minor comments: Line 364, "areal areas"?