Tropical deoxygenation sites revisited to investigate oxygen and nutrient trends

Stramma, Lothar; Schmidtko, Sunke

doi:https://doi.org/10.5194/os-17-833-2021

Articles | Volume 17, issue 3

https://doi.org/10.5194/os-17-833-2021

© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/os-17-833-2021

© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 17, issue 3

Research article

| Highlight paper

|

01 Jul 2021

Research article | Highlight paper |

| 01 Jul 2021

Tropical deoxygenation sites revisited to investigate oxygen and nutrient trends

Lothar Stramma and Sunke Schmidtko

Download

Final revised paper (published on 01 Jul 2021)
Supplement to the final revised paper
Preprint (discussion started on 28 Dec 2020)

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC1: 'Reviewer Comments', Anonymous Referee #1, 29 Jan 2021
- AC1: 'Reply on RC1', Lothar Stramma, 21 Apr 2021
RC2: 'review to Stramma&Schmidtko paper', Anonymous Referee #2, 03 Mar 2021
- AC2: 'Reply on RC2', Lothar Stramma, 21 Apr 2021

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

AR by Lothar Stramma on behalf of the Authors (21 Apr 2021) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (22 Apr 2021) by Arvind Singh

RR by Anonymous Referee #1 (27 Apr 2021)

RR by Anonymous Referee #3 (05 May 2021)

Suggestions for revision or reasons for rejection

Review of the manuscript, entitled “Oxygen and nutrient trends in the Tropical Oceans” authored by Stramma and Schmidtko.

In this manuscript, the authors have intended to describe trends of oxygen and nutrients using long time series spanning from 1950 to 2018. This work is an extension of the authors' previous paper published in 2008 by including a more recent dataset and incorporating more variables into the analysis.
The manuscript presents an extensive set of dataset and therefore, contribute in enhancing our existing understanding of the said processes. However, as noted in the earlier comments by other reviewers, I have few major concerns and they are listed below:
1. The title of the manuscript says “trend of the Tropical Oceans”. But actually, the analysis is limited to very few locations in the tropics. This is particularly true for the Indian Ocean. I can understand that data limitation is the major problem to extend this study to a broader region, but then the title is grossly misleading.
2. Another major problem with the analysis is that the calculated trends for oxygen and nutrients for most regions are either not statistically significant or the error associated with the trend is more than the trend itself. This indicates that goodness of fit for this trend calculations are very poor (can be seen easily from the plots as well), which means the spread of the data point is too large for a sensible linear fit. This casts a doubt on the reliability of these trends. However, there seems to be a definite decreasing (increasing) trend for the oxygen (nutrients) in the 300-700 layer, but the calculated trends may not be reliable.
3. Authors have interpolated data into standard vertical depths and then averaged over 50-300 m and 300-700 m to calculate the trends for the respective variables.
(a) What are these standard depths?
(b) I believe, the depth ranges are chosen as an analogous to the ocean heat content approach. However, unlike temperature data, oxygen and nutrient data are much more sparse and therefore, may pose a problem while averaging. This is particularly problematic for the upper layer as some part of it is above the thermocline and some part within the thermocline and therefore, experience a strong gradient in oxygen concentration. Now, it may possible that in some years data above the thermocline is available and for other years within the thermocline. This will result in scatter concentration in time series similar to what we see here. Moreover, as the thermocline itself show a large shift due to ocean warming and various climate modes, the averaging across the depths (in case of data coverage is not uniform across the depth and year) will itself lead to a trend. How authors have addressed this issue is not very clear.
(c) For vertical interpolation, a uniform Gaussian weighting method is applied with a uniform cut-off range in space and time across all the boxes and (top and intermediate) layers. Since the variability in the top layer is very different from the bottom layer, different weighting mechanisms may suit better for the different layers.
(d) In order to better appreciate the data inventory and the trends, is it possible to show number of data points available for each of the standard depths for each box? Maybe a plot with y-axis as depths and x-axis as years with the number of data points as colour will serve the purpose.
4. Tables showing the trends are just too many details and may not be very useful to most readers. Is it possible to convert them as barplot with error bar for each variable? This will make life easy for the readers. The detailed tables can be provided as supplementary information for interested readers.
Overall, the trend calculations are not quite robust and also doubtful. There are a large number of caveats present in the data analysis. Therefore, I believe the title and abstract are exaggerated in nature. Authors need to spell these caveats upfront in the abstract so that the readers are not misled while interpreting these results.

Hide

ED: Publish subject to minor revisions (review by editor) (14 May 2021) by Arvind Singh

AR by Lothar Stramma on behalf of the Authors (20 May 2021) Author's response Author's tracked changes Manuscript

ED: Publish as is (20 May 2021) by Arvind Singh

AR by Lothar Stramma on behalf of the Authors (21 May 2021) Author's response Manuscript

Short summary

Six tropical areas in the Pacific, Atlantic and Indian oceans were investigated for trends for the years since 1950 for temperature, salinity, oxygen and nutrients at 50 to 300 m and 300 to 700 m depth. Generally, temperature increases, oxygen decreases and nutrients often increase. Overlain variability on the trends seem to be related to climate modes. Different trends indicate that oxygen and nutrient trends cannot by completely explained by local warming.