Seasonal cycle in surface seawater isotopic composition

Reverdin, Gilles; Waelbroeck, Claire; Goeme, Lola; Voelker, Antje; Steen-Larsen, Hans-Christian

doi:10.5194/egusphere-2026-851

Preprints

https://doi.org/10.5194/egusphere-2026-851

Preprints

17 Feb 2026

| 17 Feb 2026

Seasonal cycle in surface seawater isotopic composition

Gilles Reverdin, Claire Waelbroeck, Lola Goeme, Antje Voelker, and Hans-Christian Steen-Larsen

Abstract. Data of stable isotopologs of water in the quality-controlled CISE-LOCEAN database provide an unprecedented sampling of the surface Atlantic Ocean and the southwest Indian Ocean. This investigation of δ¹⁸O, δ²H and deuterium excess (d-excess) in surface sea water indicates large variability of the regional regression lines with salinity as a function of latitude for δ¹⁸O and δ²H, with weakest slopes near equator, but also to a lesser extent with weaker slopes in the western Atlantic compared with the eastern Atlantic at low and mid-latitudes. These regressions often explain a large share of the variability in the data (40 to 70 %), except for d-excess at the higher latitudes and δ¹⁸O, δ²H near the equator. The data are sufficient to extract a seasonal variability in these relationships and an average seasonal cycle in some regions of the North Atlantic subpolar gyre with signals probably related to changes in the input of heavier sea ice melt and depleted water of Arctic origin. A seasonal cycle is also found in the central equatorial Atlantic with the strong contrast between the vicinity of the equator and the region to its north, associated with the seasonal shift of the precipitation belt and with the seasonality of the equatorial upwelling. In those regions there is also some evidence in seasonal variability of the d-excess that is not currently explained. Elsewhere, the seasonal signal usually does not come out of the noise. A 7–8 years long time series of nearly monthly surface sampling at the BATS site near Bermuda (31°50’ N, 64°10’ W) also indicates a rather well-defined seasonal cycle of SSS, δ¹⁸O, δ²H and deuterium excess. There is also some indication of interannual variability, which might not be closely related to salinity.

Received: 12 Feb 2026 – Discussion started: 17 Feb 2026

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Gilles Reverdin, Claire Waelbroeck, Lola Goeme, Antje Voelker, and Hans-Christian Steen-Larsen

Status: final response (author comments only)

RC1:
'Comment on egusphere-2026-851', Anonymous Referee #1, 22 Mar 2026
General comments:
This study, titled “Seasonal cycle in surface seawater isotopic composition” by Reverdin et al., utilizes the valuable LOCEAN database to investigate surface seawater isotopes (d¹⁸O and dD) and their regional relationships with salinity. While the data itself has been previously published (Reverdin+2022, 2025), this deeper analysis of the dataset is of high scientific value. However, the current presentation of the results makes the manuscript difficult to follow, thereby limiting the immediate impact of the findings. Some descriptive explanations of the results are not shown with plots, inhibiting a full understanding of the results. Furthermore, although the authors subdivided the study area into 20 regions (Fig. 1 and Table 1), these subregion numbers are not consistently referenced in the text. To make the manuscript suitable for publication, I believe the following points should be addressed to enhance its clarity.

Major Comments:
While the Introduction provides a good overview of previous research, the specific aims of this study are somewhat vague. To guide the reader without stress, the authors should explicitly state the study’s goal; for instance, add a clear statement such as: “This study aims to...”

LOCEAN Database (Data Section): The LOCEAN database (Reverdin et al., 2022, https://doi.org/10.5194/essd-14-2721-2022) is a relatively recent and specific dataset. Some readers need fundamental information on this dataset. Please briefly introduce the LOCEAN database in Data Section, by citing relevant references and summarizing what was previously evaluated to clarify the aim of this current paper (Reverdin et al., 2025).

While the authors define 20 subregions (Table 1), the text uses geographical names (e.g., “south-east of Greenland”) without providing the corresponding subregion numbers. This makes it very difficult to cross-reference the descriptions with the tables and figures. Please consistently identify subregions by their ID numbers (e.g., “NW NASPG [Region X]”) throughout the manuscript. additionally, these subregion numbers should be clearly annotated on the map in Figure 1.

The current plots do not indicate the data density for each regional dataset. Please provide the number of observations (N) for each monthly or season data point (e.g., Fig 3). This is essential for interpreting the differences in the magnitudes of SD and SE (e.g., Fig. 2 vs. Fig. A1). Furthermore, several descriptions (e.g., L. 268-271 and the “unexpected regressions” in L. 474-475) lack corresponding plots. Without a plot showing the relationship (e.g., salinity vs. d18O), it is difficult for the reader to assess the “unexpected” nature of these results.

Minor Comments
L21:Replace “CISE-LOCEAN” with “the isotopic platform facility at LOCEAN (hereafter “CISE-LOCEAN”)”.

L23:For clarity: “This investigation… for each of the 20 subregions indicates…”

L176-178:Phrasing check: “frozen... before been analyzed... before been refrozen.” This appears to be a typo.

L249, 251 (Salinity Climatology):Please specify the source and definition of the term “salinity climatology” used in your analysis. It is unclear whether this refers to WOA or another gridded dataset. If you compared measured bottle salinity with CCI v4.4, as suggested in the Acknowledgements, this should be explicitly described in the Methods or Results instead of only being mentioned in the Acknowledgements.

L285 & 288:please ensure consistent use of subregion names and IDs (e.g., “NW NASPG” rather than “south-east of Greenland”). As noted above, Fig. 1 should be updated with these ID annotations to help the reader follow the regional descriptions.

L309: .. larger winter-spring salinity -> higher winter-spring salinity

L333: uncertainties... are larger than between them -> .. are larger than the differences between these seasonal values.

L376-378, 448-450, 503:Refer to the corresponding subregion numbers.

L397-405: Please provide the geographic coordinate ranges for the merged subregions.

Fig. 1:The primary purpose of these two plots is currently unclear. Please provide a more detailed explanation in the caption to guide the reader. For example, the caption could state: “The LOCEAN database comprises xxx% CRDS measurements; based on this data density, the study area was subdivided into 20 subregions.” Additionally, I note that some areas containing data were not included in the subregion analysis. What was the rationale for this? Please explicitly define the criteria used to determine and classify these subregions.

Fig. 2: Please ensure that the subregions 5N–12N and 5S–5N are clearly defined and consistent across Fig. 1 and Table 1. Additionally, the text following “(--- deviations)” appears to be truncated or cut off; if so, please provide the missing texts.

Fig. 4:The color palette should be changed to a colorblind-friendly version (e.g., viridis or plasma) to comply with modern publication standards.
Citation: https://doi.org/10.5194/egusphere-2026-851-RC1
RC2:
'Comment on egusphere-2026-851', Anonymous Referee #2, 27 May 2026
Summary Comments
This study presents a detailed characterization of seasonal variability in surface seawater isotopic composition (δ18O, δ2H, and d-excess) and its relationship to sea surface salinity from the expanded CISE-LOCEAN database across the Atlantic and Southern Indian oceans. This is an important contribution, as it discusses this database’s ability to resolve regional relationships between water isotopes and salinity, as well as their seasonality. These relationships are foundational to the interpretation of many marine proxy records in the region, as such this study has implications for paleoclimate proxy studies. Furthermore, the results presented in this study could provide valuable benchmarks for isotope-enabled models. However, I have major concerns about the structure, scientific clarity, and legibility of the manuscript as it is presented currently. I have provided substantive revisions to improve this manuscript, with the goal of strengthening the scientific framing and to ensure that the observational uncertainties and sampling limitations are clearly communicated.

Major Comments
The introduction of this manuscript would benefit from more discussion of the scientific motivation and border significance of the constraints provided on seasonal variability in water isotopes. More context on the climatic insights that water isotopes and their relationships to salinity can provide, and how this may vary regionally, would improve the framing of the introduction. The authors briefly reference the application of seawater isotopes to paleoclimate proxy calibrations, however, I think this deserves more attention and is a very important application of this work. The authors could draw on a broader base of literature and paleoclimate studies from the region which use modern studies of seawater isotopes to make interpretations about past hydroclimate to ground this work. Similarly, the application of this study for isotope-enabled model validation is briefly mentioned, however, I think more discussion of these tools and the importance of this study for model development and benchmarking would make this introduction more effective. Overall, a more clear framing of what new insights the CISE-LOCEAN database brings and what kind of scientific questions are being addressed would greatly benefit this manuscript.

The results section would greatly benefit from more sub-headings, and a more concise presentation.

The figures in this manuscript would benefit from a clearer depiction of spatial sampling and coverage. Some representation of the temporal coverage of the database would be helpful. I think a figure showing the number of measurements in each domain through time would be helpful. Walter et al. (2023) figure 1 is a good example of this kind of figure: https://essd.copernicus.org/articles/15/2081/2023/. Atwood et al. (2026) figure 4 is also a good example of a figure showing temporal resolution.

In figure 1, it would be helpful to indicate on the figure which domain is the central equatorial Atlantic, North Atlantic subpolar gyre, and eastern North Atlantic regions which are referenced in later text and figures. In this same figure, it would be helpful to indicate the number of data points in each domain. Does each point represent one measurement? Or are there some sites with multiple measurements being used? If so, this should be indicated.

The limitations of spatial coverage are mentioned in reference to some of these figures; however, this topic should be treated more, as the analysis of the regional coverage of this database is an important finding which will be useful for future work. To what degree do these sampling limitations impact climatic interpretations and your ability to constrain the seasonal cycle? Moreover, you should provide the number of measurements being used to define the seasonal cycle and the regression slopes for each region.

The discussion of seasonal variations in the results section should be streamlined both with regard to the figure showing the raw seasonal cycles, which have vastly overlapping error bars, as well as seasonal variations in the slope/intercepts. These results take up too much room in the paper when there is not a robust result on seasonal variability from this database. This database seems to be a poor fit for the question of seasonal variability in water isotopes and their relationship with salinity given the sparse data available. Consider streamlining or putting these figures (especially figures 2 and 3) in the supplemental information.

The progression through topics in the discussion section should follow the order of presentation in the results section. Do not present results materials in discussion (eg. lines 518-528) and figures should be presented in the results section (e.g. Figure 6 and 7).

The BATS timeseries should not be presented in Appendix B with results presented in the discussion. Remove figure from discussion.

The discussion of the physical mechanisms that drive variability in the relationship between water isotopes and salinity is underdeveloped as is. There is some discussion about the isotopic signature of upwelling and horizontal advection in surface water masses, however, I think this could be expanded more. This is a significant scientific question, especially in regards to the use of paleoclimate proxies in the region to understand past hydrological changes. The authors should expand their literature base to include some context on the physical oceanography of the literature which can potentially drive the observed variability in these regressions. This discussion would also highlight the value of these datasets to the broader community.

Minor Comments
L45: It would be helpful to provide context to what the scientific goals of programs such as GEOTRACES and GO-SHIP are and to establish why this classification is important.

L66-67: “...as well as through the water column in order to provide a climatology to which isotope-enabled models can be referred to.” Are you referring to benchmarking efforts for isotope-enabled models? Please revise for clarity.

L70: Please clarify the regional relationship you are referring to.

L80-82: What potential biases in the data are you referring to? The database contains information on a number of potential biases so please specify. Atwood et al (2026) does not evaluate interannual or decadal variability across the entire dataset, as there are not enough time series with the temporal coverage to do so. Please clarify what you mean by “can be modelled”.

L87-88: Replace “surface seasonal variability” with “surface water isotope seasonal variability” for clarity.

L110-112: The CoralHydro2k seawater database has been published (Atwood et al., 2026): https://essd.copernicus.org/articles/18/1921/2026/essd-18-1921-2026.html

L191-194: Are all measurements open ocean? If so, please state that.

L192: Please specify the ranges used.

L196-198: Are you indicating that you are sampling so the domains are representative of identifiable oceanographic domains? Please rephrase for clarity.

L249: Please specify what climatology you are referring to.

L264-265: It would be helpful to report correlations between d-excess and SSS.

Figure 3: This figure would benefit from some representation of the number of data points the slopes are being calculated from in addition to the average plotted as a point.

L314: Please stay consistent with how you report the season (e.g. JFM versus 7-9).

L331-333: I believe this is a typo, please include “difference” in the sentence if applicable: “...are larger than between them.”

L334-335: By “individual data” do you mean 3 individual data points or 3 individual locations with data and calculated slopes and intercepts? 3 data points seems quite small to calculate a regression slope, so if that is the case it would be worth discussing associated uncertainties in the slope in more detail.

L512: This is not a stand-alone topic sentence, perhaps consider deleting or combining with previous paragraph

Table 1: Please include the number of samples used in the regression.

Figure 5: Please make the y-axis of the plot large enough so we can see all the points and uncertainties.

Table 2: Please include the number of samples used in each seasonal regression.

L554: I think the “?” in this sentence is a typo. Additionally please clarify what a good timeseries means.

L798: Please specify which climatology you are referring to.
Citation: https://doi.org/10.5194/egusphere-2026-851-RC2

Gilles Reverdin, Claire Waelbroeck, Lola Goeme, Antje Voelker, and Hans-Christian Steen-Larsen

Viewed

Total article views: 1,100 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
738	294	68	1,100	104	90

HTML: 738
PDF: 294
XML: 68
Total: 1,100
BibTeX: 104
EndNote: 90

Views and downloads (calculated since 17 Feb 2026)

Month	HTML	PDF	XML	Total
Feb 2026	280	120	38	438
Mar 2026	324	92	24	440
Apr 2026	75	41	2	118
May 2026	54	39	4	97
Jun 2026	5	2	0	7

Cumulative views and downloads (calculated since 17 Feb 2026)

Month	HTML	PDF	XML	Total
Feb 2026	280	120	38	438
Mar 2026	324	92	24	440
Apr 2026	75	41	2	118
May 2026	54	39	4	97
Jun 2026	5	2	0	7

Viewed (geographical distribution)

Total article views: 1,094 (including HTML, PDF, and XML) Thereof 1,094 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 08 Jun 2026

Short summary

The distributions of water isotopic properties (δ¹⁸O, δ²H, deuterium excess) as a function of salinity are studied in the surface Atlantic Ocean and southwest Indian Ocean using the CISE-LOCEAN data base. Large spatial gradients are found as well as a seasonal variability in the North Atlantic sub-polar gyre and in the equatorial Atlantic. Spatially-averaged seasonal cycles are established in these regions, as well as at the BATS site (31°50’ N, 64°10’ W), with weak interannual variability.


Total:	0
HTML:	0
PDF:	0
XML:	0