the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 24 Apr 2026
Assessing the potential of Individual Foraminifera Analyses (δ18O) to reconstruct variability, seasonality and extremes in the tropical Indian Ocean
Abstract. The Indian Ocean plays a critical role in the global climate system by regulating heat and moisture transport, influencing major fluxes that drive atmospheric circulation. Individual foraminiferal analyses (IFA) of the stable oxygen isotope composition (δ18O) of their test provides a powerful approach to reconstruct past climate variability beyond changes in mean conditions, with the potential to capture seasonal to interannual climatic extremes associated with phenomena such as the Indian monsoon and the Indian Ocean Dipole (IOD). However, field-based calibrations of this proxy remain sparse in the Indian Ocean, especially in the western part of the basin, limiting our understanding of how IFA records regional hydrographic signals. In this study, we combine a forward-modelling approach with IFA in Indian ocean core-top sediment samples to evaluate how δ18O variability responds to changes in temperature and salinity linked to seasonal and interannual variability. Using ocean reanalysis data, we simulate surface and thermocline conditions at several sites representative of different oceanographic provinces across the Indian Ocean and generate synthetic core-top IFA datasets through random sampling. We evaluate the sensitivity of results by artificially amplifying or reducing seasonal and interannual climate variability. Forward-modelled data are then compared with IFA data from core-tops at four key sites across the Indian Ocean. Our results show that surface-ocean individual foraminiferal δ18O changes in the western and central Indian Ocean are mainly controlled by temperature seasonality related to monsoon dynamics, whereas in the eastern Indian Ocean, seasonality is less pronounced and interannual variability is more dominant. On the other hand, thermocline δ18O variability for all sites is primarily associated with interannual variability and temperature changes, highlighting the potential of thermocline species to record interannual temperature variability. Forward-modelled IFA closely match real datasets from Late Holocene core-tops for surface and thermocline depths. This framework provides a basis for interpreting IFA in the Indian Ocean in terms of underlying climate processes and offers perspectives for reconstructing past seasonal and interannual climate variability in this important region.
- Preprint
(13489 KB) - Metadata XML
-
Supplement
(2369 KB) - BibTeX
- EndNote
Status: open (until 19 Jun 2026)
- CC1: 'Comment on egusphere-2026-2343', Giacomo Medici, 22 May 2026 reply
-
RC1: 'Comment on egusphere-2026-2343', Anonymous Referee #1, 25 May 2026
reply
The manuscript applies individual foraminifera analysis (IFA) δ18O across four oceanographically distinct sites in the tropical Indian Ocean to evaluate variability resulting from seasonality and extreme climate events. While IFA is a well-established concept, the authors extend its application to the tropical Indian Ocean by combining new measurements with previously published records. The forward modelling approach uses ORAS5 reanalysis to simulate δ18O of calcite (δ18Oc) at surface (0.5m) and thermocline (97m) depths. The forward-modelled distributions are compared with core-top IFA δ18Oc of G.ruber and N.dutertrei, except for the lack of thermocline species data from the eastern tropical Indian Ocean site. Climate sensitivity is assessed by introducing ±75% perturbations in ORAS5 temperature and salinity, modifying only seasonal amplitude at the surface and both seasonal amplitude and temporal frequency at the thermocline
The scientific premise is relevant and the manuscript falls within the scope of CP. The core-top IFA δ18Oc data are sufficient to support the broad interpretations and conclusions. However, the forward modelling framework rests on several simplifying assumptions that limit the realism of the outputs. These assumptions need to be either improved or more rigorously justified. Significant revisions to the methodology are therefore required before the manuscript can be considered for publication.
General comments:
1) The authors used ORAS5 reanalysis data for forward modelling δ18Oc at surface (0.5 m) and thermocline (97 m) depths. Since ORAS5 forms the foundation of the results and discussion, the authors should first evaluate how well ORAS5 reproduces observed conditions near the proxy sites. Comparisons with instrumental datasets (e.g., satellite-derived SST and SSS, buoy-based temperature and salinity observations) need to be presented (in supplementary). Additionally, a comparison with other reanalysis products such as HadISST (surface), GODAS (surface and thermocline), and SODA3 would strengthen confidence in the reanalysis-based approach and help assess the sensitivity of results to the choice of reanalysis product.
2) Assigning a single fixed depth for the surface (0.5m) and thermocline (97m) uniformly across all sites and through time is a significant simplification. Thermocline depth is neither spatially constant (Fig. 2A) nor temporally stable, it varies with monsoon phase, extreme climate events, and regional oceanographic dynamics. The authors should consider integrating temperature and salinity data over the mixed layer and thermocline depth ranges, accounting for their spatial and temporal variability, to produce a more physically realistic forward-modelled δ¹⁸Oc distribution.
3) The authors artificially introduced ±75% variability in temperature and salinity to simulate extreme climate events but do not adequately justify this choice beyond citing Thirumalai and Clemens (2020), whose study focuses on the northern Bay of Bengal, a region highly sensitive to salinity changes due to large freshwater fluxes and therefore oceanographically distinct from the four sites used here. The authors first need to quantify the realistic upper and lower bounds of ENSO- and IOD-related variability at each of the four oceanographically distinct sites before applying a uniform ±75% perturbation. For example, the annual SST range at site MD96-2060 (western tropical Indian Ocean) is approximately 3.5°C (Fig. 1a). Increasing this variability by 75% would yield a range of ~6.1°C. It is therefore necessary to demonstrate whether such variability is actually observed during ENSO or IOD events at this site. Without such evidence, the interpretation appears overstated.
4) The authors state (Lines 207–211 and 330–332) that only seasonal amplitude is altered at the surface, whereas both amplitude and frequency of interannual events are modified at thermocline depths. What is the rationale for not introducing interannual frequency changes at the surface as well? This methodological choice requires clarification and justification.
5) The use of site names throughout the discussion is difficult to follow. It would improve readability if the authors consistently referred to the sites using regional identifiers such as western Arabian Sea, western tropical Indian Ocean, central tropical Indian Ocean, and eastern tropical Indian Ocean. Otherwise, readers must repeatedly refer back to Figure 1 to identify the site locations.
6) The authors need to report nearby seasonal flux, and habitat-depth information for G.ruber and N.dutertrei from the FORCIS database (https://doi.org/10.1038/s41597-023-02264-2) in the Supplementary Material. Since the study interprets the SD of d18O G.ruber and N.dutertrei as reflecting climate variability, forward modelling using realistic habitat depth and seasonal flux data would be more appropriate than assuming constant carbonate flux at fixed depths (0.5 m and 97 m). Incorporating these parameters would substantially improve the paleoclimate relevance of the modelling framework.
7) At present, the manuscript relies on forward modelling mean and variability of δ18Oc at fixed depths (0.5 m and 97 m), assuming constant monthly carbonate flux, a single δ18O paleotemperature equation across species, and constant δ18Ow–salinity relationships across all sites. These assumptions move the model outputs away from realistic oceanographic conditions. To make the forward-modelling framework more robust and useful for the paleoclimate community, the following revisions are recommended: a) Define the habitat depth and its variability for G.ruber and N.dutertrei at each site using nearby plankton-net or sediment-trap observations (e.g., FORCIS database). b) Incorporate seasonal flux information from sediment-trap records near the proxy sites (e.g., FORCIS database). c) Use regional and depth-specific δ18Ow–salinity relationships for both surface and thermocline waters, given that the four sites are oceanographically distinct. d) Apply species-specific δ18O paleotemperature equations instead of using the G.bulloides equation of Bemis et al. (1998) for all species. e) Propagate uncertainties through each modelling step and include them in the final outputs. f) Introduce realistic climate perturbations at both surface and thermocline depths based on observed ENSO/IOD variability near the study sites. Such modifications would make the forward-model outputs substantially more realistic and scientifically valuable for the IFA and paleoclimate communities.
Specific comments:
1) Line 47: The authors state “drier northeast monsoon during boreal winter.” However, the northeast monsoon delivers substantial rainfall over southern India, Sri Lanka, and parts of Southeast Asia.
2) Figure 1: The authors used annual range for surface waters and SD for thermocline waters. Please specify the depth range over which the averages were calculated. Since the study objective is related to validating IFA δ18O SD, why not also present SD for surface waters in the figure?
3) Lines 160–161: Please report the mean thermocline depth and its standard deviation for the four sites based on ORAS5 reanalysis.
4) Lines 163–164: Why are species-specific equations not being used? Instead, the study applies the G. bulloides and O. universa equations from Bemis et al. (1998) to simulate δ18Oc for G. ruber and N. dutertrei.
5) Lines 166–167: The authors apply the Singh et al. (2010) δ18Ow–salinity regression for surface waters. However, this regression mainly reflects northern Indian Ocean conditions influenced by strong riverine freshwater input (https://doi.org/10.1016/j.dsr.2010.08.002). Please justify its applicability to the study locations. Please also report the R² value and associated uncertainty.
6) Lines 169–171: For subsurface δ18Ow–salinity relationships, the authors use a regression derived from a global dataset. Please show the locations of the water samples used from LeGrande and Schmidt (2006) along with proxy sites in a supplementary figure. Also provide the R² value and uncertainties in the slope and intercept.
7) Figure 2: The manuscript suddenly introduces World Ocean Atlas 2023 data, whereas ORAS5 is used throughout the forward modelling. Please use ORAS5 consistently for the depth-wise temperature and salinity plots at all four sites. The selection of 97 m as the thermocline depth appears to rely on WOA23 seasonality. However, the manuscript does not demonstrate how SD of salinity and temperature vary interannually with depth. The rationale for selecting 97 m is therefore not properly justified
8) Lines 198–200: This section is confusing. In Section 3.1.1, the authors use Singh et al. (2010) for surface waters and LeGrande and Schmidt (2006) for subsurface δ18Ow–salinity relationships. However, in Section 3.1.2, the manuscript refers to the δ18Ow–salinity relationship of Delaygue et al. (2001). Please clarify which relationship was actually used. In reality, δ18Ow–salinity relationships can vary seasonally. The authors should also evaluate how the SD of forward-modelled δ18Oc changes when different δ18Ow–salinity relationships are applied, and justify the use of a constant relationship.
9) Lines 203–204: The authors state that “Each individual foraminifera is assumed to record the average monthly environmental conditions of its calcification period.” Please provide justification based on the average lifespan of G. ruber and N. dutertrei.
10) Table 1: Please include uncertainties associated with the core-top ages.
11) Figure 3: Please include SD values in panels B and D, and discuss them in the interpretation section.
12) Lines 318–320: The authors state that “At sites NIOP 905 and U1467, thermocline variability on interannual timescales is more muted, and seasonal variations in SST and SSS at thermocline depths are larger (Fig. 3D).” However, site NIOP 905 is influenced by the Somali Current. Does the absence of interannual variability in thermocline temperature and salinity imply that the interannual strength of the Somali Current is nearly constant? Is this interpretation supported by instrumental observations?
13) Figure 4: It would be easier to follow if the panels were arranged longitudinally from west to east. This would reduce the need for readers to repeatedly identify site locations.
14) Line 446: Figure 1c represents the annual range of surface-water δ18Oc, not δ18O SD. The figure citation is therefore incorrect.
15) Line 458: Incorrect citation of Figure 6B.
16) Lines 527–529: This statement is misleading. Since interannual perturbations were not introduced at the surface, it is unclear how the authors conclude that interannual variability is observed in the surface simulations.
17) Lines 547–548: This statement appears overstated. I would describe the agreement between forward-modelled δ18O and core-top IFA δ18O as only partial, with approximately half of the sites showing agreement.
Technical comment:
1) Line 74: “Northeastern” appears incorrect; perhaps “northwestern” is intended.
Citation: https://doi.org/10.5194/egusphere-2026-2343-RC1
Data sets
Individual foraminiferal δ¹⁸O analyses (IFA) from Late Holocene core-tops sediments from Indian Ocean Y. Lichterfeld https://www.seanoe.org/data/01023/113513/
Model code and software
IFA Forward Modelling Toolbox Y. Lichterfeld https://github.com/yoyolich/Lichterfeld2026-IFA-forward-model_2026
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 298 | 343 | 27 | 668 | 50 | 26 | 28 |
- HTML: 298
- PDF: 343
- XML: 27
- Total: 668
- Supplement: 50
- BibTeX: 26
- EndNote: 28
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
General comments
Very good research on paleoclimate that proposes an isotopic approach. Follow my guidelines to improve the manuscript.
Specific comments
Line 61. “Traditional δ18O paleoclimate analysis methods focus on reconstructing average climate conditions”. Statement which is not backed-up be references. Please, insert general references on traditional δ18O approaches to paleoclimate studies:
- Lécuyer, C., Allemand, P. 1999. Modelling of the oxygen isotope evolution of seawater: implications for the climate interpretation of the δ18O of marine sediments. Geochimica et Cosmochimica Acta, 63(3-4), 351-361.
- Medici G, Marianelli D, Cornacchia I, Gori F, Brandano M 2026. Multi-disciplinary approach to paleokarst occurrence in the Eocene–Oligocene succession of the Apulia Carbonate Platform (Salento, Italy). Facies, 72(2), 17.
Line 78. Clarify the general aim or principal research question of your paleoclimate research.
Line 78. They look more specific objectives than the steps of a workflow. I would use the term “specific objectives” at the end of your introduction.
Line 89. “Regional water dynamics”. Define the ranges for the spatial scale of your regional movement of water-masses.
Line 685. Link your sedimentation rate to your specific paleoenvironment / sedimentary environment in this paragraph. If you are clear about that, you could the attract more attention from the audience.
Lines 685-720. I would delete empty lines (e.g., 690).
Figures and tables
There are issues on resolution on several figures. I will be more specific below.
Figure 2. Letters and numbers on the axes of all the three graphs are too small. There is room for improvement.
Figures 3-5. Same scenario for these for these graphs.
Figure 6. Increase the graphic resolution maybe rising the dpi. Some words and numbers are difficult to read.
Figure 7. Do you need to show equations for the correlation lines?
Figure 7. Legend and words/numbers on axes are difficult to read.