Review of Alonso-Garcia et al., “Sea-level and monsoonal control on the Maldives carbonate platform (Indian Ocean) over the last 1.3 million years, by Jesse Farmer

Alonso-Garcia and colleagues present x-ray fluorescence (XRF)-derived records of carbonate source (Sr/Ca), summer monsoon intensity (Fe/K), and primary productivity (Br) from a sediment record (IODP Site U1467) in the Maldives Inner Sea covering the last 1.3 million years. The authors note a first-order correlation between Sr/Ca and glacial-interglacial cycles indicative of changes in carbonate source (periplatform vs. pelagic), which they attribute to longstanding theories of how carbonate platform productivity is affected by sea level. However, they also note discrepancies between a simple sea-level driver of carbonate production. These discrepancies are highlighted for interglacials MIS 5, 11, and 13, and are attributed to a combination of summer monsoon intensity and Indian Ocean Dipole state. Last, they note long-term changes in carbonate production reaching the Maldives Inner Sea apparently coherent with the Mid-Pleistocene Transition and Mid-Brunhes Event.

Overall, this is an exciting dataset and interpretation that is worthy of publication in Climate of the Past. With that said, I think the manuscript would benefit from major revisions and another round of review. The current draft is a bit scattered; key background information appears in the discussion, and the rationale for evaluating factors other than sea level for carbonate productivity is not entirely clear. I’ve included a few major comments and line-by-line edits up through the discussion. I’d be happy to look at a revised manuscript and provide a more in-depth review of the discussion at that point.

Major comments.

Introduction. As a non-expert in periplatform carbonate sedimentation, I felt that necessary background was missing in the introduction to distinguish Sr-rich carbonate production during highstands vs. Sr-poor carbonate production during lowstands. Later on, the first 1.5 paragraphs of the discussion (L224 – 245) were incredibly useful background context; I’d urge the authors to move this content from the Discussion to the Introduction.

Oxygen isotope sea level proxies and their (quantitative) utility. Whereas deep ocean d18O-based sea level reconstructions usefully indicate the glacial-interglacial character of sea level change, they are not sufficiently precise to be employed for comparing highstand sea levels:

• Typical precision on d18O-based sea level reconstructions is on the order of ± 20 m accounting solely from uncertainty on calculated d18Osw (see Ford and Raymo, 2020 - https://doi.org/10.1130/G46546.1, and note also that the 2sigma uncertainty in the Spratt and Lisiecki, 2016 stack is similar). This magnitude of precision is not sufficient to constrain sea level differences within or between mid/late Pleistocene highstands.
• d18O-based sea level reconstructions also appear to have accuracy problems, possibly more apparent during highstands. For instance, the Elderfield Site 1123 d18O record would suggest MIS 11 sea level of + 40 m, which is hard to take seriously given geological constraints of < 13 m; Raymo & Mitrovica, 2012 (https://doi.org/10.1038/nature10891). Additionally, d18O-based sea level reconstructions appear to greatly overestimate MIS 3 sea level (Dalton et al., 2022 - https://doi.org/10.1016/j.gloplacha.2022.103814).

While such a detailed view of d18O-based sea level reconstructions might seem tangential to the current manuscript, it is necessary because the authors’ motivate their investigation of additional mechanisms using the discrepancy between highstand sea level (from d18O) and Sr/Ca (L259-260). I’m not sure this avenue of motivation holds up given the imprecision on the sea level reconstructions. At the very least, the error on the sea level estimates needs to be presented. I’d also urge the authors to consider something like a crossplot of Sr/Ca vs. sea level (with uncertainty shown) to illustrate the assertion that Sr/Ca and sea level are decoupled – such decoupling is not particularly apparent in Figures 5 and 6 to warrant a long discussion of additional mechanisms.

XRF Br_n. Is this more reflective of productivity or diagenetic alteration? The most notable feature to me is the apparent “burn down” of Br in glacial maxima from MIS 2 to MIS 8, with effectively constant Br_n around small orbital-scale variations before this time. Could the small orbital-scale variations also be related to diagenesis?

Minor/line-by-line comments.

Abstract.  I found this challenging to follow as written and suggest several clarifying edits.

L23. Suggest deleting “factories” and changing to “controlling the production of carbonate platforms”.

L24. Change “a very sensitive region” to “is sensitive”

L24. Delete unnecessary comma after “Sea”

L26-27. This is rather confusing as written; suggest changing to “…(XRF) core-scanning from International Ocean Discovery Program (IODP) Site U1467 in the Maldives Inner Sea, are used in this work…”

L28-30. The verb tense changes how the reader understands this. As written, “are” implies that HSAC deposition is a known modern process (and not necessarily a result from the current study). I think the authors wish to say that “HSAC from neritic settings were deposited in the Maldives Inner Sea…” as this appears to be a new result from the current study.

L29-30. Note that Sr/Ca is already a ratio; it would be simpler to say “increasing Sr/Ca values” than “increasing the values of the Sr/Ca ratio”.

L30-31. Swap the order here with the result first and interpretation after – “In contrast, low Sr/Ca values are observed coincident with sea-level lowstands, suggesting that large areas of the atolls were exposed or unable to grow and the demise in the carbonate production and sediment export” (from the platform?)

L31-33. This sentence confuses me. Don’t you expect higher Sr/Ca during interglacial periods? If so, then “several interglacial periods before and after the Mid-Brunhes event (MBE, ~430 ka) indicate high carbonate production (high Sr/Ca)” would not be surprising. Perhaps this meant to say glacial, or some other dynamic?

L34. It is worth summarizing the evidence to support this statement: “The intensity of the summer monsoon and the Indian Ocean Dipole probably modulated the production at the atolls.” (namely, Fe/K)

Introduction.

L45. Change “patterns is too low” to “forcing is too small”

L51. Note typo here and throughout – McClymont instead of Mcclymont

L53-54. A bit awkward as written; suggest changing to something like “…when it has been proposed that Antarctic ice volume increased, lowering sea level and allowing for further growth of Northern Hemisphere ice sheets”

L59-62. And in the Southern Ocean e.g. Saavedra-Pellitero et al. (2017), https://doi.org/10.1016/j.quascirev.2016.12.020

L64. Remove “very” (unnecessary modifier)

L69-76. This paragraph is missing the basic background for why high sea level results in HSAC. This would be the place to include the further background that is in the discussion (see major comment).

L77-100. These two paragraphs, focused on what climate states lead to enhanced mixing of the Maldives Inner Sea, are missing an important link to how enhanced mixing would be related to HSAC/productivity in the Inner Sea.

L114. Suggest changing “can be” to “appear”

Geological setting.

L126. “…and in the fossil” (as changes not necessary)

Materials and Methods.

L145-150. Suggest moving this short paragraph up to Geological Setting (to start paragraph on L129). This all appears to be previous work, so unless the splice or age model were updated, it can go in Section 2.

L165. Remove “very” – also, can you comment more on the veracity of the correlation between methodologies? For instance R² for the elements used in this study (Fe, K, Br, and Sr)?

L169. To my knowledge, natural logarithm is ln, not Ln.

L180-181. Sentence is missing a verb

Results.

L197-204. Note that comparisons to sea level and monsoon proxies are Discussion and should be left for Section 5.

Comments to the manuscript submitted by Alonso-Garcia et al.

The manuscript submitted by Alonso-Garcia and coauthors presents a very interesting record of climate variability in the Inner Sea of the Maldives archipelago along the last 1.3 My. The study is mainly based on a detailed analysis of carbonate production in the carbonate platform and export to the inner sea, which is mainly controlled by sea level and regional climate variability linked to the Indian monsoons. The manuscript is well organized and well written and deserves publication in Climate of the Past.

Below I present some suggestions that can be considered by the authors in their interpretation.

I suggest that the authors consider and discuss in the text the possibility that part of the siliciclastic particles come from river plumes. The Ganges and Brahmaputra river plumes reach today southern India to the north of Sri Lanka. These rivers generate extensive river plumes that extend along the Eastern coast of India more than 2000 km. Fournier et al., 2018 Modulation of the Ganges-Brahmaputra River Plume by the Indian Ocean Dipole and Eddies Inferred From Satellite Observations.             A proxy such as Ti/Al or Zr/Al could help in this discussion.

It is clear that the variability of Sr/Ca is opposite in phase to that of Fe normalized. I suggest the authors analyze the relationship between iron contents and carbonate, by analyzing the Fe/Ca and Fe/Sr or better the Fe/(Ca+Sr) ratio. It would be important to compare the relationship between Fe/(Ca+Sr) and Fe normalized. If these two records follow the same trend, then the record of Fe normalized probably reflects the mineralogical fraction that is not calcite or aragonite and therefore the balance between carbonate and eolian supply. Because there are two main sources of carbonate supply, aragonite coming from the inner sea and the pelagic calcite source, and considering that the pelagic source is less variable, the periods with less carbonate (aragonitic) supply from the inner sea would result in higher Fe concentration in the sediments, but this does not necessarily reflect higher eolian supply, instead it could be lower aragonitic supply from the shelf. To solve this question a proxy for wind intensity could help, assuming that more intense winds can carry more eolian particles. For example, the authors can use a proxy such as Ti/Al or Ti/K, or Zr+Ti/Al+K that could reflect more the intensity of the winds and its ability to transport coarser particles to the sea. The advantage of using a proxy like this is that it is not influenced by the “closed sum” effect between carbonate and siliciclastic particles.

To better understand this, it would be worth calculating Iron accumulation rates and Ca and Sr accumulation rates to have an idea about the relative contribution of the different elements.

I agree that the Fe/K ratio reflects the balance between the chemical and physical weathering and that Fe should increase during episodes of intense summer monsoon, but the supply to the Maldives inner sea must have taken place during winter monsoon that carries eolian dust from the continent to the sea. Summer monsoon winds normally flow from the ocean to the continent and do not transport dust.

I agree with the authors that the strong calcification event during MIS 11 is a remarkable event in the Maldives, which is also observed in the marine microplankton, and that this high surface calcification caused the Mid Brunhes dissolution event in the deep sea. This is a very important record for carbonate platforms and surface water calcification.