General comments

The present contribution traces the evolution of the upwelling system in the NW Indian Ocean (and its atmospheric and oceanic drivers) over the time interval from 15 to 9 million years ago (mid- Miocene) and is based on data raised from sediment cores of ODP Site 722 on Owen Ridge. Counts of calcareous nannofossil taxa, siliceous fragments and planktonic foraminifers are presented in context with data on sedimentology, geochemistry and isotopic composition of the sediment sequence. The new microfossil assemblage data are interpreted as indicators of nutrient conditions in the mixed layer and in the thermocline, and the record is proposed to illustrate interactions between different plankton groups and varying nutrient levels and ratios. The data and interpretations of local developments are then discussed in perspective of regional and global changes in the coupled ocean-atmosphere system, which at that time experienced major reorganisation in response to tectonic processes and opening/closing of ocean gateways.

Novelty of the manuscript is in the statistical investigation of quantitative nannofossil data in 71 samples (for nannofossils and siliceous debris) and planktonic foraminifers in 28 samples and their implications for reconstruction of trophic state, nutrient limitations, and paleoproductivity in the mixed ocean surface and thermocline. The statistical analysis of these data yields four “taphogroups” (plus subgroups) that are proposed to be indicative of specific local surface water conditions, mainly of nutrient concentrations and nutrient ratios, and are employed to track the evolution of conditions at Site 722 through time. This reviewer is not qualified to evaluate whether the approach of using size- and morphology-based traits as expressions of nutrient regimes is standard practise, particularly in light of the stated “weak support for individual clusters reflecting the overall strong similarities in the assemblage composition of the studied samples” (line 212).  The abstract of the Paasche (2010) publication on “Roles of nitrogen and phosphorus in coccolith formation in Emiliania huxleyi" suggests that reactions of this particular modern prymnesiophyte to nutrient limitations are much less straightforward than stated here.

Ancillary proxies, such as XRF data (dust and Mn), carbonate-C and organic-C, delta15N, have previously been published in Bialik et al. ( 2020). Why these are suitable proxies is unclear, because the introduction is incomplete and sections of text appear to be out of sequence and/or truncated in the presentation of proxies used to track wind, upwelling and OMZ (lines 93-101). In many cases the authors use these ancillary proxies to bolster their arguments based on the nannofossil assemblage, which occasionally results in circular reasoning.

As a main conclusion, the authors propose that the atmospheric (SW monsoon) driver and an  incipient oxygen minimum zone existed in the western Arabian Sea, but that a one-million-year lag in productivity and community response to upwelling marks a teleconnection to southern hemisphere intermediate water formation: Only after intermediate waters formed in the Antarctic Convergence imported new nutrients into the thermocline of the northern Arabian Sea did the upwelling elicit a typical high-productivity response in the plankton community. Furthermore, they propose that changing nutrient ratios (caused by variable denitrification in a variable oxygen minimum zone) subsequently dictated changing patterns of planktonic ecosystems at continued high nutrient supply.

The manuscript thus offers interesting observational evidence of previously published modeling results and is an extension of a previous publication of some of the authors. The positive view is dampened somewhat by the (frequently unreflected, sometimes counterfactual and occasionally contradictory) rendition or reiteration of statements that are at least worthy of discussion. This particularly concerns the interactions between atmospheric drivers, water mass distributions and their nutrients, and dynamics of the oxygen minimum zone in the various subchapters of the discussion and the discussion of global implications. These parts of the manuscript definitely need careful scrutiny. 

The manuscript fits the scope of the journal, in particular in a special issue dedicated to Dick Kroon, who has laid grounds for the reconstruction of the monsoonal upwelling in the Arabian Sea. It is reasonably well written, but comparatively long for what the authors have to say, and the list of references is very long indeed. In a revision, text should be edited with care and preferably by a native speaker. Some passages of text on intricacies of placolith morphologies should be omitted. The continual use of comparatives without reference to what it is compared to must be corrected. Figures are ok, but may possibly be condensed by plotting factor scores instead of single species´ abundances.

Detailed comments and questions:

Many of the data and many of the arguments in the discussion echo the Bialik et al. (2020; Paleoceanography and Paleoclimatology) publication: It is necessary to highlight the novelty of the present paper more clearly and succinctly.

A number of conclusions in that publication are taken for granted here, so that some of the questions below are addressed to the precursor publication. The following more or less relevant points came to my mind when reading the paper:

• The observed patterns are attributed to nutrient import from sub-thermocline waters originating in the Antarctic Confluence. What about changes in Indonesian Throughflow that progressively limited water exchange and initiated formation of the Indian Ocean Central Water mass that is the principal source of upwelling water offshore Oman (e.g., Kuhnt et al., 2004; You and Tomczak, 1993; You, 1997)? Not all readers may know what water masses are involved, so that a brief rendition of circulation the modern Indian Ocean that was established near the end of the investigated time interval is in order.
• What about the role of uplift and changes in elevation in central Asia as a driver for the SW (and NE) monsoon inception? The text only refers to latitudinal temperature gradients and emergence of the Arabian Peninsula.
• Is upwelling above Owen Ridge indeed driven by the Findlater Jet, or by wind stress curl? How does emergence of the Arabian Pensinsula (which is discussed as a decisive factor for unclear reasons) influence that?
• Seasonality is an important influence on plankton succession and alternating eutrophic/oligotrophic deposition at the transition from coastal to open-ocean upwelling (such as is the case over Owen Ridge). Is it possible that the entire phytoplankton assemblage between 13 and 10 mya reflects extreme, but variable seasonality, not just at times of a dominant taphogroup 3?
• How did an OMZ form at low productivity and C-flux around 13 Ma?
• Figure 2 and 3 plot abundances of individual taxa that are representative of specific conditions. Would a representation (similar to Fig. 5) of scores for the clusters (or factors) in Figure 4 show a clear pattern of changing conditions in mixed-layer plankton, monsoon strength and nutrients?
• Why not calculate accumulation rates of TOC and opal instead of concentrations (often used in the text to indicate accumulation, which is not correct due to the role of dilution) from the age model and GRAPE values?
• Delta15N values from Site 722 appear to be significantly lower than late Pleistocene values. What is the reason? Apparently, the data were raised on acidified samples (according to methods section in Bialik et al., 2020), which introduces spurious results. A “denitrification threshold” at 6 permil is not likely, when thermocline nitrate originating in the AA confluence has a value of more than 5 permil then and now.
• The origin of coastlines used in Figure 6 is unclear and they do not really illustrate the prominent role of emergence of the Arabian Peninsula proposed in the text. In fact, the panels all look pretty similar to me (except the hand-drawn lines supposedly illustrating nutrient import): what is the line and label N:P/Si supposed to mean? As far as I know, SAMW is not the upwelling water mass today as suggested in the figure, but that should possibly be treated somewhere in the text (for modern, aka post-Miocene conditions).

Detailed comments keyed to line numbers:

17 I find the statement in first sentence of the abstract difficult to understand. Why is that so?

20 In my understanding, upwelling cells are localized spots of high upwelling intensity caused by the interactions of wind and local topography. Many of these cells combined form upwelling systems. One is the WAS upwelling system, the only major western boundary upwelling system.

37 Concentration of TOC in the sediment says nothing about accumulation! Take care to not use these two terms synonymously….

54 Upwelling acts as both sink and source of CO2 to the atmosphere, and each system differs in the net balance.

200 does this mean number of nannos per g of CaCO3 or divided by CaCO3? Unusual annotation! Are two digits after the decimal within the confidence limits of your method?

231ff So, are the results statistically robust, or not?

236 …2, whereas

258 placoliths are the small plates, arent´t they? How can they proliferate?

263-266 How can a “highly productive open marine environment” be nitrogen limited? This means an excess of phosphate and at that point, nitrogen fixation should kick in to make up the deficit. Please explain.

268 elevated sources?

269 N-limited nutrient sources meaning low N:P ratios in upwelling thermocline water?

273 elevated nutrient levels compared to what? Frequently the comparative is used throughout the text without reference to what the comparison refers to.

277-278 The thermocline and nutricline coincide usually and there the nutrient levels are high at the base – that is why the plants are there in the first place even at low light levels (deep chlorophyll maximum) …..so, what does “elevated nutrient conditions” mean?

279 How does this setting differ from conditions of TG 1(b)?

286 concentrations, not accumulation rates!

287 Why should that high dust flux increase productivity? Or is that a consequence of enhanced ballasting and export flux?

290 more abundant than what?

294 How could such a nitrogen-excess situation arise? Very difficult to imagine at fixed N:P ratios in thermocline waters and active denitrification in the OMZ! It appears that you put a lot of trust in a limited set of culture data for recent N-cell clones of Emiliania huxleyi.

295 strong upwelling

299-301 weaker, stronger, higher than what?

304 P-limitation (see query above)

310 what is an active OMZ?

333 (and 344) more limited than what? If there is no upwelling and only low productivity, how would an OMZ form in the first place? What is the exact link between monsoon strength and the OMZ without an intermediate link created by organic matter rain rate and water mass residence time?

359 How and why is intensified upwelling linked to high delta15N?

390 what amplifies? Declining upwelling?

401 I may have missed an explanation, how upwelling intensity is recorded in delta15N values.

411 What is the role of the temperature gradient today? Is it the influence on the sea level pressure gradient? How is the deep-water temperature gradient (between which end members?) involved?

415 gradients, thereby

428 indicated by instead of related to?

437 I must have missed the link between Mn/Al ratios and productivity. OMZ intensity comes to mind, but how is that linked to low productivity in a low nutrient regime?

440 As a result of? Explain the link between SST, sea level and upwelling!

443 why poorly ventilated? In line 461 below you state that that increase/decrease is indication of a shallow and poorly vented thermocline - what changed?

464 this is not accumulation! You might have a lot of TOC raining down, but when it is diluted by a lot of dust, for example, TOC concentrations are low!

468 formation of nutrients? Are they formed, or are they not used up because of light limitation?

482 mineralizing primary producers – is that a commonly used term? Other people use that word for dissimilation of organic matter and nutrient release

505-514 This entire discussion is very difficult to follow and possibly not suitable here: You infer from a size shift in one genus that the nutrient regime changed, but then discount this explanation and invoke changing nutrient limitation, but do not state the nature of that limitation.  

518 – 520 the concept of Mn-redirection was lost in the introduction. Do you talk about sediments, or water? Are high concentrations in sediments seen at the top and bottom of the OMZ where it intercepts the margin?

531 Shifts in nutrient saturation? I don´t think that you can saturate seawater in nutrients.

551 are you talking about N:P:Si ratios?

558 intermixing with

560 There is abundant literature on iron supply from continental margin sediments, particularly when they are situated in an OMZ

570 indicate a change in

575 quantity of nutrient enrichment?

579 Explain how that affects the northern AS (see above)! I am not entirely convinced that the record from the AS is compelling evidence….there may be other factors at play.

583 what is “nutrient rejuvenation”?

588 I am not sure I understand the argument for an increasing wind regime.

592 Explain how and why wind shear increases, then causes a global shift in ocean-atmosphere circulation, and deepens the thermocline. In my view, increasing wind shear causes open ocean upwelling and shallowing of the thermocline!

613 nutrient poor. But how then do you explain the OMZ that is apparently evident at that time?

623 became

624-626 According to You and Tomczak, 1993 and You, 1997, the upwelling taps essentially Northern Indian Ocean Central Water mixed with Red Sea/Persian Gulf waters.

635 That drop in SST is certainly not exclusively linked to a specific water mass, or is it? Not to enhanced upwelling?

662 what kind of shift? Excess phosphate? Less silicate?

666 you never refer to fluxes, but to concentrations. Would it not be simple to use your age model to actually calculate component fluxes from 722 GRAPE data?

667 delta13C is not shown in Fig. 3

669 which environmental stressors aside from nutrients?

References not cited in the manuscript:

You, Y. and Tomczak, M, 1993. Thermocline circulation and ventilation in the Indian Ocean derived from water mass analysis. DSR I, 40-1, 13-46.

You, Y., 1997. Seasonal variations of thermocline circulation and ventilation in the Indian Ocean. JGR, 102/C5, 10391-10422.

In the manuscript entitled “Biotic Response of Plankton Communities to Middle to Late Miocene Monsoon Wind and Nutrient Flux Changes in the Oman Margin Upwelling Zone” submitted to “Climate of the Past”, G. Auer and co-authors retrace the evolution of the upwelling cell in the Western Arabian Sea over the Middle to Late Miocene interval ~15 to 8.5 Ma. The authors integrate counts of calcareous nannofossils and diatom frustules in 71 samples and counts of planktonic foraminifers in 28 samples (all converted to abundance %) with published geochemical data (XRF-scanning elemental data, carbonate, organic carbon and nitrogen isotope measurements) from Bialik et al. (2020). Statistical methods including cluster analysis are applied to investigate associations between microfossil abundances and environmental parameters derived from geochemical data.

The main findings of the research are: (1) the onset of upwelling at ~14 Ma along the Oman margin and development of full monsoonal conditions after ~13 Ma were closely linked to the evolution of regional tectonics and global climate; (2) a high-productivity regime was gradually established at ~12-11 Ma in tandem with high-latitude re-organization of intermediate-water formation (AAIW and SAMW); (3) peak upwelling productivity between ~12 and 9.6 Ma was driven by enhanced nutrient fluxes from increased AAIW and SAMW production; (4) upwelling productivity declined after ~9.6 Ma due to the waning of monsoonal winds.

The manuscript presents interesting, important results concerning a climate sensitive region within the Asian Monsoon system and it targets a period of the Middle to Late Miocene that has remained highly enigmatic. The multiproxy approach, combining records from calcareous and siliceous microfossil groups as well as geochemical data, provides valuable insight into the biotic response to long-term changes in upwelling-driven productivity along the Oman margin. The manuscript is well-suited to the scope of “Climate of the Past”, in particular to an issue dedicated to Dick Kroon, who pioneered paleo-monsoon research in the Arabian Sea. Despite these positive aspects, I feel, however, that the manuscript requires some substantial revision. Please find below some major and minor issues that should be addressed before the manuscript can be considered for publication in “Climate of the Past”.

Major issues

This work extends previous research published by Bialik et al. (2020) in “Paleoceanography and Paleoclimatology”. However, the amount of overlap between the two contributions is at times equivocal and will need to be clarified during revision. Clarification is needed, in particular, concerning the originality of the data and interpretations. Some reiteration of background information and reference to previous results from Bialik et al. (2020) appear in various parts of the manuscript, but relevant information is not always provided or easy to locate. For instance, Line 144 mentions that portions of the nannofossil data set was already published, but does not indicate how many samples were previously analysed.  Overall, the novelty of the findings in the current manuscript needs to be more clearly highlighted.

           The manuscript is relatively long and I feel that the discussion on Lines 253-514 (Sections 5.1-5.3 of Discussion) could be streamlined to avoid some internal redundancy. In addition, the authors might consider summarizing the temporal progression of environmental changes outlined in Section 5.2 into a table for greater clarity. It would also be useful to include this information into one of the figures or into an additional summary figure. In the current version, the temporal evolution of upwelling productivity is not indicated on any of the figures and one has to search through the dense text to find this key information. The color coding in Figs. 2-3 and 5 only refers to the cluster assignment, based on the nannofossil assemblages.

            The manuscript contains abundant references to the recent literature to support the interpretation of the results and the discussion. To strengthen the interpretations and to highlight the key findings presented on Lines 515-691, it would be useful to add a synthesis figure that provides a direct comparison of results with some of the cited published records. At present, only the temperature data from Zhuang et al. (2017) are shown in Fig. 3. Adding a synthesis figure would be especially important to demonstrate, for instance, that major changes in global climate and ocean circulation are synchronous with changes in regional upwelling-driven productivity, as proposed in the discussion. Overall, such a figure would considerably help to bolster the interpretations.

            The paper contains little information on present day circulation patterns in the Indian Ocean and on the origin of the water masses that upwell today in the Arabian Sea. As the authors put a great deal of emphasis on the role of distant circulation changes in controlling nutrient availability and ultimately productivity in the upwelling cell along the Oman margin, it would be useful to relate their reconstructions of past circulation in Fig. 6 to the modern scenario. I find the expansion of SAMW and AAIW to water depths of 1000 and 1500 m north of the equator during the Miocene somewhat surprising. However, I accept that very little is known about Miocene Indian Ocean circulation and that it may have markedly differed. Nevertheless, I still feel that other potential influences should be considered, such as the role of the Indonesian Throughflow, when the Indo-Pacific gateway was fully open. I am also puzzled that the water depth of Site 722 is not taken into consideration (see comment below for Lines 121-124), when assessing the intensity of upwelling.

Minor issues

Abstract

Lines 25-26: the sentence starting with “We combine….” is unclear. Please revise and clarify.

Line 30: the duration of the MCO, based on distinct isotopic events, is considered to be 16.9 to 14.7 Ma. A global d18O decrease at 16.9 Ma signals the onset of the MCO and a global d18O increase at 14.7 Ma marks the first step in ice sheet expansion and global cooling during the MMCT (cf. Holbourn et al., 2014, 2015).

Lines 32-34: please specify time interval. Do you mean after 12 Ma?

Lines 35-36: please break up into two shorter sentences.

Line 36: replace “beginning” by “the onset”.

Line 39: unclear what “SAM” refers to here. Acronym not previously explained.

Line 40: “The absence of full correspondence…”. Not really clear what is meant here.

Line 43: omit “fossil” here.

Introduction

Line 48: “a” missing before “biomass”

Lines 54-60; upwelling areas can also be sources of CO2. Please revise.

Line 68: on glacial-interglacial timescales rather than “in”.

Line 80: becomes established rather than “establishes.

Line 93: sentence starting with “To date…” needs attention (seems incomplete).

Line 102: see comment for Line 30 about duration of MCO.

Line 104: verb should be plural (were established).

Lines 106-107: the MMCT usually refers to the Middle Miocene interval 14.7-13.8 Ma, and does not correspond to the Middle to Late Miocene, as implied here.

Section 2

Line 119: “location” more appropriate than “locale”?

Line 120: please add water depth of Site 722.

Lines 121-124: (a) Need to clarify relationship of Indian Ocean OMZ and Arabian Sea OMZ.

(b) Present-day water depth of Site 772 is 2028 m, which is well below the Arabian Sea OMZ (given as between 200 and 1000 m on Line 123). Please check references on the OMZ vertical extent at site location.

Line 125: according to ODP/IODP convention, site needs to be capitalized only when referring to a specific site (e.g., Site 722).

Line 127: should be Hole 722B (one of the holes drilled at Site 722) and not Site 722B.

Line 128: I guess you refer to Bialik et al., 2020a, when you mentioned “data used in this study”. This is a bit confusing, so please give reference here.

Section 3

Line 147: Fig. 2 does not really show how the correction factor was derived and applied. Relevant information is required in Methods.

Line 173: replace “less than” by “fewer than”.

Line 190: data usually plural.

Section 4

Lines 191-193: verb appears to be missing.

Lines 197-199: verb tense should be consistent (either past or present).

Line 237: please provide brief information on preservation, as you did for calcareous nannofossils.

Section 5

Line 317: should be “delineate”.

Lines 353, 376, 443, 451, 486: please use either ”foraminifer” or “foraminifera” consistently throughout the ms.

Line 384: verb should be plural (are).

Line 387: AABW and NADW usually referred to as Antarctic Bottom Water and North Atlantic Deep Water (not Waters).

Line 388: Did Woodruff and Savin (1989) specifically referred to this time interval? Precise dating of NADW expansion and AAWB intensification in this part of the Late Miocene remains controversial.

Line 412: why refer to Fig. 1 here?

Line 461: should be “foraminifera” to be consistent with remainder of text.

Lines 466-467: reference needed for this shift at 12 Ma.

Line 515: there is a problem with the numbering of sections: duplication of headers (sections 5.1 and 5.2).

Lines 521-523: last part of sentence “paired with an expanded OMZ” appears disconnected from the first part of the sentence.

Line 634: Taucher et al., 2022 is not an appropriate reference here.

Line 711: these references are not appropriate here.

Line 731: ODP/IODP needs to be acknowledged for providing samples.

Line 1234: Kuhnt et al. 2015 misspelt.

Figures

Fig. 3: please specify calibration applied for TEX86 SST data from Zhuang et al. (2017).

Fig. 5: (a) Not really clear what is meant by “(note the abundance scaling of N*109/g)”, which also appears in the plot of % diatom frustules. (b) The label “Planktonic foraminifera” should be indicated on the last plot. One has to read the caption to find out what is plotted there. (c) please provide reference(s) in figure caption for previously published data, where appropriate.

References:

Holbourn, A.E., Kuhnt, W., Lyle, M., Schneider, L., Romero, O., and Andersen. N., 2014. Middle Miocene climate cooling linked to intensification of eastern equatorial Pacific upwelling. Geology, 42, 19–22, doi:10.1130/G34890.1.

Holbourn, A.E., Kuhnt, W., Kochhann, K.G.D., Andersen. N., and Meier, K.J.S., 2015. Global perturbation of the carbon cycle at the onset of the Miocene climatic optimum. Geology, 43, 123–126, doi:10.1130/G36317.1.