Review comments egusphere-2022-1144 (Letulle et al.)

Dear EGUsphere, dear authors,

Thank you for inviting me to review the manuscript titled “Multi-proxy assessment of brachiopod shell calcite as a potential archive of seawater temperature and oxygen isotope composition” by Letulle et al.  In their manuscript, the authors present a dataset consisting of various geochemical proxies measured in brachiopods grown at known temperatures. The goal of the study is to compare and assess the temperature dependence of various geochemical proxies in brachiopod shell calcite.

The authors describe a very comprehensive dataset, and their analyses has clear implications for paleoclimate studies. I would commend the authors on their coverage of the literature background required to follow the discussion in the study. This is not straightforward since the dataset describes various proxies and techniques, but I think they did this well. The description of the methods and results of this study and the line of reasoning put forward in the discussion also mostly makes sense to me.

My only major disagreement with the author’s assessment is the need for a taxon-specific calibration for clumped isotope-temperature reconstructions in brachiopods. In my opinion, the authors arrive at this conclusion a bit too hastily, only comparing their findings to one pre-existing clumped isotope-temperature calibration (by Anderson et al., 2021), even though previous studies have noted that this equation tends to underestimate temperatures (see detailed comments below). This conclusion of a brachiopod-specific paleotemperature equation is, in my opinion, also in disagreement with the conclusions of disequilibrium fractionation in brachiopods put forward by papers by David Bajnai (which are cited and discussed in the manuscript). I do not agree with the approach of circumventing this disequilibrium issue by inventing a new empirical temperature equation for using clumped isotopes for temperature reconstructions from brachiopod calcite. I therefore think the authors should revise this conclusion before the paper becomes acceptable.

Besides this point, I think this is a well written manuscript about a well-thought-out study which will be of interest to the paleoclimate community and merits publication. Below I provide some line-by-line feedback which I hope will help the authors to improve their manuscript.

Major comments

In the Introduction (lines 49-89), several potential temperature proxies in brachiopod calcite are introduced one by one. While these introductions are important and well-written, the individual paragraphs are quite detached from each other and disrupt the flow of the manuscript somewhat. I suggest the authors either tie the paragraph a bit better into the rest of the manuscript or place them in a separate “Background” paragraph.

Lines 124-129: Perhaps the authors can provide a citation or reason for why they used this pre-treatment method. There is some literature suggesting that pre-treatment with oxidizing agents might influence the (clumped) isotope or trace element composition of the carbonate. Personally, I am of the opinion that excess pre-treatment with such substances should be avoided in these types of studies. However, if the authors have convincing evidence (either by their own research or from the literature) that this treatment is warranted in this case, I am happy to support it.

Section 3.1: It is not clear from this section how the authors dealt with uncertainty on the temperature and salinity/d18Ow value associated with the samples. Since brachiopod samples are large and do not always represent mean annual averages (i.e. due to sampling of less than a full year, see lines 144-146, or due to variations in growth rate over the year) I think the authors should take into account the seasonal cycle in temperature and d18Ow (or salinity) at the sampling locality as uncertainty on their regression. I assume from the text in this section that the authors used a normal linear regression (not including errors on measurements or on the independent variables). To incorporate uncertainty on the independent variable, the authors could use a Deming regression which takes into account measurement error as well as errors on the “known” variable (in this case temperature and d18Ow). Judging from Figure 1 (which I assume shows uncertainties on temperature), these uncertainties are significant.

Line 348-351: I suggest the authors also compare the clumped isotope-temperature relationship in brachiopod calcite with the values obtained by applying the Meinicke et al. (2020) calibration, which was recently updated to the I-CDES scale (Meinicke et al., 2021). A recent study by de Winter et al. (2022) demonstrated that the Anderson et al. equation likely induces a cold bias on shallow water carbonates which might explain part of the observed offset in clumped isotope values in this study. I wonder if the brachiopod-specific clumped isotope equation proposed by the authors is significantly different from Meinicke et al. if projected on the I-CDES scale. If so, it might not be warranted to propose a new clumped isotope calibration as brachiopods might be calibrated with general calcite calibrations.

Meinicke, N., Ho, S. L., Hannisdal, B., Nürnberg, D., Tripati, A., Schiebel, R., and Meckler, A. N.: A robust calibration of the clumped isotopes to temperature relationship for foraminifers, Geochimica et Cosmochimica Acta, 270, 160–183, https://doi.org/10.1016/j.gca.2019.11.022, 2020.

Meinicke, N., Reimi, M. A., Ravelo, A. C., and Meckler, A. N.: Coupled Mg/Ca and Clumped Isotope Measurements Indicate Lack of Substantial Mixed Layer Cooling in the Western Pacific Warm Pool During the Last ∼5 Million Years, Paleoceanography and Paleoclimatology, 36, e2020PA004115, https://doi.org/10.1029/2020PA004115, 2021.

de Winter, N. J., Witbaard, R., Kocken, I. J., Müller, I. A., Guo, J., Goudsmit, B., and Ziegler, M.: Temperature Dependence of Clumped Isotopes (∆47) in Aragonite, Geophysical Research Letters, 49, e2022GL099479, https://doi.org/10.1029/2022GL099479, 2022.

Minor comments

Line 29: “but is significantly offset” rephrase to “are significantly offset” (“D47 values” is plural)

Line 85-87: Perhaps the authors could add here that current empirical clumped isotope-temperature calibrations also show good agreement with ab initio models of the carbonate isotope system, as was demonstrated in Jautzy et al. (2020). This is another argument in favor of the use of common clumped isotope thermometers in a variety of carbonate materials.

Jautzy, J. J., Savard, M. M., Dhillon, R. S., Bernasconi, S. M., and Smirnoff, A.: Clumped isotope temperature calibration for calcite: Bridging theory and experimentation, Geochemical Perspectives Letters, 14, 36–41, 2020.

Line 142: “graving bit” should this read “engraving bit”?

Line 262-263: “At temperate and polar temperatures (20 to 0°C) our equation has a steeper slope than that of Brand et al. (2019)” The authors need to explain this in more detail and/or refer to a figure where the reader can spot this effect. The authors refer to figure 2A later, but it is not clear which line in this figure represents the equation by Brand. I do not understand how the slope can be different with temperature if a linear equation (with a constant slope) is compared, but I might misunderstand what the authors are trying to say. Later on, the authors mention that the Brand et al. equation is non-linear, but I still have trouble following the description of the comparison in this section.

Line 342: “In consequence” rephrase to “As a consequence”

Line 415: “constrain” should read “constraint”

Line 416: “should be privileged for trace-element-based paleotemperatures reconstructions” consider rephrasing to “should be selected/prioritized for trace element-based paleotemperature reconstructions”

Line 443-466: This clear grouping based on trace element content is an interesting observation. I wonder if the authors considered whether there might be a relationship with growth rate. Does the “high” or the “low” group show significantly faster growth than the other? If so, this could be an explanation for the difference in shell composition, as trace element concentrations in calcifiers often show a correlation with growth rate. In addition, in this section about grouping of specimens based on trace element content, adding a figure showing the differences in concentration would be helpful.

Line 475: The authors might consider rephrasing the title of this section to: “Precipitation of brachiopod shell calcite out of equilibrium with seawater”

Line 493-495: The authors might consider citing the recent study by Garbelli et al. (2022) here who also interpret changes in isotopic composition of (fossil) brachiopod shells as seasonal variability.

Garbelli, C., Angiolini, L., Posenato, R., Harper, E. M., Lamare, M. D., Shi, G. R., and Shen, S.: Isotopic time-series (δ13C and δ18O) obtained from the columnar layer of Permian brachiopod shells are a reliable archive of seasonal variations, Palaeogeography, Palaeoclimatology, Palaeoecology, 607, 111264, https://doi.org/10.1016/j.palaeo.2022.111264, 2022.

Lines 516-531: It seems that some of the discussion of kinetic (growth rate-related) effects in trace element composition could be a useful addition to the section above where the observation of differences in trace element composition between brachiopod groups is discussed (see my comment on lines 443-466).

Line 532: “isotopic” should read “isotopic”

Line 539-540: “Given the highly…may be coincidental.” I think the authors should explain this line of reasoning in a bit more detail.

Line 582: Please add the missing “delta” before “¹⁸O_sw”.

Line 586: “central values” is a bit of a cryptic term, perhaps the authors mean median or mean/average values?

Lines 556-610: I think the addition of a fossil case study is nice, but it is not essential for the study. If the authors would like to keep their manuscript more concise, this is a section that could be significantly shortened or removed in my opinion.

Table 4: As mentioned in one of my previous comments, it might be worthwhile to add the clumped isotope-based temperature reconstructions based on the Meinicke et al. calibration, since these are likely more accurate and more closely in line with the modern brachiopod data in this study as well.

Line 623-624: The temperature underestimation by ~3 degrees is very similar to the offset found in de Winter et al. (2022; see comment above) and this nice corroboration between multiple datasets is worth mentioning.

Line 624-625: As mentioned above, I tend towards disagreeing with this call for a brachiopod-specific clumped isotope calibration, since most of the offset in clumped isotope values may be explained by the Anderson et al. equation underestimating temperatures in general (not just for brachiopods). The authors should consider this explanation before suggesting a taxon-specific calibration is needed.

Dear Editor, Dear Authors,

Thank you for allowing me to read this scholarly MS. Letulle and colleagues submit an MS documenting an in-depth analysis of different palaeothermometers making use of the brachiopod shell carbonate archive and discuss strengths and weaknesses of the different methods aiming to reconstruct seawater palaeotemperatures and oxygen isotope signatures. The authors propose a new fractionation equation between brachiopod shell calcite and seawater. I want to emphasise that I am, at best, a learnt amateur in clumped isotope geochemistry; hence, I approach the MS from the viewpoint of carbonate archive research, biomineralisation, geochemistry and palaeo-oceanography.

I like the concepts presented and think this paper should be useful for the community. The science seems mostly fine, and I did not recognise many significant technical problems. I list a limited number of scientific comments below.

From a more critical perspective, the writing ranges from good to less-than-acceptable for a paper submitted to an international journal. The paper is inconsistently written, both in its style and the quality of the technical English. In any case, many grammatical and spelling mistakes should be fixed. I sympathise with all non-native speakers (I am one), but there are now several sophisticated language editing software packages that we can easily download for free. These are very useful for those among us struggling with a foreign language. I will use some examples in the abstract to document why I think the authors must significantly improve portions of the MS concerning language, logic and clarity.

I also found the MS partly convoluted and often long-winded. This feature limited my ability to extract science and information. Some of the text must be shortened and focussed.

My review is not a word-by-word cookbook job but focuses on points that caught my attention while reading the text and are listed in the order in which they appear.

Abstract:

Poorly written. Many deficits in the logic and precision of the language. I list several points that caught my attention; there are others.

Ln 13. Please use `seawater´ rather than `ocean'.  Delta¹⁸O measurements. That is jargon. A `measurement' is the analytical step(s) we perform to generate the data, the analysis so to speak. The reconstruction of past seawater properties is based on oxygen isotope data (not the measurement thereof).

Ln. 18. Missing word: commonly used `archives' in studies….

Ln. 20. Unclear wording: …resistant to diagenetic alteration for decades. Do you mean brachiopod shells do not alter over the time span of several decades? Or do you mean that over the past decades, scientists have considered brachiopod shells to be resistant to diagenetic alteration?

Ln. 20. I am not sure what a `growing temperature' is (also referred to as `living temperature' elsewhere)? I did google the term to make sure I did not miss something. The only paper that matched is yours (this discussion version) on the EGU sphere webpage. Do you mean the ambient seawater's temperature during the brachiopod's lifetime? Use `ambient seawater temperature', I suggest. Other than that, please use proper terminology: seawater d<sup>18</sup>O `values' or similar.

Ln. 21. Again, I can only guess what a `supposed´ carbonate-based palaeothermometre is. Do you mean `novel´ or `less well established´? Moreover, the palaeothermometre is NOT based on carbonate but uses the archive data (geochemical properties) recorded in carbonate (note the difference between the terms archive and proxy).

Ln. 33. Missing word: …with `seawater´ temperatures…

Ln. 34. What do you mean by `relatively good´. In agreement with the measured temperatures within xy degree Celsius?

Introduction:

Much better written, but still some language deficits similar to those listed for the case of the abstract. Please consider.

Ln. 42. What are past seas, and what is the difference to oceans? Do you mean epeiric seas as opposed to genuine oceanic bluewater?

Ln. 52. Many inconsistencies concerning technicalities of cited references. See, for example, ln. 52. Brand et al., (2013) should read Brand et al. (2013).

Ln. 91. That is a scientific criticism. The authors argue about the question of whether shell carbon (DIC) and oxygen isotope values are in equilibrium with the seawater from which the shell carbonate precipitated or not. Please allow me to clarify that brachiopod biominerals are secreted from bodily fluids, NOT seawater. The problem is threefold: (i) What is the isotopic value of the bodily fluid relative to that of the ambient seawater? (ii) Does the isotopic value of the bodily fluid change during active versus passive cycles in the brachiopod metabolism cycle and during the brachiopods life span? Juvenile brachiopods grow rapidly, mature slow down. (iii) What is the fractionation factor between bodily fluid and brachiopod biomineral, and is it constant during the lifetime of a brachiopod? In some cases, brachiopod bodily fluids are isotopically close to the ambient seawater; in others not. In short, it is complicated. The authors provide text about thermodynamics and kinetics but less so about these metabolic effects and biomineralization pathways. In my oppinion, that is a weakness of the paper. Please see the discussion and references cited in:

Immenhauser, A., Schöne, B., Hoffmann, R. and Niedermayr, A. (2016) Mollusc and brachiopod skeletal hard parts: Intricate archives of their marine environment. Sedimentology 63, 1-59.

I emphasise that you do not need to cite my paper! That is entirely up to you. It simply saves the reviewer time when being able to refer to the text and the cited references in a published paper. Please consider.

Material and Methods:

No major comments, looks o.k. One exception, please avoid acronyms in titles (2.3) and please refer to `values'. d¹³C of modern brachiopods is jargon. Please use carbon isotope values of modern….

Results:

Header chapter 3.1. Please do not use `stable' as a synonym for carbon and oxygen isotope values; science knows about 120 stable isotopes.

Ln. 210. I always wonder, what is the meaning of the second decimal in a range of isotope values resulting from bulk samples? What is the meaning of -2.24 permil in this context? My opinion, the second decimal is meaningless. You analyse a bulk sample from a brachiopod shell, and I would refer to that as pseudo-precision. I suggest providing one-decimal values. Bulk samples and second decimals do not match. Particularly as you mix bulk sample data and data from the inner and outer shells (see Table 2).

Discussion:

General comment: This chapter is longwinded and, in part, difficult to follow. I advise streamlining the text and shortening it by at least 20%.

I wonder if the `holy trilogy´ of scientific writing consisting of Data Presentation, Data Interpretation and Discussion is applied here? If so, where is the discussion? Consider rephrasing the header as `Interpretation and Discussion'.

Ln. 282. What are `independent´ brachiopods? Please explain.

Chapter 4.1.2 is poorly written. Quite some problems regarding grammar and formalities (citations etc.). Please clean up.

Chapter 4.1.4 All good science but very longwinded. Could you streamline that? This is not easy to follow and this is not something you want to hear from the readers.

Chapter 4.2 Here, we need much more emphasis on metabolism and biomineral secretion from bodily fluids. The authors deal with the topic as if brachiopod biomineralization pathways were an inorganic precipitation experiment. These are super complicated little `bio-machines', and they are fascinating since each individual is a case on its own. Please see papers from the marine biology community (mainly aquaria monitoring experiments but also field observations).

Conclusion(s):  

Please use the plural, I suggest that you list more than one conclusion here.

This chapter is very much written in a discussion style. Please consider coming up with genuine conclusions style text rather than a short (renewed) discussion. The last statement is an anti-climax. First, you present all of these data and text. Then you tell the reader that you advise considering the variability in brachiopod live habitats, environmental conditions, metabolic effects, seasonal effects etc.? I must admit, have read very similar concluding statement in many papers published a decade or more years ago. Please consider.

The figures are reasonably well done.

As I said, I generally like the paper and the concepts and the data. I listed some of the numerous problems regarding syntax, grammar, and logic in an exemplary manner and indicated several scientific issues that require the authors' attention.

I look forward to reading (and citing) this article in its final incarnation.

Sincerely,

Adrian Immenhauser

The study by Latulle and colleagues provides new geochemical data from modern brachiopod shells (δ¹⁸O, Δ47 and Element/Ca ratios) to test their reliability as archive to reconstruct ambient temperatures and seawater δ¹⁸O. The novelty of this study is a revised equation of oxygen isotope fractionation that slightly differs from the previously published ones by Brand et al. (2013, 2019) by using a linear relationship for temperatures below 50°C. Further, the calibrated area extends to polar latitudes with MAT below 0°C. The authors discuss the accuracy and precision of the new equation in comparison to other carbonate-based paleothermometers by using the published brachiopod-derived oxygen isotope data of Bajnaj et al. (2018). In addition, the brachiopod-shell Δ47 values correlate with growing temperatures, but with an offset relative to other biogenic and abiogenic calcium carbonate minerals. The Element/Ca ratios instead do not show meaningful relations with temperatures.

The results of Latulle and colleagues supplement the results of previous studies and are thus a contribution for the improved development of brachiopod shells as a paleoclimate archive.  Brand et al. (2019) delivered alternative fractionation equations of oxygen isotopes based on an extensive data set of modern specimen. Bajnaj et al. (2018) identified kinetic effects of biomineralisation of brachiopods by the analysis of clumped isotopes, and Ullmann et al. (2017) identified the kinetic effects to differ taxonomically among different orders of brachiopods especially the suborder of Terebratellidina.

The here supposed combined approach of using brachiopod-shell-derived Δ47 and δ¹⁸O delivers growing temperatures and seawater δ¹⁸O, and has a good potential for the reconstruction of past environments, particularly in epi-continental and shallow marine settings where other carbonate archives as well-preserved foraminifera are rare. I recommend the study for publication. In the following are some comments the authors should address in a revised version.

• Comparison of the new equation of the oxygen isotope fractionation with those of Brand et al. (2019). Both equations are similar in the temperature range between 10-25°C, but differ in the low-temperature field (<10°C). Since the Brand et al brachiopod-data set is by far much better constrained by data points, the authors should provide a more in-depth discussion about causes of the offset. Further, they should strengthen their arguments why it is necessary to introduce the new equation, and why it shall be an improvement.
• In this context, please also indicate the MAT range covered by the Bajnaj et al. 2018 brachiopod data set (line 281).
• The supposed sampling procedure avoids specialized parts of the shell as umbo, edges, muscle scars, primary layers. However, Ullmann et al. (2017) observed additional significant taxon-specific ranges in their intra-specific high-resolution oxygen isotope data. How has this observation an effect for the results of this study? Is intra-specific variability smoothed by the sample size? Which degree of uncertainty introduce specimens of the suborder Terebratellidina to the fractionation equations? Please comment on this.
• The Jurassic example is not well executed and not the scope of this study. New and very few data points are introduced first time in the discussion. I recommend the removal of this part of the manuscript, since its focus is on modern brachiopod taxa.
• Please, explain all parameters and abbreviations in Supplementary Table S1.

Please check the manuscript for spelling errors, here are some I spotted:

• “Rhynchonellida” in Fig. 4 – revise spelling two “l”
• Line 86: Delete “previous”
• Line 98-99: incomplete sentence
• Line 375: Enter a space between “regression derived”
• Line 532: spelling of “isotopic fractionation”
• Typos in supplementary File S3 (Sheet Description)