Establishing Baseline Assessment Levels for Monitoring Coastal Heavy Metals Using Foraminiferal Shells: A Case Study from the Southeastern Mediterranean

Round 1

Reviewer 1 Report

Establishing baseline assessment levels for monitoring coastal heavy metals using foraminiferal shells: A case study from the South-eastern Mediterranean 

Authors: Lin Hoober, Danna Titelboim, Sigal Abramovich, Barak Herut, Nadya Teutsch, Tal Benaltabet, Adi Torfstein

 

The proposed manuscript focus on the use of geochemical composition of benthic foraminifera to monitor heavy metal pollution in the environment. More specifically, the authors propose a procedure to define baseline assessment levels (BAL) for 3 heavy metal (Zn, Cu, Pb). For this purpose, they benefit from a very unfortunate but also nice opportunity to compare “unpolluted” geochemical composition of foraminiferal calcite with the one of specimens exposed to tar pollution event that impacted Israeli coast in 2021.

The manuscript is really well written and easy to follow. The idea of defining BAL for foraminifera is nice and innovative. This work will definitely interest the scientific community and I hope the bio-monitoring agencies. The authors chose to show the minimum of data within the manuscript and present a lot of supplementary material. I understand that they want to keep the manuscript straightforward but this is too extreme to my opinion. This is a scientific paper and data should be discussed in more details and part of the currently in Sup. Mat. should definitely be presented in the body of the manuscript.

Moreover, the discussion is too short. Although we see the purpose of the work done here, the authors should open the discussion: why using this specific way to calculate BAL knowing that these kind of dataset will be hardly available in monitoring studies? How applicable are these BAL to other study area? To other species? What is the benefit/necessity to use LA-ICPMS data rather than bulk analyses to define the BAL concentrations?

Please find below a list of major revisions and comments on the manuscript as well as editing remarks that have to be taken into account before the manuscript can be accepted for publication.

Major revisions

Introduction

• The authors explain clearly the limitation of direct HM measurements in seawater. Maybe they could discuss a bit more the interest of measuring HM concentrations in foraminiferal shell compare to measuring concentrations directly in the sediment where BAL are already defined.
• “The incorporation of chemical impurities (such as HM) in the calcite shells reflects, among other parameters, the composition of the seawater in which they were precipitated [10,18,19,23–34].”: To my opinion, some of the references are not appropriate since they do not refer to the record of element composition of seawater into the shell or they are based on fossil records. I would suggest to remove 26, 27, 31, 32 (that refers to temperature, pCO2 and salinity experiments) and 30 (fossil record). On the other hand, I would suggest to add De Nooijer et al. (2007), Barras et al. (2018), Schmidt et al. (2022), culture experiments dealing respectively with Cu, Mn and HM mixture incorporation.

Material and methods:

• “a total of 200 winter and 150 summer P. calcariformata tests were divided into two and three replicates (100 and 50 tests each, respectively)”: There seems to be a big difference of size between Winter and Summer specimens for this species. What could be the effect on metal incorporation?
• “more extensive bleaching treatment to remove any residual contamination related to the exposure of the shells to Rose Bengal”. I don’t really understand why the Rose Bengal stained forams would need more bleaching than the living unstained ones. Rose Bengal is staining cytoplasm of the individual so if cytoplasm is removed, there should be no RB left. However it is easier to see cytoplasm remains when it is pink! Are you sure that the 30mn in total in the oxidative solution was enough for the 2021 samples?

Results:

• Table S3.2 and S3.3: why are there no data for Cu/Ca and Zn/Ca for bulk analyses in 2013? While there are data for bulk analyses in 2021?? This should be said and explained in the text. Also, I am surprised that Mg/Ca data are not used/shown in the paper. These data would be useful for comparison between LA and bulk analyses?
• “Methods: Potential differences between whole shell solution ICP-MS and single chamber LA-ICP-MS were evaluated … indicating that the change in HM contents between both years is not related to analytical biases.”: The comparison between bulk analyses and LA-ICPMS analyses should be presented within the manuscript in a figure (or within figure 1) in the main text, not only in supplementary table.
• “Years: Zn/Ca, Cu/Ca, Pb/Ca were significantly higher in 2021 compared with 2013 in both species (Tables S4.3, S4.6).” This is not true for Cu/Ca in P. calcariformata. The same assumption is made in the discussion. Please correct.
• Methods: Tables S8 and S11 are missing, although I’m wondering if the authors are not referring to Tables S4.1 and 4.4. Please check and correct.
• Figure 2:
  • What is the error bar representing in the graphs (SD, 2SD, SE?). Please give this information in the legend.
  • I am quite surprised to see such high SD for the bulk analyses (e.g. for Lachlanella for Cu&Pb in W and S2021, Zn in S2021, or P. calcariformata for W2021). Could you compare with other studies and give an explanation? Would contamination (i.e. problem of cleaning efficiency) be a potential explanation for such a high variability?
  • Why Mg/Ca data are not presented? This is an interesting geochemical proxy since it would give information about temperature variations between Summer and Winter knowing that temperature may have an impact of growth rate and therefore on metal incorporation.

Discussion:

• The authors explain very well how they calculate the BAL values and what is the logic behind it. However, this approach requires a lot of data that might not be available for other study area and/or other species (i.e. LA-ICPMS data from pristine environment, information about biological natural variability for the metal considered thanks to culture experiments). On a pragmatic point of view, I think that it would be interesting to discuss the applicability of such approach in bio-monitoring study in other study areas and with other species. This would fill out the discussion part which is at the moment very short:
  • What alternative could be proposed in case initial database are not available? In fine, I have the idea that BAL values are really close to the mode of the distribution. Would that be the same for other species? Would it be an option to consider this value as BAL if data are not available to apply the proposed BAL calculation?
  • Discussion on the generalisation potential or limitations of the BAL proposed to transfer elsewhere.
  • Discussion about the benefit/necessity to use LA-ICPMS data rather than bulk analyses to define BAL values. As the authors are writing in the introduction, bulk analyses has the advantage of being more accessible with respect to laboratory facilities…
  • Discussion about the possibility to define BAL using dead shell when pristine data are not available.
  • Moreover, as a comparison, authors could describe how are defined BAL values in other cases, e.g. in mussels for example (on hard shells and/or on soft tissues).
• Point 3. There is a problem in the appropriateness between the reference cited and what is written… Titelboim et al. (2017) does not deal within culture experiments on both species for each metal of interest in this study. It looks like there have been confusion between two studies of Titelboim et al. On the one hand, Titelboim et al. (2017) corresponds to an in situ study which deals with both species but no culturing experiments are reported and this study is only dealing with Mg/Ca ratios. On the other hand, Titelboim et al. (2021) deals with culture experiments on Amphistegina (and not on both species considered in the present study) looking at Zn and Pb, but not Cu. The authors need to clarify this part and be more precise in the data they are using to evaluate this biological effect.
• It is missing a discussion about why a difference in Cu/Ca ratios is observed for Lachlanella between 2013 and 2021 and not for P. calcariformata.
• “This contamination event is further supported by the fact that the whole shell Zn/Ca and Cu/Ca of both species from 2021 overlap with the upper tail of the 2013-2014 LA-ICPMS records.” I’m sorry but I’m not sure I understand the idea… How to explain such high values in 2013 since there was no oil spill at that time?
• I don’t think that there is a discussion to explain such high variability and variable variability in the bulk data? From 2013? And from 2021
• Why do you have such difference in Zn/Ca between W and S 2021 for P. calcariformata? Could this come from the release of Zn through time from the tar? But why is it not visible in the other species (although the variability is much higher in Summer) or for other pollutants (Cu, Pb)? Van Dijk et al. (2017) observed a relatively high effect of carbonate chemistry on Zn/Ca ratios in Ammonia (culture exp). Do you think that this could explain part of the significant difference observed? Or maybe another environmental parameter? I think this would be interesting to discuss.

Conclusions

• “Seawater HM concentrations at the studied Nachsholim site (Table S3.1) were comparable with corresponding…”. This should not arrive only in the conclusions but be part of the Results and Discussion. This is an important argument.

Minor revisions

• Abstract : give the full name of the genus of P. calcariformata

Introduction:

• “foraminiferal shell chemistry could be used as a proxy for HM contamination [16–23]”: the authors could add the reference of Oron et al. (2021)
• “baseline levels of any environment in which they exit so that a reference”: change “exit” to “exist”
• “The incorporation of chemical impurities (such as HM) in the calcite shells reflects…”: it looks like it is incorporated as particles whereas elements are most of the time incorporated as dissolved ion within the crystal lattice. Maybe change “impurities” for “heavy metal ions”?
• “for 12 specimens of each species: 246 of Lachlanella and 167 of P. calcariformata, representing the entire sampling period.”: this is a very practical information that should not arrive in the introduction to my opinion.

Material and methods:

• Add a space before the “34” in the coordinates
• “February 16 and effected almost the entire”, change “effected” by “affected”
• “Two sampling campaigns were carried out…”: add at the start of the sentence “For this study,“
• Missing a dot after the sentence “(100 and 50 tests each, respectively)”
• “(see S1 and Figure S1 for more details)”. S1 is the same figure as Figure S1. This should be referenced only once and only 1 supp mat S1 should be available (the one including the text description of the different steps).
• Text S1: “Several cleaning steps of the specimens were placed in precleaned 2mL vials, were performed”. Please rephrase. Change “remnant” to “remaining”. In “H2O2”, use indices for the numbers. In “rinsed in in Milli-Q water”, remove second “in”.
• Change space around the comma in “(Agilent ,8900)”
• “Rstudio software”: the software itself is R, not Rstudio
• Table S2.1: Change “analyses session” to “analytical sessions”. Write Amphistegina in italics. Change the SD “1.5” to “2” for March 2021 Red Sea sediments.
• Table S2.2: homogenise the number of digits after the comma (between value and SD, and between dates for 1 given element (since values are in the same range)). Why values for Ca are not given?

Results:

• “Theses variabilities”, change to “These variabilities”
• “2,364 vs 2,806 [μmol/mol]”, remove the square brackets. Express these values in mmol/mol here and in Fig. 2.
• “the most substantial increases were in Pb/Ca is in both species”: remove “is” in the sentence
• Add in the legend of Fig. 2 “(see text for explanation about the calculation of BAL ratio)” after “Red horizontal lines mark the BAL ratio for each anthropogenic metal (Cu, Zn and Pb)”
• Table S4.1: add a column with the info about the statistical test used as in the other tables. Table S4.2, the last column heading should be “Results”, not “statistical test”. Table S4.3: precise in the legend that 2013-2014 data are LA-ICPMS and 2021 are bulk measurements.
• “Zn/Ca, Cu/Ca, Pb/Ca were significantly higher in 2021 compared with 2013 in both species (Tables S4.3, S4.6).” According to the legend of the tables, you are comparing the 2021 data to the all dataset from 2013 AND 2014. Please correct in the sentence accordingly and also in the rest of the paragraph.

Discussion:

• In point 3, change “present” to “percent”
• I’m not convinced that the log scale is pertinent in that case. Maybe using 2 different graphs (with linear scale) for each species would be more appropriate to better appreciate the proportion of each contribution (1-2-3) to the BAL value.

References:

• Ref 23: Please check the official year for this publication. I think it should be 2021.
• Ref 16: not properly referenced

 

References listed in this review:

Barras C., Mouret A., Nardelli M. P., Metzger E., Petersen J., La C., Filipsson H. L. and Jorissen F. (2018) Experimental calibration of manganese incorporation in foraminiferal calcite. Geochim. Cosmochim. Acta 237, 49–64.

de Nooijer L. J., Reichart G. J., Dueñas-Bohórquez A., Wolthers M., Ernst S. R., Mason P. R. D. and van der Zwaan G. J. (2007) Copper incorporation in foraminiferal calcite: Results from culturing experiments. Biogeosciences 4, 493–504.

Oron S., Sadekov A., Katz T. and Goodman-Tchernov B. (2021) Benthic foraminifera geochemistry as a monitoring tool for heavy metal and phosphorus pollution - A post fish-farm removal case study. Mar. Pollut. Bull. 168, 112443.

Schmidt S., Hathorne E. C., Schönfeld J. and Garbe-Schönberg D. (2022) Heavy metal uptake of nearshore benthic foraminifera during multi-metal culturing experiments. Biogeosciences 19, 629–664.

 

 

Author Response

Thank you very much for review. Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The work presented here by Hoober and coauthors introduces a new concept to use BAL of HM to for biomonitoring purposes. The concept is interesting, and might be developed further in a new standard for biomonitoring. At the moment, however, I think the manuscript is lacking method development to become a true guideline. In my opinion the manuscript is mainly lacking two things: the details on determining BAL and the embedding & comparison of the obtained data with existing literature studies, which will hamper the development and application of this biomonitoring tool. The method development is very minimal and the guidelines on how to determine this are sometimes lacking (e.g. how many measurements are needed, how to determine exactly the 3 different components).

Since this manuscript is an introduction of a new concept, it is crucial to clearly convince readers to use, add and improve this dataset on BAL. The manuscript, as is, need some further work and “fleshing out” to reach this point.

Major comments (minor comments are marked in the pdf)

Note: for now, I did not in detail correct typos, spelling etc. I strongly advise a second round of review, since in my opinion there will be a lot of changes and addition in text.

The dataset presented in this manuscript is interesting, but in my opinion lacking content. The authors should consider ‘fleshing out’ the manuscript, by two ways:

1) investigating the foraminiferal shells from both before and after the spill using e.g. SEM to study potential shell/ultrastructural changes induced by the oil spill (see also Frontalini et al., 2017). There are quite some other papers that investigate the link between shell deformation and heavy metal concentrations, and I think the manuscript as is an excellent opportunity to investigate this and it would really benefit from a more comprehensive approach.

2) A more thorough comparison to E/Ca from the literature. There are more studies on the effect of contamination or high heavy metal concentrations on foraminiferal geochemistry from both the field and the laboratory, like Nardelli et al., 2016 or Oron et al., 2021). By comparing partition coefficient and trends (for example, the observed higher or lower E/Ca for hyaline or milliod species, is this consistent with data from other species?) between these studies, even if it is not the exact same species (!) is very interesting from a biomineralization point of view, and also important to see if BAL from one hyaline species could be used for other one. By presenting an overview of the BAL calculated in this manuscript compared to values from field studies will give the approach more strength and depth.

This manuscript proposes guideline for a new method to determine HM BAL. Therefore, details on how to determine BAL (and why and how these 3 components are chosen) should be incredible clear, since it will be used by other scientists to increase the BAL database. I miss some details in the method development, like how many specimens should minimally be measured (this can be investigated by resampling the database using Monte Carlo simulations with the existing dataset). ow does the natural variability compare between the culture studies and the field samples (which is a combination of biological variability and seasonality), are the culture studies necessary to determine this? If it is possible to only use the field samples, how much time (seasons) do the samples have cover to include the natural variability?

In the discussion section, on application of BAL, I would like to see some arguments if it is feasible to determine BAL, and if species-specific BAL calibrations would be needed. The latter would complicate the applicability, since we would need to establish first the BAL for key species.

 

Suppl mat:

Check the phrazing of the first sentences of S1

Table S3.2. and 3.3.: Does 'replicates' mean single measurements on individual foraminifera, or the number of LA ICP MS measurements on all specimens combined? And they are average values +/- 1SD?

Check species names (in italics, and with genus and species where possible) and super/subscripts on chemical notation.

 

Comments for author File: Comments.pdf

Author Response

Thank you very much for review. Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

   Dear authors,

I would like to thank you for the modifications according to my comments.

Please take into account these last minor comments so that the manuscript can be published in Water.

 

• Use subscript for the 3 in HNO3 instead of lowering the size of the text. Same remark for Dmetal, Cacalcite, Casw in legend of table 1.
• Table 3: Check the alignment of text in all columns. And either repeat the reference for each species (line) or draw a line between different reference or use curly brackets to clarify.
• Table 3: In this study, the authors are dealing with high-Mg species. Since their table do not seems to aim to be exhaustive, I would suggest to remove low-Mg species from the table i.e. Asterigerina carinata and Amphistegina gibbosa from ref 42 and all species from reference 30.
• Figure 3 legend: “Letters show statistically significant differences between the measured groups.” Please be more clear. E.g. “Letters show statistically significant differences between analytical techniques”. Remove the “of” that is unnecessary after “ICPMS analyses”.
• Discussion : add a dot at the end of the sentence “Moreover, the metal/Ca levels among species of the same lineage can also be different
• BAL point 1: thanks for including this part on how the pivot value is calculated in sediment analyses. If I understood correctly, from “the picot value are supposed … “ to “lowest tail of ranked concentrations”, this corresponds to the methodology for sediment, this is not what the authors did here. Just to avoid confusion (and if I got it right), I would suggest this modification “The pivot values are supposed to be the lowest possible concentrations of normalized heavy metals, and the assessment of the pivot values IS performed by comparing them to the ranked heavy metal concentrations of all available data [44,45]. The pivot values are supposed to be close to the lowest tail of ranked concentrations WHICH EXPLAINS WHY WE CHOSE THE 5^th LOWER PERCENTILE IN THIS STUDY.”
• BAL point 3: “The relative standard deviation (RSD%) for each metal/Ca is used to evaluate the natural biological variability between and within specimens”. I’m sorry about MY confusion since indeed Titelboim et al. (2017) are presenting culture data. However, there are no data published for Cu/Ca, Zn/Ca, Pb/Ca in this article (neither in sup mat) to my knowledge. Therefore, I don’t understand how the SD for each metal can be evaluated. Either the authors used unpublished data (Cu/Ca, Zn/Ca, Pb/Ca data) from the experiments performed by Titelboim et al. (2017). In this case, this should be stated. Or they used the Mg/Ca SD and applied it to the other elements but this is not written this way. Please clarify.
• Chapter 4.2: Add a comma between Pb/Ca and Cu/Ca
• Conclusion : Please give a reference to this statement “Previous recommendations suggest that 6 LA-ICPMS spots will reduce the potential biological variability within a single specimens to less than 10%.”

Author Response

Reviewer 1:

Use subscript for the 3 in HNO3 instead of lowering the size of the text. Same remark for Dmetal, Cacalcite, Casw in legend of table 1.

Subscripts were used in the original file. It might be related to a format differences between software versions.

 

Table 3: Check the alignment of text in all columns. And either repeat the reference for each species (line) or draw a line between different reference or use curly brackets to clarify.

Corrected. Columns are aligned to the center and we added lines to Table 3.

 

Table 3: In this study, the authors are dealing with high-Mg species. Since their table do not seems to aim to be exhaustive, I would suggest to remove low-Mg species from the table i.e. Asterigerina carinata and Amphistegina gibbosa from ref 42 and all species from reference 30.

Corrected. low-Mg species were removed from Table 3.

 

Figure 3 legend: “Letters show statistically significant differences between the measured groups.” Please be more clear. E.g. “Letters show statistically significant differences between analytical techniques”. Remove the “of” that is unnecessary after “ICPMS analyses”.

Corrected.

 

Discussion : add a dot at the end of the sentence “Moreover, the metal/Ca levels among species of the same lineage can also be different

Corrected.

 

BAL point 1: thanks for including this part on how the pivot value is calculated in sediment analyses. If I understood correctly, from “the picot value are supposed … “ to “lowest tail of ranked concentrations”, this corresponds to the methodology for sediment, this is not what the authors did here. Just to avoid confusion (and if I got it right), I would suggest this modification “The pivot values are supposed to be the lowest possible concentrations of normalized heavy metals, and the assessment of the pivot values IS performed by comparing them to the ranked heavy metal concentrations of all available data [44,45]. The pivot values are supposed to be close to the lowest tail of ranked concentrations WHICH EXPLAINS WHY WE CHOSE THE 5th LOWER PERCENTILE IN THIS STUDY.”

Corrected. We changed the text according to the reviewer's suggestion.

 

BAL point 3: “The relative standard deviation (RSD%) for each metal/Ca is used to evaluate the natural biological variability between and within specimens”. I’m sorry about MY confusion since indeed Titelboim et al. (2017) are presenting culture data. However, there are no data published for Cu/Ca, Zn/Ca, Pb/Ca in this article (neither in sup mat) to my knowledge. Therefore, I don’t understand how the SD for each metal can be evaluated. Either the authors used unpublished data (Cu/Ca, Zn/Ca, Pb/Ca data) from the experiments performed by Titelboim et al. (2017). In this case, this should be stated. Or they used the Mg/Ca SD and applied it to the other elements but this is not written this way. Please clarify.
Correct. We used unpublished data (Cu/Ca, Zn/Ca, Pb/Ca data) from experiments performed by Titelboim et al. (2017). We added an appropriate reference (page 9, third paragraph) as well as an additional electronic appendix with the missing data.

 

Chapter 4.2: Add a comma between Pb/Ca and Cu/Ca.

Corrected

 

Conclusion : Please give a reference to this statement “Previous recommendations suggest that 6 LA-ICPMS spots will reduce the potential biological variability within a single specimens to less than 10%.”

Corrected. The relevant reference was added.

Reviewer 2 Report

Thank you for providing a revised version of the manuscript. Even though the discussion is in my opinion still not going in-depth enough, I will refrain from making new suggestions, since apparently it is not in the scope of the paper to do so. I would like to suggest the authors to next time also include references to the corrections / changed text (line and page number) in the reply letter to the reviewers, to more easily see the changed made to accommodate specific comments. It will save a lot of time for the reviewers and gives a better and easier overview,

Author Response

Thank you for providing a revised version of the manuscript. Even though the discussion is in my opinion still not going in-depth enough, I will refrain from making new suggestions, since apparently it is not in the scope of the paper to do so. I would like to suggest the authors to next time also include references to the corrections / changed text (line and page number) in the reply letter to the reviewers, to more easily see the changed made to accommodate specific comments. It will save a lot of time for the reviewers and gives a better and easier overview,

We thank again the reviewer for his/her support and thoughtful comments.