Foraminiferal Distribution in Two Estuarine Intertidal Mudflats of the French Atlantic Coast: Testing the Marine Influence Index

Round 1

Reviewer 1 Report

This submission presents the foraminiferal distribution in two estuarine intertidal mudflats (Auray and Vie) of the French Atlantic coast. In order to determine the factors controlling their distribution, authors, besides micropaleontological analysis, performed sedimentological and chemical analyses and multivariate statistical analysis. Additionally, they tested the applicability to intertidal estuarine mudflats of the foraminiferal indices of environmental quality commonly used in transitional waters, as well as the newly proposed Marine Influence Index.

The manuscript is well written and organized. The introduction provides sufficient background, the methods are adequately described, the results are clearly presented and the presented data confirm the proposed interpretation. The determination and testing of biomonitoring indices in areas affected both by natural and anthropogenic stress is of high scientific significance.

My overall conclusion is that the manuscript is properly constructed, suitable for the journal. In my opinion, this manuscript could be accepted for publication after minor revision.

In particular, I highlighted the following minor issues:

• I would like to suggest the authors not to use first person in the text and change it to third person :

Paragraph 117-127: “Our study….”, “We will identify...”, “…we will investigate...”, “We will apply…”

Line 131: “…we will study..”

Line 192: “…we used the percentage..”

Line 195: “…we used the classification..”

Line 204: “…we used Thorium..”

Line 207-208: “…we used samples..”

Line 249: “We privileged studies combining...”

Line 273: “We tested...”

Line 249: “We privileged studies combining...”

Line 300: “…we decided...”

Line 681: “…we will investigate..”

Also please check chapter 5. Conclusion

• Figure 1a illustrates 8 sampling sites for Auray estuary, but in chapter 2.2, line 159, seven sites are mentioned.
• Line 577: “..of medium sand, and the salinity is low (about 16).” Please add the measurement unit for salinity (16ppt?)
• Lines 600-603: “Therefore, a detailed comparison with our results is not possible, although some observations are given as supplementary material. The main conclusion is that the community did not change substantially over the last 25 years.” In my opinion, the conclusion that the foraminiferal assemblage did not change substantially over the last 25 years is of great importance, therefore I would like to suggest authors to incorporate, in brief, the comparison to previous studies in discussion instead.
• Lines 747-748: “…we assume that the Auray and Vie estuaries are not subjected to strong anthropogenic pollution.” I think that the presented data are adequate to replace “assume” with “it is indicated” or “it is suggested that the Auray and Vie estuaries…..”

All the best

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

Dear Editor,

 

the paper entitled "Foraminiferal distribution in two estuarine intertidal mudflats of the French Atlantic coast: testing the Marine Influence Index" by Marie Fouet and nine co-authors from the BIAF team at Angers University, France. The study found a gradient in living benthic foraminiferal associations from assemblages dominated by calcareous species in the mouth, and agglutinated taxa dominating the faunas in the inner part of two estuaries at the French Atlantic coast. Salinity, its variability and submergence time appeared the main controlling factors for this distribution pattern, which is parameterised by a new Marine Influence Index. As both estuaries are almost pristine environments, any deviations from this patterns are expected by the authors to indicate anthropogenic pollution influences.

 

Major concerns

 

The Introduction is a very balanced, comprehensive overview of the state of the art, well written and interesting to read. The only thing that worries me is that the present paper acts as a case study for another paper which has been submitted in parallel to the same journal by the same group of authors, cited as "[43] Jorissen, F.J., Fouet, M.P.A., Singer, D. and Howa, H.: The Marine Influence Index (MII): A tool to describe foraminiferal assemblages in estuarine intertidal mudflats." It is evident that the present study is dedicated as a diligent but routine piece of work of the PhD student Marie Fouet, whereas the conceptual paper with more impact and attention to be attracted is first-authored by her supervisor, Frans Jorissen. With respect to the progress and establishment of more young women in science, it is necessary that both papers are strictly separated, and that the merit of invention of the new index is granted to the very person who did the work. I recommend the Editor to undertake all necessary measures ensuring the authenticity and acknowledgement of her invention.

 

Despite of that, the authors state that they intended to test the MII index by using the data set from this study. Even though it sounds to me that this index has been developed by using data from this study too, the approach must fail because salinity data were, with one exception on page 17, line 577, not reported or considered in the present study.

 

The authors discussed the utility and applicability of foraminiferal indices for environmental quality assessments in estuaries and marginal marine environments on pages 18 and 19 of the present manuscript. Thereby, authors refer to assignments of foraminiferal species to certain ecological groups as given in the literature. These assignments can only be applied in the present study once the same species are meant. However, the determination of three out of twenty species as documented on Plate 1, i.e. not less than 15 %, is doubtful. The taxonomic uncertainty challenges the calculation of ecological indices and environmental interpretations in the present study.

 

Minor comments

 

Line 146: Figure 1 deserves different signatures for the geographical features. At present, they are hardly discernable. All cities have to be named. The "other areas " (plain grey) have to be specified as well (forests, farm land, etc.).

Line 165: an appropriate sedimentological term has to be provided instead of " beach-like structures directly on the basement rock".

Line 170: why was the surficial layer subdivided in two 0.5 cm thick slices in the Auray estuary?

Line 170: The FOBIMO protocol requires a sample size of 50 cc and not more than 70 cc as in the present study. Hence, the authors must state that they "largely" or "mainly" followed the FOBIMO recommendations here.

Line 175ff: far more explanations and accuracy estimates are needed in this paragraph. First, a theodolite is not a surveyor's level, hence less accurate. Please provide the trade mark and model of the manufacturer of the instrument used, the precision as given in the manual, and an accuracy estimate of your levelling and distance measurements. The elevation of the highest halophytic fauna is quite variable, also with reference to the prevailing salinity in the area. The relative height, and species inventory of the highest halophytes has to be provided for those points, where reference to a geodetic benchmark is available at short distance. The tidal curve is not a sinus. The authors must state which model they used, linear, sinusoidal, or astronomical, and how this model fits to the local tide curve based on actual water level data from a tide gauge in the study area.

Line 188: in the Introduction, the authors stated that the salinity is a limiting environmental variable in estuaries. Why was the prevailing salinity at the sampling sites not measured, and if so, why are the data, with one exception, not reported in this paper?

Line 208: How is the migration of certain elements along the redox gradient in the near surface sediments during burial taken into account to obtain geochemical background concentrations? Were regolith samples from the catchment area of the rivers analysed as well? They provide a much better reference.

Line 244: were the samples picked wet or dry?

Line 247: how can you test this index when you do not use or have the salinity data?

Line 255: were they determined to phylotypes, and was that applicable to all specimens found in the samples?

Line 308: the y-axes and appropriate scaling is missing in both panels. It is not explained why minor enrichment commences at ca. 1.3 fold and not at 1.0 fold as compared to Holocene sediments. A relation to regolith samples from the catchment area would provide a better relationship (see also comment on Line 208).

Line 460: after having read the previous paragraphs, it becomes evident that the indistinct correlations to environmental parameters are a matter of the different principle faunal composition in both estuaries, which you have put together for the Principle Component Analysis. The faunal data from each estuary has to be analysed separately.

Line 483 ff: I would have wished to read a description of the MII distribution along the estuaries first, and then I would expect a thorough comparison with the abundance of individual species. The authors did probably have zugzwang to find a presentable relationship at all, as depicted by a correlation with an appreciable low p-value.

Line 486: it is clear that this correlation is driven by three outliers from Vie estuary.

Line 494: not necessarily true. One must separate the different Elphidium species, because each of them has unique ecological preferences (see above).

Line 509: the authors should regard the salinity data in order to support this contention.

Line 515: as in any estuarine system, the mud is deposited during slack water and not during the outflowing tide. This is also evident from the statement in the next line.

Line 529: I wonder why the authors do not take into account the d13C values of soils in the area. This terrestrial organic matter is a main component of the total organic carbon in intertidal sediments.

Line 533: a lot of copper is required to show recognisable effects. Are there shipyards? If so, is sand blasting and painting done frequently?

Line 561: if fisheries, touristic and oyster farming are obviously done in an environmentally friendly manner, why do you seek for (rare) evidences for human impact in the foraminiferal assemblages?

Line 579: "(Figure 2). It " should read "(Figure 2). It " (remove double blank).

Line 580: the authors should consider the common knowledge that the abundance of arenaceous taxa in marginal marine environments is not a matter of the sand content of their substrate.

Line 596 ff: the main results of the cited earlier studies have to be summarized here and not in the supplements, otherwise this paragraph is incomprehensible.

Line 608 ff: the environmental quality was good to high, no significant impact of pollution could be found. This is a given fact. It is not clear why this again has to be tested in the following. The results are circumconclusive (see lines 616-619).

Line 691: with such a large scatter as shown in Figure 8a, and an R2 of 0.19, the correlation may be statistically significant but not mechanistically plausible.

Line 724 ff: foraminifers on mud flats live longer than just 30 days, and waters in an estuary are oscillating. The lower salinity during discharge phases certainly had an influence.

Line 743: I extract from the paper that the distribution of the foraminiferal assemblages is well understood, but some of the environmental drivers are less constrained.

Line 803: the important intertidal Elphidium williamsoni has not been mentioned in the text before, why?

Line 807: the Quinqueloculina of Figure 12 largely differs from Q. jugosa in apertural characteristics, size and density of costae. It is certainly another species, probably Quinqueloculina limbata d'Orbigny, 1826, which shows the same collar around the aperture.

Line 808: the oblique inner chambers rather resemble Quinqeloculina bosciana d'Orbigny, 1839, than Pseudotriloculina oblonga as stated in this paper

Line 809: the specimen is obviously an Elphidium translucens Natland, 1938.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 3 Report

Review of the manuscript “Foraminiferal distribution in two estuarine intertidal mudflats of the French Atlantic coast: testing the Marine Influence Index”, written by Fouet et al.

 

This manuscript is about to test a new index to describe the ecological status of intertidal environments. The authors have studied two estuaries and have applied and compared different indices to test whether is suitable for this kind of environment. The manuscript is well written and of high importance for the benthic community because it shows that the new MII-Index seems to work for intertidal environments that are different from other shallow water or marginal marine environments where other indices successfully have been developed in the past. One difficulty was that the manuscript, introducing the new index, is currently not available and, hence, could not read. A have some comments and suggestions, most of them are minor and can be found below. However, I would encourage the authors to think about the statistics that can be used for this kind of data (see also below).

 

Introduction:

 

Line 45: sea-level rise could be added here as an example.

Line 49: Estuaries “support? multiple economical activities”? Can the authors reword this phrase?

Line 54-55: “use-d” has been written twice. In line 55 it could be written e.g., “organisms are proofed as bioindicators”.

Line 60: There is new publication showing that foraminifera also can life in freshwater and soils (Holzmann et al., 2021) - an information that could be added here.

Line 101-109: I would assume that non-natural vegetation in combination with grazing also can have an influence on the benthic community.

 

Material and methods:

 

General: The authors use the new MII method to recognize the ecological status of intertidal environments. I wonder why they have selected two sites with a similar ecologic status. To test the MII, it would have been better to use to sites with contrasting ecological status. Hopefully, it will be tested in the near future (or is already tested in Jorissen et al., unpublished?)

 

Line 131: Please use present tense.

Lien 132-133: The authors should provide references here. The same is true for line 136-137, line 139-141, and line 149-150.

Line 159-161: The two to three stations, that have been sampled per site, are replicates? If so, then please report the distance between these stations.

Line 164-167: In the Vie estuary samples from the seaward side have been taken but not in the other estuary and some samples in the Vie estuary have been taken “on beach-like structures and “on mud-covered ramps”. Then, in one estuary 1-cm-thick samples have been used but in the other 0.5-cm-samples. What is the rationale behind this sampling strategy? A short summarizing sentence would help to understand this.

Line 191: Why the samples had not been decarbonated before grain-size analysis?

Line 242: The samples have been counted wet?

Line 262: Although often used, I would suggest not to apply PCA for biological data. The method is based in linear regression which is not an adequate method to describe the more unimodal behavior of species (see Legendre and Legendre, 1998). Next, PCA can only be applied to normal-distributed species data which seldom is the case (a transformation may help here (such as arcsine (perc/100) for center-log for relative abundance data)). It further overemphasizes the most dominant species if no transformation is applied. It is not mentioned but have the authors applied a DCA analysis prior to PCA analysis to analyze the species response? This easily can be done in package vegan (function decorana). The gradient length gives an indication whether species in data set have a more linear or more unimodal reaction norm. Leps and Smilauer (2003)/Legendre and Birks (2012) stated that a gradient length of <3/<2.5 points to a linear and a gradient length >4/>3 to a unimodal species response. An alternative ordination method is non-metrical multidimensional scaling (NMDS; metaMDS in vegan) that has some advantages over the classical ordination methods (e.g., Bray-Curtis distance (better for relative abundance data) can be used instead of Euclidian distance which is used in PCA; it has no arch effects as observed for correspondence analysis or horseshoe effects as observed for PCA in the case of real data; and finally, there is no need to test for normal distribution).

Line 266-269: The authors fit the environmental variables to the PCA axes. Why they have not used a direct ordination method such as (RDA, linear) or CCA (unimodal)? These ordination methods are also implemented in the vegan package.

Line 276: The MII looks interesting but I cannot say something to it because the paper introducing this index isn’t available yet.

 

Results:

 

Line 294: An R of 0.37 (R² of 0.14) points to a week correlation. The p value is near insignificant – was it > or < 0.05 or exactly 0.05? Please write R, p in italics in the manuscript.

Line 307 and 474: Please start each subchapter by some text and then present the figure(s).

Line 340: It is not critical for the presented study to remove station 8B (Auray estuary) from correlation analysis but “outliers” are better identified on a statistical basis by using the interquartile range or standard deviation around the mean.

Line 359: Inner-organic linings have been observed in these samples?

Line 376-377: It should be mentioned in line 259 that these diversity indices have been calculated.

Line 423-424: “the phylotype was determined for 50 randomly sampled specimens”. This is not mentioned in the methods. Please describe there.

Line 441-442: See my comments to line 262.

Line 455: Normalization is not mentioned in the methods. A method to make environmental parameters comparable is standardization à (x-mean (=normalization)) / standard deviation. There is a different font in line 455-456. Please correct.

Line 455-457: I would suggest to describe the species (samples) - environment relationships in figure 7 here.

Line 483ff: Why are the MII results itself are not shown and described? They are presented in the supplement (Table 11). Please mention accordingly. For the regression analyses: some of them are linear regressions but they based on the species and MII comparisons are not. Please describe, in the methods, which kind of regression has been used and whether the variables (particularly the species) fulfill the requirements for the regression methods used. If not, better use generalized linear models.

 

Discussion:

 

Line 505-509: This short paragraph could be deleted because it is a repetition.

Line 527-528: It this own observation in the field?

Line 597: “Erreur ! Signet non défini.”?

Line 596-603: Although there are differences in the size fractions used, it would make sense to add the information from the supplement here to support the statement that the foraminiferal community was almost similar in past 25 years. The authors only studied the live populations. Have they also checked the empty tests (as a time-average signal) and compared them to the live populations? If I have understood correctly, the authors have samples replicates? Is there a difference between them with regard to the benthic community?

Line 628-630: The correlations are weak, particularly for emergence time and MII.

Line 630-631: I cannot really follow this statement here?

Line 658: Please add “in our study area” here.

Line 725: What is RD/CS?

 

Figures/ Tables/ Plates

 

Figure 4: The authors have done a regression analysis here? If so, this should be mentioned in the methods.

Figure 6: Please add a y-axis heading. The figure looks a bit incomplete: “100” on the y-axis is capped.

Plate 1: Some of the “most abundant” species shown here have been not mentioned in the text. Please mention them or remove them from the plate. Species 12-14 should be better arranged by presenting them with the aperture upward.

 

Supplements:

 

Suppl. tables 8 and 9: The complete census counts should be provided here or in an online archive.

Line 1093-1095: I would assume that this could also be an environmental signal. Bolivina striatula should also be found in the >125 µm fraction if the species is present

 

References cited:

 

-Holzmann, M., Gooday, A.J., Siemensma, F., Pawlowski, J., 2021. Review: Freshwater and Soil Foraminifera – A Story of Long-Forgotten Relatives. J Foraminiferal Res 51, 318-331.

-Legendre, P., Birks, H.J.B., 2012. From Classical to Canonical Ordination., in: Birks, H.B.J., Lotter, A.F., Juggins, S., Smol, J.P. (Eds.), Tracking Environmental Change Using Lake Sediments. Springer Science+Business Media B.V., Dordrecht Heidelberg New York London, pp. 201-248.

-Legendre, P., Legendre, L., 1998. Numerical Ecology. Elsevier, Amsterdam (Netherlands).

-Leps, J., Smilauer, P., 2003. Multivariate analysis of ecological data using CANOCO. Cambridge University Press, Cambridge.

Author Response

Please see the attachment

Author Response File: Author Response.docx