Autochthonous Versus Allochthonous Resources in a Tropical Rocky Shore Trophic Web Adjacent to a Marine Riparian Area

Round 1

Reviewer 1 Report

The manuscript 'Autochthonous versus allochthonous resources in a tropical rocky shore trophic web adjacent to a marine riparian area' from Pires-Teixeira and collaborators bring a very interesting question about the contribution of terrestrial vegetation bordering marine systems to the aquatic food web. I found the question very interesting, and even more, because the authors are evaluating the contribution of the Atlantic Forest to the marine consumers and do not the mangroves as usually evaluated. This brings the novelty of this work. However, the manuscript needs to be strongly improved in the sense of results presentation and discussion clarity. I suggest major revisions before being considered for publication.

The results section is very poorly presented, it is mandatory the improvement of the figures, tables, and the properly results presentation. The authors forgot to include one of the most important tables where the isotopic values of the organisms should be presented.

The discussion also must be improved. Please, see the details below.

Abstract
Line 15: “we used isotopes” – we used stable isotopes.
Line 21: “their eating habits” – their feeding habits

Introduction
Line 30: ‘occur and gradients form” – form gradients
Lines 54-65: Suggestion: delete. Stable isotopic analysis are very common currently and there is no need in explaining the basics of the technique in the introduction anymore.
Lines 72-74: I think you could provide some examples of what has been done, as analysing the importance of other terrestrial systems like mangroves, saltmarshes, etc to the marine adjacent area, showing how these systems many times are very important supporting marine communities. Then, after showing the importance of other terrestrial systems, you can explore that the Atlantic forest has been overlooked in this aspect. Some literature that may help:

 Bruno D. O., L. Riccialdelli, F. Botto & E. M. Acha. 2017. Organic matter sources for fish larvae and juveniles in a marine-estuarine interface (Mar Chiquita lagoon, Argentina). Environmental Biology of Fishes 100(12): 1609–1622.

Buelow C. & M. Sheaves. 2015. A birds-eye view of biological connectivity in mangrove systems. Estuarine, Coastal and Shelf Science 152: 33-43.

Taylor M. D., T. F. Gaston & V. Raoult. 2018. The economic value of fisheries harvest supported by saltmarsh and mangrove productivity in two Australian estuaries. Ecological Indicators 84: 701–709.

Line 90: “and test which basal” - I do not think you tested, I would recommend to use another world, like 'estimate' or 'evaluate'.

Material and Methods
Line 96: ‘state of Rio de Janeiro, has” - state of Rio de Janeiro, Brazil, has
Line 119: replace for "The relative abundance of benthic invertebrates was estimated by SCUBA diving on the ..."
Line 150: Please, describe the method (concentration of acid, which acid, drops x fumigation, etc...)
Line 159-160: Please, remove “dividing the isotope ratio of the sample by the isotope ratio of an internationally accepted standard multiplied by one thousand”
Line 167: No need for this topic. You could simply add a paragraph explaining that the trophic groups separation was used in the mixing models, and the organisms were categorized depending on their diet and cite table 1.
Line 170: reliable instead desired

Table 1: caption is missing.
Table 1: Try to not separate the name of the descriptor. I think you could try to better fit this table by reducing the text on the 3rd column and making more room in the second column to fit the name of the species and descriptors.

Results
I am having a hard time with the lack of the isotopic values of the sources. I read the text twice already, and still could not detect the isotopic values of phyto x SOM x C3 plants. A table is missing with the averages +-sd, as well as you could provide the results in the text. Also, please revise the captions and numbers of your figures and tables.

Line 197-205: Try summarizing this using trophic groups instead of the species. Species can be seen in the table.
Line 207: Please, include here average+- sd
Line 213: Sediment is not really a food resources, usually we measure the are estimating the isotopic value of the organic matter in the sediments (or debris). I suggest using SOM (sedimentary organic matter) instead through all the text.
Lines 213-216: Please, remove it. This is not a result; this is a premise for the technique. If your consumers were outside the sources polygon, you would not be able to use the technique, and no results to show here. Please, remove this sentence. Also, you need to provide the model diagnostics in the supplementary material. if your sources are too overlapped the model has difficulty distinguishing their contributions and you should probably use them together (but just if this makes ecological sense). Please, provide the diagnostic plots.
Lines 216-219: The is no information regarding the isotopic values in table 1. please, provide.
 

Fig.2: Very hard to read this figure. you could increase the scale and plot each individual consumer as a point in the biplot, so avoid the overlapping lines from the sd of each consumer into the resources.

Line 228: maybe it would be good to describe the trophic groups together with the species name? this way you make it easier to be applied to other systems... repeat this pattern throughout the results and discussion sections
Line 243: “except for the 242 crustacean M. hispidus (Figure 2).” - It is not possible to see this result in figure 2
Lines 246-247: “ The resource most assimilated resource in this group” – please, rephrase it

Figure 3 (it is listed as figure 1): nice figure but need to modify the lines because it is very difficult to differentiate 1-25% from 26-50% of resource contribution. also, as a suggestion, it would be very useful and nice to use this (modified) figure to summarize the results... if you modify this figure as a web of the interaction in the coral system, it would be very informative and could be used to help the discussion. you could layer the organisms in trophic levels and make a food web, all organisms in the same web instead of partitioning them into trophic groups. Then, you could substitute this figure for a graph with the contributions for each trophic group (would be easier to read the % contributions using a bar graph, for example), and use table 1 in the supplementary material.

Table 2: make this a supplementary material and use the data to build a bar graph with the contributions.

Discussion
You could make the reader's life a little bit easier here. In this current format is very confusing to read because we need to come back all the time to look at what kind of trophic category you are referring to. I think a good configuration for the discussion would be to separate in trophic groups section and at the end summarize with the modification of figure 3(listed as 1) as I suggested above.

Lines 272-274: No new information. C3 plants have much-depleted d13C and this is one reason why we can distinguish most terrestrial from marine contributions (C3 x C4). Please, remove it.
Lines 283-284: Why?
Lines 287-289: didn't understand in which context this information is here, what is the purpose?
Lines 304-305: Please, join the paragraphs to not break the flow of the reading.
Lines 308-310: maybe it would be important to evaluate how much of the mangrove vegetation is composing the SOM, and this could be an important source for the detritivores in the reef system…
Line 333: BIG - IGB?
Line 346: “values were central” - please, provide more accurate information. you could also run a model for SOM and better define the major contributor to marine debris in your system to improve this discussion
Lines 353-355: likely because of J. adhaerens abundance, right? maybe should state that.
Lines 357-389: would like to see the results…
Line 367: BIG again – IGB
Line 376: ‘occasionally a major resource” - in which case? results did not show that...

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

see attached

Comments for author File: Comments.pdf

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

This is my second review of the manuscript and, although major improvement has been made, now that the authors presented the results with more clarity, I could detect the methodological issue I was suspecting they had. I do not feel comfortable with the mixing models they presented considering they have more sources than consumers (as we can see in Table S1, Figure S1). Fortunately, it is fixable. My suggestion would be instead of running models for each species, they should consider grouping the species in the trophic groups and run only 6-7 models. This would give them a higher number of samples/consumers to return more reliable results. 

Also, some convergence diagnostics are missing because Figure S1 doe snot includes diagnostics for all the models that were run. I appreciate that the authors mentioned Gelman diagnostics review in the methods section.

I really like the idea of this manuscript, but the, methodological issues cannot be neglected. In this sense, I strongly encourage a second round of revision.

Suggested reading: Phillips DL, Inger R, Bearhop S, Jackson AL, Moore JW, Parnell AC, Semmens BX, Ward EJ. Best practices for use of stable isotope mixing models in food-web studies. Canadian Journal of Zoology. 2014;92(10):823-35.

Also, you should consider adding a phrase or two about the importance/extension of the Atlantic Forest (species that it harbors, diversity, ecosystem services, etc).

Author Response

Dear reviewer, thank you for your comments.

We ran the mixing model by trophic group (L159; L234-237; L292-296; Table S3) and have included a brief explanation of multiple food sources in the discussion (L392-400).

We have included the convergence diagnostics in the supplementary material (Table S3). In fact, some diagnostic plots were missing. We carried out six mixing models, one for each trophic group, placing the species as a group in the argument of the function. For some reason, the simmr package function generates the diagnostic plot from only one specie ("group 1") by default. However, we managed to generate the diagnostic plot by including the group argument in the function and put all plots in the supplementary material.

We have also included the information about the importance/extension of the Atlantic Forest (L65-67).

Reviewer 2 Report

Comments for author File: Comments.docx

Author Response

Dear reviewer, thank you for your comments.

Source contributions are now included in Table S2 in the supplementary material. If appropriate, we can reintroduce the table into the manuscript.

We included the mixing models with the gastropod and crustaceans grouped together. (L158-159; L201-205; L226; Table S2).

SOM is included in the mixed model of only two depositivores (Isostichopus badionotus and Holothuria grisea), and neither consume plankton. We did not combine these resources to avoid including a resource in a diet that they are not part of.

We have included a brief explanation of multiple food sources in the discussion (L392-400).

Lines 218-220: Due to the multiple versions in pdf and with and without track changes we could not find in the indicated lines what this comment referred to!