Influence of Suspended Mussel Aquaculture and an Associated Invasive Ascidian on Benthic Macroinvertebrate Communities

Round 1

Reviewer 1 Report

Review

Paper title: Influence of suspended mussel aquaculture and an associated invasive ascidian on benthic macroinvertebrate communities

The authors conducted a field study to reveal fluctuations in the structure and abundance of gastropod benthic assemblages inside and outside of mussel aquaculture sites. The authors found a grater abundance and biomass macroinvertebrates at sites inside of leases. Additionally, the presence of the invasive ascidian Styela clava did not affect these parameters. The authors concluded that the  presence  of  macroinvertebrates  on  suspended  mussels offsets  impacts  on  the  abundance, biomass and productivity of macroinvertebrate communities in mussel farms. This study may have important implications for bivalve aquaculture.

All these reasons explain the relevance of the paper by Lisa Robichaud and co-authors submitted to "Water".

General scores.

The data presented by the authors are original and significant. The study is correctly designed and the authors used appropriate sampling methods. In general, statistical analyses are performed with good technical standards. The authors conducted careful work that may attract the attention of a wide range of specialists focused on freshwater ecology.

Remarks.

The authors should format the abstract as continuous text without subsections.

Materials and methods. The authors should include the full Latin name of the mussel species cultured at the sites examined.

Specific comments.

L 30. Change “impacts to” to “impacts on”

L 62. Change “structure” to “structures”

L 87. Change “This is also true on” to “This is also true for”

L 152. Change “Shaw 1998” to “[ 64 ]” and check the order of references.

L 272. Change “was resampled for bays” to “were resampled for bays”

L 281. Change “used” to “was used”

Author Response

Review

Paper title: Influence of suspended mussel aquaculture and an associated invasive ascidian on benthic macroinvertebrate communities

The authors conducted a field study to reveal fluctuations in the structure and abundance of gastropod benthic assemblages inside and outside of mussel aquaculture sites. The authors found a grater abundance and biomass macroinvertebrates at sites inside of leases. Additionally, the presence of the invasive ascidian Styela clava did not affect these parameters. The authors concluded that the  presence  of  macroinvertebrates  on  suspended  mussels offsets  impacts  on  the  abundance, biomass and productivity of macroinvertebrate communities in mussel farms. This study may have important implications for bivalve aquaculture.

 All these reasons explain the relevance of the paper by Lisa Robichaud and co-authors submitted to "Water".

 General scores.

 The data presented by the authors are original and significant. The study is correctly designed and the authors used appropriate sampling methods. In general, statistical analyses are performed with good technical standards. The authors conducted careful work that may attract the attention of a wide range of specialists focused on freshwater ecology.

Thank for your kind words on this manuscript.

Remarks.

The authors should format the abstract as continuous text without subsections. – Done

Materials and methods. The authors should include the full Latin name of the mussel species cultured at the sites examined. - Done

Specific comments.

L 30. Change “impacts to” to “impacts on” – Done

L 62. Change “structure” to “structures” – Done

L 87. Change “This is also true on” to “This is also true for” – Done

L 152. Change “Shaw 1998” to “[ 64 ]” and check the order of references. – Done

L 272. Change “was resampled for bays” to “were resampled for bays” – Done

L 281. Change “used” to “was used” – Done

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper of Robichaud et al. examined the effect of suspended mussel aquaculture and S. clava on benthic macroinvertebrates. In general, the paper provided a sound science in terms of analysis. However, there are major flaws in terms of presentation. For one, some parts of the materials and methods should be placed either in the result or in the discussion section. Please see attached file as I have placed my comments directly on the manuscript. 

Comments for author File: Comments.pdf

Author Response

The paper of Robichaud et al. examined the effect of suspended mussel aquaculture and S. clava on benthic macroinvertebrates. In general, the paper provided a sound science in terms of analysis. However, there are major flaws in terms of presentation. For one, some parts of the materials and methods should be placed either in the result or in the discussion section. Please see attached file as I have placed my comments directly on the manuscript. 

Line 11: Please remove the partitions in the abstract and re-write it completely. Check on the journals website on how the abstract should be written.

We have rewritten the abstract to adhere to the journal guidelines.

Line 40: Please cite the recent work of Reyes et al who also studied aquacultute impact using biomarkers

Reyes, V.P.; Ventura, M.A.; Amarillo, P.B. Ecotoxicological Assessment of Water and Sediment in Areas of Taal Lake with Heavy Aquaculture Practices Using Allium Cepa and Daphnia Magna Assay. Philipp J. Sci. 2022, 151, 969–974.

This was not done. The suggested reference was very focused on contrasting areas with and without fish aquaculture using two assays whereas the cited references focus on identifying threats or stressors to marine and coastal areas to put our work in context. Clearly, if the reviewer insists, we are open to citing further relevant publications. We, in fact, replaced one of the older references with an updated one that was more general in terms of identifying general stressors in the marine environment.

Lines 54-58: Please re-write this sentence.

This was done, as suggested.

Lines 75-77: This sentence (we suggest that consideration of both sediment macroinvertebrates and mussel sock macroinvertebrates may provide a more holistic understanding of the effect of bivalve aquaculture on “benthic” communities) should be in the conclusion section.

We agree and have added text to this point in the conclusion (lines 563-568).

Lines 80-83: This (We predict that the inclusion of the fraction of the benthic macroinvertebrate community associated with suspended cultured bivalves in the calculation of macrobenthic community parameters will change such relationships to show either neutral or positive influences of suspended bivalve culture on macrobenthic community parameters.) should be re-phrased and be included in the objective of the study.

We agree and have reworded the objectives to reflect this (lines 111-115).

Lines 106-108: . This sentence (We also thus predict that the additional biogenic structure due to the presence of S. clava further augments the influence of suspended bivalve culture on macrobenthic community parameters, relative to benthic communities associated with mussel socks that lack this additional structure) should be re-phrased or put in as objective of the study.

We agree and have reworded the objectives to reflect this (lines 118-121).

Lines 159-162: Remove this (“Control sites in all studied bays were of similar depth to mussel farm sites (4.58 m ±1.42 SD inside vs. 4.52 m ±1.64 SD outside of mussel leases, respectively) and were at least 300 m from aquaculture sites, thus ensuring that they would not be directly influenced by sedimentation from mussel aquaculture [45-47].”) in materials and methods, put in discussion.

We have not done this as we feel that it is important to the methods to indicate that the control sites are, other than not having aquaculture operations, similar to the areas with mussel farms.

Lines 188-199: Should be in result or discussion: “The mean length of mussels per subsample ranged from 49.1-67.6 mm (mean = 56.7 ± 9.6 mm, n = 1116), excluding mussels shorter than approximately 3 cm which were considered to be secondary set (i.e. to have settled on the socks). Styela clava collected in mussel sock samples were up to 145 mm long. When present, the total number and biomass of S. clava ranged from 25 to 658 individuals and 1.92 to 334.83 g per sample (Table 2), respectively, although biomass was likely underestimated because frozen S. clava became dehydrated after thawing. Although adult S. clava were not observed in Marchwater, 112 small S. clava individuals were observed in one sample upon processing (total biomass = 1.24 g). Only socks containing S. clava were sampled in bays infested by this species. Water depths for individual sediment samples were estimated from hydrographic charts using GPS points taken in the field. Depths of mussel sock samples were estimated from their relative positions on the socks.”

We have not done most of this as we believe that this information is relevant to describing the environment in which the study was done and the relevant assumptions. We did remove reference to Styela clava biomass being reduced as this is moot as this data is not included in the calculation of any of the indices.

Line 402: Which figure (s) or table (s) can show this (“found significantly lower abundances of sediment macroinvertebrates”) result?

We have modified the text to refer to the appropriate figure.

Line 567: Is this (“see above”) necessary?

No, this is not necessary. We have deleted this text.

Line 570. Conclusion should not contain any references. Should be the summary of your study and the future direction of this research.

Not sure what to do here. Many papers have references in the conclusions to show how the findings advance ideas on a given subject. This includes 1 of the 4 papers already published in this special issue of Water. Some direction from the editor would be of value here.

Author Response File: Author Response.pdf

Reviewer 3 Report

The manuscript ‘Influence of suspended mussel aquaculture and an associated invasive ascidian on benthic macroinvertebrate communities’ is a potential valuable contribution dealing with two common disturbances on marine ecosystems (invasions and aquaculture). 

The topic addressed by the authors is interesting particularly being in mind the novelty of including the macrofaunal assemblages associated with mussel socks to evaluate the effects of mussel farming on benthic assemblages.

 My major concern is about statistical analyses. Authors explain that ‘Variation in macroinvertebrate abundance, biomass and productivity was analyzed by ANOVA’ however, these analyses are based on an unbalanced design because they have 3 bays invaded with S. clava and 5 without Styela clava, in this way they are violating ANOVA assumptions and these analyses are incorrect (Please check Underwood, 1997. Experiments in Ecology: Their Logical Design and Interpretation Using Analysis of Variances). The best option would be to include only 3 bays without Styela clava to meet with ANOVA assumptions. Other option could be using PERMANOVA that can be based on unbalanced designs. However, authors should consider that even with PERMANOVA the unbalanced design is conditioning the results.

Another weak point is stress values of MDS representations. According to different manuals (for instance PRIMER) stress values near 0.3 could indicate that the points are close to being randomly placed, such as in figure 4. Moreover, only stress values < 0.1 corresponds to a good ordination. However, stress values on figure 6 ranged between 0.16 and 0.20. Authors should consider stress values. 

Authors have discussed some limitations of the study for instance, the potential effect of the higher level of taxonomic resolution used. According to material and methods, mussel farming has started on different dates in some studied bays. Authors could discuss the potential influence of differences on the beginning of mussel farming at different bays on their results.

Author Response

The manuscript ‘Influence of suspended mussel aquaculture and an associated invasive ascidian on benthic macroinvertebrate communities’ is a potential valuable contribution dealing with two common disturbances on marine ecosystems (invasions and aquaculture). 

The topic addressed by the authors is interesting particularly being in mind the novelty of including the macrofaunal assemblages associated with mussel socks to evaluate the effects of mussel farming on benthic assemblages.

Thank you for your words of encouragement.

 My major concern is about statistical analyses. Authors explain that ‘Variation in macroinvertebrate abundance, biomass and productivity was analyzed by ANOVA’ however, these analyses are based on an unbalanced design because they have 3 bays invaded with S. clava and 5 without Styela clava, in this way they are violating ANOVA assumptions and these analyses are incorrect (Please check Underwood, 1997. Experiments in Ecology: Their Logical Design and Interpretation Using Analysis of Variances). The best option would be to include only 3 bays without Styela clava to meet with ANOVA assumptions. Other option could be using PERMANOVA that can be based on unbalanced designs. However, authors should consider that even with PERMANOVA the unbalanced design is conditioning the results.

With respect to “violating assumptions” according to Underwood (1997), we agree that this is the case. That said, Underwood (1997) has a very restrictive sense of what is acceptable and not acceptable with respect to statistical analyses and the notion of balance in experimental design (indeed, the software he developed – GMAV – will only accept fully balanced designs). This is in contrast to the vast number of works that provide methods to handle such unbalanced designs (e.g. Quinn GP, Keough MJ (2002) Experimental Design and Data Analysis for Biologists. Cambridge University Press, Cambridge). Also, we did all our analyses and experimental design under the supervision of Gaetan Daigle in the Department of Mathematics & Statistics at Université Laval. When we did this experiment, there were only 3 bays with established Styela clava infestations and thus that was the maximum number of bays that all we choose to include in our sampling design. Yes, we could use PERMANOVA but the underlying assumptions with respect to issues for unbalanced ANOVA designs as outlined by Underwood (1997) also hold true for using PERMANOVA. We have thus kept our analyses as presented with the unbalanced experimental design.

Another weak point is stress values of MDS representations. According to different manuals (for instance PRIMER) stress values near 0.3 could indicate that the points are close to being randomly placed, such as in figure 4. Moreover, only stress values < 0.1 corresponds to a good ordination. However, stress values on figure 6 ranged between 0.16 and 0.20. Authors should consider stress values. 

With respect to stress values approaching 0.20, Dexter et al. (Dexter E, Rollwagen‐Bollens G, Bollens SM (2018) The trouble with stress: A flexible method for the evaluation of nonmetric multidimensional scaling. Limnol Oceanogr Methods 16:434-443) point out that Clarke (Clarke KR (1993) Non-parametric multivariate analyses of changes in community structure. Aust J Ecol 18:117-143), who’s work led to development of the multivariate approach used in the PRIMER suite of programs, gives an explicit warning about an over-reliance on stress values. Moreover, in conversations I have had with Dr Clarke in the past (he was a visiting fellow where I did my postdoctoral studies – AJ Underwood’s lab), he was quite emphatic that the various cut-offs that he suggested were for use as guides only and that common sense had to be used such that if the expected patterns were evident (e.g. in the present case – that infaunal samples under mussel lines differ from those distant from farm sites in the same embayment), then interpretations are likely justified. This is particularly true given that the guidelines were developed for fairly small sample sizes and stress increases with sample size (Dexter et al. 2018). Moreover, Sturrock and Rocha (Sturrock K, Rocha J (2000) A multidimensional scaling stress evaluation table. Field Methods 12:49-60) provide estimates of the distribution of critical stress values that arise from unstructured data, given a specific sample size and dimensionality, suggesting that meaningful interpretations may be derived from MDS plots with stress values < 0.388 for n=80 samples. We thus suggest that the stress values as indicated in our figures do not indicate that the points on the plots “are close to being randomly placed.”

Authors have discussed some limitations of the study for instance, the potential effect of the higher level of taxonomic resolution used. According to material and methods, mussel farming has started on different dates in some studied bays. Authors could discuss the potential influence of differences on the beginning of mussel farming at different bays on their results.

We have added text on this (lines 443-447). Related text on lines 434-436 also discusses the time since start of farming operations.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

The comments and suggestions were addressed.

Author Response

According to the handling editor (Allison Wu), "You have already addressed comments of reviewer 2's report 1, and he said "The comments and suggestions were addressed." in report 2. So you don't need to revise according to reviewer 2 again."