Round 1

Reviewer 1 Report

Overview:

Zupan and colleagues describe  The authors emphasize that ecological succession is well-documented in other systems, but this process is insufficiently documented in fouling communities. So, the authors document change in the macro fouling community at two wind farms in the north sea over 11 years. The authors outline the expectation that, initially, random taxa will comprise the community. However, the authors also expect that the communities will eventually converge and stabilize, indicating the climax community. The authors found the opposite, communities diverged and did not stabilize within 11 years.

 

General impressions:

Zupan and colleagues' paper represents a valuable contribution to an important subject: multiple groups aim to evaluate the fouling communities on offshore wind farms, and this paper adds valuable empirical context. The paper is well-written and organized and includes a nice review of the literature on wind farms and general ecology.

 

Comments:

The paper includes many valuable figures, yet it could help if there was an additional figure illustrating the empirical change in community composition. This figure could help readers interpret existing figures, e.g., figures 2, and 3. The authors do show community change through time, at a high taxonomic level in figure 6, but the analysis (table 2, figure 2,3) is done at a different taxonomic level, so an additional (or replacement) figure would help readers intuitively interpret the model results. The authors also illustrate change by taxa in figure S4, but this is only for a subset of the taxa. The authors might consider a figure similar to Fig 2 in Nat Commun 12, 6402 (2021). https://doi.org/10.1038/s41467-021-26543-x

 

Similarly, it could help readers if the authors include a figure that empirically shows the change in richness and diversity through time. The authors show the results of the GAM in figure 4, but it would be helpful to see the empirical data, either in a separate figure or behind the curves in figure 4.

 

The authors use multiple diagnostics tests, it would help the reader if the authors could outline the logic for using each test: do the results of one motivate the use of another? For example, do they use the GLM (table 2) to test the significance of the main effects? And then only if the main effects were significant, do they probe the importance of each predictor (figure 2) and then the direction of change (figure 3).

 

The authors nicely motivate the first set of analyses for community composition, as a means to identify convergence. The authors also nicely motivate the inclusion of specific biological predictor variables (foundation species, keystones, etc.). However, it is not clear why the authors use these biological predictor variables to understand changes in richness and diversity, instead of continuing to focus on changes in community composition. Overall, I think the paper would benefit from a clearer motivation for evaluating changes in richness and diversity. Do the authors assess richness and diversity because these variables indicate something about succession? Or, because the community did not reach a climax community, and richness and diversity are useful tools for interpreting the continual change?

 

Overall, the authors discuss the results within the context of the literature. Nevertheless, I think the discussion would benefit from a broader discussion of the main take-home message: the community never reaches a stable climax community. What are the implications of not achieving a stable community? e.g., for scientific interpretation, future management or regulatory decisions, conservation, etc. Also, can the authors identify other ecological examples where communities fail to reach a stable climax? For these other situations, what were the implications? e.g., for scientific interpretation, future management or regulatory decisions, conservation, etc.

 

Minor comments

Line 221  - is this referring to the GLM in the previous paragraph? Not clear as written. Move this statement in the paragraph above, or move the last sentence from the paragraph above here to clarify.

 

Line 240: "temporal trends in community dynamics" - the distinction from previous paragraphs is unclear as all of 2.3.1 focuses asses change through time.

 

Line 349 Font size changes for "The" 

Line 547  Font size changes

Author Response

Please see the attachment. 

Author Response File: Author Response.pdf

Reviewer 2 Report

It is an interesting research that after some methodological and analitical corrections should be considered. I made several comments and corrections at the attached revisedfile.

The main ones are:

1. You have two counting techniques (solitary and colonial species) but you just considered the solitary in the analyses. It is very well referenced that colonial species are usually dominat as hydrozoans, bryozoans, tunicates, etc. How can you consider just their presence in all analyzes? How did you compute presence in mathematical/numerical analyses? This is not right and definitely interfered in all your results and consequently in the discussion and conclusion. This is why we usually treat all these different organisms as coverage percentage. If you cannot do this, I suggest to change the title and highlight that it is solitary fouling species.

2. The dominat species (J. herdman) represents > 75% in both OWFs, so it seems that they converged, no? Please, review this.

3. 50 kms offshore means open sea; thus a scarcity of benthic larvae is expected no? It seems that the sedentary species that were sampled come from the sea bottom. You need to discuss how sampling design might influence your results.

4. The OWFs need to be better described as total depth, local current velocity, visibility and chlorophyll at the sampling area. All these data might and will influence the results.

5. At the discussion section, you don´t even mention anything about the colonial species. You need to review this (in the analitical methods, in your results and in the discussion sections).

6. At the discussion section you need to include a paragraph with a self- criticism related to the flaws of the manuscript and how this could have influenced your results.

Comments for author File: Comments.pdf

Author Response

Please see the attachment. 

Author Response File: Author Response.pdf

Reviewer 3 Report

The aims presented in this manuscript are too ambitious considering the data available. The data is based on a monitoring program of two offshore wind farms, which was collected for other purposes than the aim of the manuscript, although monitoring data can be very useful for long-term studies.

Furthermore, the definitions of the different roles of the macroinvertebrates in the community are confusing. The authors have misunderstood the concept of keystone species – which are species present in low densities with disproportionally large effects on the community. Some top predators are keystone species that in their absence can have a cascading effect down the food web, and in a way change completely the community. Well known examples are the wolfs from Yellowstone park or the sea otter removal from a kelp ecosystem. Some sea stars have proved to be keystone species, such as the classical Paine experiments of Pisaster removal from an intertidal rocky shore.

However, if a predator is present in large densities, it is no more a keystone species. Eventually, it can temporarily modify the other species' abundances and diversity because it is an opportunist species. Or a pest! But will not originate a highly diversified and stable community.

The authors have considered the sea star Asterias rubens and the sea urchin Psammechinus miliaris as keystone predators. Asterias rubens should be in fact considered a pest. It is able to decimate a mollusc production in a few months. Psammechinus miliaris is an interesting species that can help to reduce fouling organisms’ density. Should it be considered a keystone species?

Analyzing the raw data, the variability among the small square samples taken is huge. Furthermore, the sampling periodicity is annual. Most benthic invertebrates have short life histories, they are able to rapidly cover and replace other species. This means that it is not possible to detect any effect of a possible keystone species in the fouling community with the monitoring sampling design carried out. The possible effect of a predator in the community cannot be detected. If by chance a predator is present in high density (not a keystone species), it can decimate the preferential preys in some days, and then move to another area. In one year, the previously decimated community will have recovered and the effects of a predator will not be detected by the sampling.

According to the raw data, Psammechinus miliaris highest density was 1440 /m2. That sampled square was also covered by 48000/m2 Jassa amphipods and 688 Asterias rubens /m2 together with more 7000 organisms /m2 belonging to 28 other taxa. Can P.miliaris and A.rubens fill the concept of keystone species? Or should they be considered opportunistic predators? What would happen to the fouling community if one or both species are not present? That can only be answered by controlled testing experiments.

The increase in species richness with the increase of Asterias rubens although significant is very low (from around 18 taxa to around 28 taxa) (Figure 5), and please notice the high variability in species richness when no Asterias rubens are present, ranging from 1 to around 35 taxa (much higher than with 20 or 25 five sea stars present in 0.0625 /m2). Can this be indicative of a Keystone species? I do not think so. (I can’t understand and interpret the YAxes scale of Shannon diversity for Asterias rubens).

Nevertheless, the statistical data analysis is interesting but perhaps the authors can change the theoretical approach only to stable or unstable communities and work around this concept as well as on the opportunist behavior of some species, including opportunistic predators such as Asterias rubens and the importance of foundation species or ecosystem engineers or dominant species.

Author Response

Please see the attachment. 

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

I just suggest the title: Long-term fouling succession on offshore wind farms ...

In the text (abstract also) you don´t have to say long-term time serie. Just long-term series

Author Response

Dear Reviewer 2,

Thank you again for your comments on our manuscript diversity-2052692, initially entitled “Long term succession on artificial reefs and the role of species interactions: Foundation species, Keystone Predators and Space Occupiers”, for publication in Diversity. Based on the suggestions from Reviewer 3 and Reviewer 2, we have decided to change the title once again to “Long term succession on offshore wind farms and the role of species interactions”.

Below, we have added our answers (in italics) to your  comments (regular font).

We would like to thank you once again for the detailed and constructive comments.

Sincerely,

on behalf of all the authors,

Mirta Zupan

 

Reviewer 2 comments

I just suggest the title: Long-term fouling succession on offshore wind farms ...

In the text (abstract also) you don´t have to say long-term time serie. Just long-term series

Based on your suggestions, we have decided to change the title to ‘Long term succession on artificial reefs and the role of species interactions.’

Through the text, we have used ‘long-term series’ instead of ‘long-term time series’. 

Reviewer 3 Report

The new version of the paper is improved, but there are still some imprecisions, and clarifications are needed.

 Lines 102 -105.

It is necessary to distinguish between top predators and keystone species. One possibility is to change the text to "Low abundance top predators may increase species diversity and behave as keystone species, but when they become dominant ....

Table 1. Improve this table. It is a mess!

Horizontal lines are not necessary, remove all. Remove the lines "All turbines combined". That information can easily be gathered from the other lines of the table. Keep the “Total” - but remove "Grand". The Cell with "Offshore Wind farm" (and all the line) can be removed, it is already in the caption.

 Lines 224-230

The "non-contable species" might have had a high cover percentage and dominance!

Line 531 – please remove “keystone” – You cannot conclude you have any keystone species”

Line 576 – replace “but” by “and”

Lines 575-576. The abundance might be relatively low when compared with the extremely high abundances of the small amphipod Jassa and other taxa. But if biomasses have been determined, the relative biomass would be much higher, which changes the possible status of the sea star A. rubens to dominate / pest species instead of a candidate to a "keystone" predator.

Lines 586-589 How can you conclude that keystone predators (which keystone species?) were found “in lower abundances in offshore communities compared to natural rocky reef communities” ? The cited references present most of their data in percentage cover. A. rubens in you study presented densities up to 2400 individuals m-2. Can you consider this a low abundance?

589 – I do not understand the “However”.

Author Response

Please see attachment

Author Response File: Author Response.docx