Assessing the Effectiveness of 3D-Printed Ceramic Structures for Coral Restoration: Growth, Survivorship, and Biodiversity Using Visual Surveys and eDNA

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This is an interesting study offering a new approach for coral reef restoration using 3-D printed ceramic tiles with complex structure. It is definitely worth to be presented for scientific audience and I believe the quality of the manuscript allow to accept it as it is without any additions.  However, the described novel approach it is just one more technique of artificial coral culture in addition to more than dozen such methods known to date, including such well-known worldwide technologies as ReefBall, ReefRock, Reefscapers, etc. One of the limitations of the presented approach is the necessity of corresponding technical support for production of such 3-D tiles.  It is not realistic for many coral reef restoration projects. Authors also decided to consider dynamics of all possible levels of coral ecosystem including corals themselves, indicator functional groups of fish and invertebrates and even ASVs and incorporate all data into single article, although ASVs and eDNA application could be considered as a separate study and publication.  Anyway, I believe the authors succeeded in presenting their results in full and wish them further success in their studies.   

Author Response

Comment 1: This is an interesting study offering a new approach for coral reef restoration using 3-D printed ceramic tiles with complex structure. It is definitely worth to be presented for scientific audience and I believe the quality of the manuscript allow to accept it as it is without any additions.  However, the described novel approach it is just one more technique of artificial coral culture in addition to more than dozen such methods known to date, including such well-known worldwide technologies as ReefBall, ReefRock, Reefscapers, etc. One of the limitations of the presented approach is the necessity of corresponding technical support for production of such 3-D tiles.  It is not realistic for many coral reef restoration projects. Authors also decided to consider dynamics of all possible levels of coral ecosystem including corals themselves, indicator functional groups of fish and invertebrates and even ASVs and incorporate all data into single article, although ASVs and eDNA application could be considered as a separate study and publication.  Anyway, I believe the authors succeeded in presenting their results in full and wish them further success in their studies.   

Response 1: Thank you for your positive reply and timely feedback. We are happy to hear that you appreciate both the eDNA dataset and traditional growth and visual survey data being presented as one manuscript. In light of your comment on the technical requirements for this restoration approach, we have de-emphasized “scalability” within the introduction and discussion of the paper (removing all but one reference to scalability, which is in the context of the tiles having a modular design on line 58). This change has been made in order to keep the focus of this manuscript on the measured ecological outcomes and approach to monitoring a restoration project. Thank you again for your time!

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you for the opportunity to review this innovative and well-executed manuscript. The study presents a compelling and methodologically rigorous approach to coral reef restoration using ceramic 3D-printed artificial reef tiles. It integrates ecological engineering, coral transplantation, and biodiversity monitoring through both traditional visual surveys and environmental DNA (eDNA) metabarcoding. The manuscript is timely, aligns well with the scope of JMSE, and contributes meaningfully to the advancement of ecosystem-based reef rehabilitation.

Below is a detailed review organized by manuscript sections, followed by specific suggestions to enhance clarity, analytical rigor, and ecological interpretation.

Introduction

The Introduction is conceptually strong, well-researched, and clearly structured. It presents a compelling rationale for coral reef restoration using artificial reef (AR) technologies, particularly ceramic 3D-printed tiles. The articulation of the global coral reef crisis is effective, and the integration of eDNA with traditional biodiversity surveys is cutting-edge.

Suggestions for Improvement:

• Explicit Research Questions and Hypotheses:
  Consider adding a paragraph that clearly outlines the study’s guiding question and hypotheses:
  Research Question: Can ceramic 3D-printed reef tiles support scalable coral restoration and enhance biodiversity?
  Hypotheses: (1) Ceramic tiles will promote recruitment of cryptic taxa; (2) Tiles will improve ecological function compared to conventional ARs.
• Clarify Study Scope and Transition:
  Revise the final paragraph to better articulate the study’s scope and provide a smoother transition to the Materials and Methods section:
  “In this study, we tested the ecological performance of ceramic 3D-printed reef tiles deployed in a subtropical, urbanized marine environment. We assessed coral survivorship, community reassembly, and biodiversity enhancement using a combination of visual surveys and eDNA metabarcoding. This approach aims to provide a replicable model for ecosystem-based reef rehabilitation.”

Materials and Methods

This section is impressively thorough and demonstrates a high level of ecological and engineering integration. The study site is well-described, and the rationale for its selection is ecologically justified. The biomimetic and hydrodynamic considerations in tile design enhance ecological relevance, and the monitoring protocol is rigorous.

Suggestions for Improvement and Clarification:

• Environmental Parameters:
  Consider including measurements of environmental variables (e.g., turbidity, nutrient levels, temperature) to strengthen ecological context and interpretation.
• Clarify Species Selection:
  Briefly explain the rationale for selecting Reef Check indicator species. Why are these taxa considered reliable indicators of restoration success?
• Bacillariophyta Justification:
  Define the ecological role of Bacillariophyta and justify their exclusion from comparative analyses.
• eDNA Replication and Filtering:
  Specify the number of eDNA samples per site retained after quality control filtering. Clarify how merging ASVs by site affects downstream diversity estimates—does this approach inflate richness or obscure within-site variability?
• Statistical Assumptions:
  Indicate how assumptions for ANOVA and other parametric tests were tested and handled (e.g., transformations, non-parametric alternatives).
• Beta Diversity Analysis:
  Clarify whether multivariate dispersion tests (e.g., PERMDISP) were applied to validate group differences in PERMANOVA.

Results

The Results section is rich in data and well-structured across ecological scales. It clearly demonstrates the effectiveness of ceramic reef tiles in supporting coral survivorship, growth, and biodiversity enhancement. The integration of coral performance metrics, visual bioindicator surveys, and eDNA-based community profiling provides a multidimensional view of restoration outcomes.

Suggestions for Improvement:

• Clarify Statistical Significance:
  Soften interpretation of non-significant ANOVA results:
  “Although the restored site exhibited higher mean ASV richness, these differences were not statistically significant (p > 0.1), suggesting that observed trends may reflect natural variability.”
• Ecological Implications of ASV Patterns:
  “The high proportion of site-specific ASVs suggests localized community differentiation, potentially driven by substrate type and habitat complexity.”
• Statistical Reporting Enhancements:
• Include degrees of freedom for Kruskal-Wallis and ANOVA tests.
• Clarify whether chi-square tests were adjusted for multiple comparisons.
• Confirm whether PERMDISP was used to validate PERMANOVA results.
• Interpret Bacillariophyta Dominance:
  “The dominance of Bacillariophyta, a diatom-rich phylum, likely reflects its role as a primary colonizer and contributor to benthic productivity in stable reef environments.”
• Clarify Chi-Square Value Consistency:
  If the identical chi-square value (χ² = 138.1) across tests is correct, explain why it remains unchanged despite excluding Bacillariophyta.
• Interpret Mollusca and Annelida Abundance:
  “Elevated Mollusca and Annelida abundances at the unrestored site may reflect opportunistic colonization in less structurally complex habitats.”
• Bridge Richness and Abundance Metrics:
  “To complement richness-based assessments, relative read abundance was analyzed using Hellinger transformation to capture compositional differences across sites.”

Discussion

The Discussion is well-structured and effectively synthesizes ecological outcomes. It aligns with contemporary coral restoration frameworks and is well-supported by literature. The integration of trophic structure, functional roles, and biodiversity metrics is commendable.

Suggestions for Refinement:

• Clarify Growth Metrics:
  Specify baseline and final values for “mean maximum linear extension” or reference the Results section directly.
• Synthesize Acropora Observations:
  “The observed high survivorship of Acropora, despite its susceptibility to breakage, suggests that structural fragility may not compromise long-term viability when substrate stability is maintained.”
• Strengthen Mixed Assemblage Interpretation:
  “Incorporating diverse morpho-functional groups not only enhances structural complexity but also buffers restored communities against environmental fluctuations, particularly in eutrophic urban settings.”
• Address Pavona Detachment:
  “To mitigate detachment, future designs could incorporate bioadhesives or modular plug systems to enhance substrate adherence.”
• Clarify Fish Aggregation vs. Recruitment:
  “While aggregation may initially drive increased fish abundance, long-term monitoring is needed to determine whether restored habitats support sustained recruitment and population growth.”
• Expand on Functional Redundancy:
  “Functional redundancy ensures that key ecosystem processes persist even if specific taxa decline, thereby enhancing resilience to environmental disturbances.”
• Complement eDNA Limitations:
  “To mitigate detection biases, future studies should integrate eDNA with visual surveys, baited remote underwater video (BRUV), and substrate sampling to capture cryptic and sessile taxa.”
• Frame Ecological Succession Explicitly:
  “The observed community differences likely reflect early successional stages, with ongoing recruitment and turnover expected to shape long-term assemblage trajectories.”
• Clarify Richness vs. Functional Convergence:
  “While richness was higher at the restoration site, compositional differences suggest that the restored community is functionally distinct rather than converging toward the reference assemblage.”

Conclusion

The manuscript presents a scientifically sound, methodologically innovative, and ecologically relevant contribution to coral reef restoration science. With the above refinements, the manuscript will be well-positioned for publication and will offer valuable insights into scalable, ecosystem-based reef rehabilitation strategies.

Style suggestion:

Introduction

1. Text: In response, Artificial Reefs (ARs) have emerged as one of the most widely used interventions...

Suggested: In response, artificial reefs (ARs) have emerged as widely adopted interventions...

2. Text: Yet, the effectiveness of these design choices requires empirical validation.

Suggested: However, the effectiveness of these design features requires empirical validation.

Materials and methods

3. Text: Lying just south of the Tropic of Cancer, Hong Kong supports a surprising level of coral diversity...

Suggested: Located just south of the Tropic of Cancer, Hong Kong supports unexpectedly high coral diversity...

4. Text: These dynamic conditions, combined with proximity to urbanized coastal zones, positions HHWMP...

Suggested: These dynamic conditions, combined with the proximity to urbanized coastal zones, position HHWMP...

5. Text: Each reef tile was seeded with six evenly spaced coral fragments...

Suggested: Each reef tile was seeded with six coral fragments, evenly spaced to ensure uniform coverage...

Results

6. Text: Survivorship of transplanted coral fragments remained high throughout the four-year monitoring period...

Suggested: Transplanted coral fragments exhibited high survivorship throughout the four-year monitoring period...

7. Text: Mortality and detachment rates remained low and stable over time, indicating sustained substrate stability and fragment viability...

Suggested: Low and stable mortality and detachment rates suggest sustained substrate stability and long-term fragment viability...

8. Text: Pavona had the lowest survivorship rate, with only 76% classified as healthy, 3% dead, and 21% detached...

Suggested: Pavona exhibited the lowest survivorship, with 76% of fragments classified as healthy, 3% as dead, and 21% detached...

9. Text: The Bray-Curtis distance analysis, which is calculated based on both occurrence and abundance...

Suggested: Bray–Curtis dissimilarity, incorporating both occurrence and abundance data, revealed distinct site clusters with minimal overlap..

10. Text: Bacillariophyta ASV richness was highest at the reference site (863 ASVs, 50.9% of site ASVs)...

Suggested: Bacillariophyta ASV richness peaked at the reference site (863 ASVs, 50.9% of total site ASVs)...

Discussion

11. Text: Coral restoration has two primary ecological goals, broadly: coral population enhancement, and the preservation of community biodiversity and ecosystem function.

Suggested: Coral restoration aims to achieve two primary ecological goals: enhancing coral populations and preserving community biodiversity and ecosystem function.

12. Text: This survivorship rate significantly exceeded the average of 66% reported for coral transplantation projects — a figure often considered an overestimate…

Suggested: This rate significantly exceeds the reported average of 66% for coral transplantation projects, which may be inflated due to short monitoring durations...

13. Text: Diver surveys using color indices found no signs of coral bleaching or disease.

Suggested: Diver-based assessments using standardized color indices detected no signs of bleaching or disease.

14. Text: Coral growth rates can vary considerably among species and morphologies...

Suggested: Coral growth rates vary widely across species and morphologies...

15. Text: One of the principal considerations in designing the reef tiles was to create an artificial structure that would attract and support a diverse community of marine taxa.

Suggested: A principal design goal of the reef tiles was to attract and support a diverse assemblage of marine taxa.

16. Text: Moderate densities of sea urchins and small herbivorous fishes on and near the tiles indicate promising progress...

Suggested: The presence of moderate densities of sea urchins and small herbivorous fishes suggests early functional establishment of the restored reef.

17. Original: The increased abundance of fish and macroinvertebrates at the restoration site may reflect either true enhancement of local carrying capacity or merely an aggregation...

Suggested: Elevated fish and macroinvertebrate abundances at the restoration site may reflect either genuine increases in local carrying capacity or aggregation effects...

18. Original: The rugose tile surfaces, combined with the growth of transplanted coral fragments, served to enhance the structural complexity...

Suggested: Rugose tile surfaces, together with coral growth, enhanced structural complexity and resource availability...

Author Response

Comment 1:  Thank you for the opportunity to review this innovative and well-executed manuscript. The study presents a compelling and methodologically rigorous approach to coral reef restoration using ceramic 3D-printed artificial reef tiles. It integrates ecological engineering, coral transplantation, and biodiversity monitoring through both traditional visual surveys and environmental DNA (eDNA) metabarcoding. The manuscript is timely, aligns well with the scope of JMSE, and contributes meaningfully to the advancement of ecosystem-based reef rehabilitation. Below is a detailed review organized by manuscript sections, followed by specific suggestions to enhance clarity, analytical rigor, and ecological interpretation.

Response 1: Thank you for your detailed and constructive review comments. In particular, we feel your suggested improvements to our statistical methods – such as adding a PERMDISP to the eDNA beta diversity analyses, bolstering the site descriptions with environmental data, and advice on how to more clearly communicate our results – have greatly strengthened our paper. It is clear you put time into writing out clear and direct ways to refine the text, and it was a pleasure implementing those changes. Thank you as well for the positive review, your feedback was greatly appreciated.

 

Introduction

The Introduction is conceptually strong, well-researched, and clearly structured. It presents a compelling rationale for coral reef restoration using artificial reef (AR) technologies, particularly ceramic 3D-printed tiles. The articulation of the global coral reef crisis is effective, and the integration of eDNA with traditional biodiversity surveys is cutting-edge.

Suggestions for Improvement:

Comment 2:  Explicit Research Questions and Hypotheses:
Consider adding a paragraph that clearly outlines the study’s guiding question and hypotheses:
Research Question:  Can ceramic 3D-printed reef tiles support scalable coral restoration and enhance biodiversity?
Hypotheses: (1) Ceramic tiles will promote recruitment of cryptic taxa; (2) Tiles will improve ecological function compared to conventional ARs.

Response 2: We have edited the final paragraph of the introduction to more clearly frame a research question and have added hypotheses (lines 88-100). The hypotheses have been altered from those suggested here to align more closely with what we were able to test with the collected data (“We hypothesized that deployment of the ceramic tiles would produce 1) greater fish and macroinvertebrate abundances and 2) eukaryotic eDNA communities that were richer and distinct from those detected at an unrestored seabed.”)

 

Comment 3: Clarify Study Scope and Transition:
Revise the final paragraph to better articulate the study’s scope and provide a smoother transition to the Materials and Methods section:
“In this study, we tested the ecological performance of ceramic 3D-printed reef tiles deployed in a subtropical, urbanized marine environment. We assessed coral survivorship, community reassembly, and biodiversity enhancement using a combination of visual surveys and eDNA metabarcoding. This approach aims to provide a replicable model for ecosystem-based reef rehabilitation.”

Response 3: Amended as suggested, with some modifications to integrate with our response to Comment 2. The end of the final paragraph now reads: “In this study, we tested the ecological performance of ceramic 3D-printed reef tiles deployed in a subtropical, urbanized marine environment. We assessed coral survivorship, community reassembly, and biodiversity enhancement using a combination of visual surveys and eDNA metabarcoding. This approach aims to provide a replicable model for ecosystem-based reef rehabilitation while evaluating the performance of the tiles according to the following research question: Can this engineered substrate support coral restoration and simultaneously enhance overall biodiversity? We hypothesized that deployment of the ceramic tiles would produce 1) greater fish and macroinvertebrate abundances and 2) eukaryotic eDNA communities that were richer and distinct from those detected at an unrestored seabed. In addressing this question, our broader goal is to establish a framework for effective reef rehabilitation that not only restores coral cover but also measures and promotes the recovery of diverse reef-associated communities”

 

Materials and Methods

This section is impressively thorough and demonstrates a high level of ecological and engineering integration. The study site is well-described, and the rationale for its selection is ecologically justified. The biomimetic and hydrodynamic considerations in tile design enhance ecological relevance, and the monitoring protocol is rigorous.

Suggestions for Improvement and Clarification:

Comment 4: Environmental Parameters:
Consider including measurements of environmental variables (e.g., turbidity, nutrient levels, temperature) to strengthen ecological context and interpretation.

Response 4: We have added a new reference (Geereart et al., 2021) and further descriptions of the nutrient regimes in Eastern Hong Kong, We have also supplemented the site description with average total nitrogen, turbidity, and salinity values from three nearby water quality monitoring stations (data provisioned by the Hong Kong Environmental Protection department; lines 119-136). We do not currently have any of our own direct measurements to report for the site, although we will be implementing these into our ongoing monitoring visits of the site.

 

Comment 5: Clarify Species Selection:
Briefly explain the rationale for selecting Reef Check indicator species. Why are these taxa considered reliable indicators of restoration success?

Response 5: Thank you for this comment.   Upon consideration of your comment, we have decided to avoid the language of “bioindicators” to avoid confusion (such language has now been changed to simply “fish and invertebrates” or “taxa” as appropriate). While the Reef Check organization refers to these taxa as “bioindicator” species, we were unable to find literature that clearly connects each organism to a specific and relevant (to our study/urban restoration contexts) stressor. These taxa are now described within the manuscript only in terms of the different trophic positions each represents. We have updated the methods as follows (and amended table 1 to include the diet of each taxon): “For this study taxa were selected to represent a range of trophic strategies – with groupers as high-level carnivores, wrasse and sweetlips feeding on smaller fish and macroinvertebrates, urchins as herbivorous grazers, and sea cucumbers as detritivores. All fish chosen for these surveys, as well as the black sea cucumber, are also of commercial interest (Sadovy and Cornish, 2000)”.

 

Comment 6: Bacillariophyta Justification:
Define the ecological role of Bacillariophyta and justify their exclusion from comparative analyses.

Response 6: Bacillariophyta were not excluded on an ecological basis but for quantitative reasons. We have re-written this section of the methods (lines 371-381) for clarity and have added two references. We acknowledge that this re-write now includes some content that may be better suited to the results and discussion, but feel it is relevant for justifying the statistical approach we took with the data: “Amplicon sequence variant (ASV) richness for Bacillariophyta was markedly higher across all samples (1,612 ASVs) compared to the other 34 phyla, which were represented by 1 to 389 ASVs (median = 17, mean = 111 ASVs; Figure A2). Bacillariophyta are comprised primarily of diatoms, a particularly diverse group of organisms that exhibit high levels of both speciation and intraspecific variation [Godhe and Rynearson, 2017; Julius and Theriot, 2010]. Although this high diversity is not unexpected biologically, it presents a potential source of statistical bias, potentially masking significant differences among less diverse phyla and increasing the likelihood of false negatives. To mitigate the bias resulting from uneven ASV richness...”

 

Comment 7: eDNA Replication and Filtering:
Specify the number of eDNA samples per site retained after quality control filtering. Clarify how merging ASVs by site affects downstream diversity estimates—does this approach inflate richness or obscure within-site variability?

Response 7:  We have amended the text to clarify the number of samples remaining per site: “The mock sample was then removed from the dataset: Five sample replicates for the Restoration and Reference sites and four replicates for the Unrestored sites thus remained for all downstream filtering and analysis.” Regarding the second part of the comment, samples are only merged by site for certain metrics—specifically, ASVs per site (Figure 8A), eukaryotic phylum richness by site (Figure 10), and taxonomic composition by site (Figures 11B and 11C). The majority of analyses, however, are based on unmerged sample data. For Figures 8A and 11B/C, we also present the data unmerged by site (Figure 8B and Figure 11A), allowing both diversity estimates to be visualized. We chose to include both merged and unmerged versions of these figures because merging by site can obscure within-site variability. However, we also aimed not to overemphasize within-site variability, since the prevalence filtering step—which retained only ASVs present in three samples from the same site—could itself obscure within-site heterogeneity and artificially lower richness estimates. This effect would be particularly pronounced at sites with greater heterogeneity in community composition. While this prevalence filtering approach may result in an underestimation of richness, we found that using a more conventional threshold (i.e., removing only sequences appearing in a single sample) risked greatly overestimating richness due to the inclusion of spurious ASVs produced by background sequencing errors during apparent over-sequencing, as referenced in the Methods.

 

Comment 8: Statistical Assumptions:
Indicate how assumptions for ANOVA and other parametric tests were tested and handled (e.g., transformations, non-parametric alternatives).

Response 8: Regarding the visual survey count data, the following text has been added: “As both the fish and macroinvertebrate count data exhibited non-normal distribution (Shapiro-Wilk test p < 0.001) and non-homogenous variance (Levene test p < 0.001), a non-parametric Kruskal-Wallis rank sum test was performed using the kruskal.test() function of the stats package”. See “Homogeneity of variance and normality were verified by Levene’s and Shapiro’s tests using the rstatix package (v0.7)” for how assumptions were tested for the ANOVA run on ASV richness (no further transformations/alternatives needed as data met the assumption.

 

Comment 9: Beta Diversity Analysis:
Clarify whether multivariate dispersion tests (e.g., PERMDISP) were applied to validate group differences in PERMANOVA.

Response 9: PERMDISP was not included in the original manuscript. Following your comment, we have run and incorporated the analysis. The following text has been added to the methods: “To assess whether the observed differences in community composition among sites were influenced by variations in group dispersion (heterogeneity of variances), a permutational analysis of multivariate dispersions (PERMDISP) was conducted using the permutest function of vegan.”

 

 

Results

The Results section is rich in data and well-structured across ecological scales. It clearly demonstrates the effectiveness of ceramic reef tiles in supporting coral survivorship, growth, and biodiversity enhancement. The integration of coral performance metrics, visual bioindicator surveys, and eDNA-based community profiling provides a multidimensional view of restoration outcomes.

Suggestions for Improvement:

Comment 10: Clarify Statistical Significance:
Soften interpretation of non-significant ANOVA results:
“Although the restored site exhibited higher mean ASV richness, these differences were not statistically significant (p > 0.1), suggesting that observed trends may reflect natural variability.”

Response 10: Amended as suggested

 

Comment 11: Ecological Implications of ASV Patterns:
“The high proportion of site-specific ASVs suggests localized community differentiation, potentially driven by substrate type and habitat complexity.”

Response 11: Thank you, this is a great suggestion. Amended as suggested to the relevant discussion section (4.1.2.2)

Statistical Reporting Enhancements:

Comment 12: Include degrees of freedom for Kruskal-Wallis and ANOVA tests.

Response 12: The degrees of freedom for the Kruskal-Wallis tests have been noted in parentheses next to the H value (e.g., Kruskal-Wallis H(2) = 13.6, p = 0.001..., where 2 is the degrees of freedom), the degrees of freedom for the ANOVA test are similarly formatted (e.g., (F(2, 11) = 2.225 where 2 is the degrees of freedom are 11 are the residuals).

 

Comment 13: Clarify whether chi-square tests were adjusted for multiple comparisons.

Response 13: We have now added a post-hoc multiple comparison test with Holm-Bonferroni adjusted p-values to the results. The methods section has been amended as follows: “Following the detection of significant differences in phylum richness between sites, a post-hoc multiple comparisons test was performed with Holm-Bonferroni adjusted p-values, using the chisq.posthoc.test package (Ebbert, 2025).”. The results have been amended as follows: “Bacillariophyta was the only phylum for which multiple comparisons testing indicated a significant difference in ASV richness between sites (Holm-Bonferroni adjusted p < 0.001).” (referring here to the first chi-sq test that was performed) & “Multiple comparisons testing indicated Cnidaria was the only phylum with a significant difference in richness between sites (Holm-Bonferroni adjusted p = 0.002), which contributed 17.0% to the calculated χ² value.” (later in the paragraph, when describing the second test where Bacillariophyta has been removed)

 

Comment 14: Confirm whether PERMDISP was used to validate PERMANOVA results.

Response 14: The following has been added to the results: “No significant differences in dispersion among sites were found for either ordination (PERMDISP; Jaccard distance, F(2,11) = 0.29, p = 0.75; Bray-Curtis distance, F(2,11) = 0.86, p = 0.472). This suggests that the homogeneity of variances assumption was met, and the observed differences in community composition, as detected by PERMANOVA, are likely due to differences in group centroids rather than dispersion.” Previous text relating to differences in clustering between sites has also been removed, as they were not supported by the PERMDISP results.

 

Comment 15: Interpret Bacillariophyta Dominance: “The dominance of Bacillariophyta, a diatom-rich phylum, likely reflects its role as a primary colonizer and contributor to benthic productivity in stable reef environments.”

Response 15: Amended to section 4.1.2.2 (Discussion, Cryptobiome) with minor formatting adjustments for flow, and two added references: “The dominance of Bacillariophyta, a diatom-rich phylum, particularly among samples from the restoration and reference sites, likely reflects the role of these organisms as  primary colonizers and contributors to benthic productivity in stable reef environments (Dang and Lovell et al., 2016; Virta et al., 2019)”

 

Comment 16: Clarify Chi-Square Value Consistency: If the identical chi-square value (χ² = 138.1) across tests is correct, explain why it remains unchanged despite excluding Bacillariophyta.

Response 16: Thank you for catching this mistake. We have re-run the analyses to confirm the correct values, and have amended the results: “In this analysis, a significant difference in ASV richness was similarly detected (χ² = 102.2, df = 46, p = <0.001)”

 

Comment 17: Interpret Mollusca and Annelida Abundance: “Elevated Mollusca and Annelida abundances at the unrestored site may reflect opportunistic colonization in less structurally complex habitats.”

Response 17: Amended as suggested

 

Comment 18: Bridge Richness and Abundance Metrics: “To complement richness-based assessments, relative read abundance was analyzed using Hellinger transformation to capture compositional differences across sites.”

Response 18: Amended as suggested

 

Discussion

The Discussion is well-structured and effectively synthesizes ecological outcomes. It aligns with contemporary coral restoration frameworks and is well-supported by literature. The integration of trophic structure, functional roles, and biodiversity metrics is commendable.

Suggestions for Refinement:

Comment 19: Clarify Growth Metrics:
Specify baseline and final values for “mean maximum linear extension” or reference the Results section directly.

Response 19: Referenced to Results Fig. 5 in the discussion (line 741).

 

Comment 20: Synthesize Acropora Observations: “The observed high survivorship of Acropora, despite its susceptibility to breakage, suggests that structural fragility may not compromise long-term viability when substrate stability is maintained.”

Response 20: Amended as suggested.

 

Comment 21: Strengthen Mixed Assemblage Interpretation: “Incorporating diverse morpho-functional groups not only enhances structural complexity but also buffers restored communities against environmental fluctuations, particularly in eutrophic urban settings.”

Response 21: Amended as suggested.

 

Comment 22: Address Pavona Detachment: “To mitigate detachment, future designs could incorporate bioadhesives or modular plug systems to enhance substrate adherence.”

Response 22: Amended as suggested, with added bioadhesive reference from Liao et al. 2025.

 

Comment 23: Clarify Fish Aggregation vs. Recruitment: “While aggregation may initially drive increased fish abundance, long-term monitoring is needed to determine whether restored habitats support sustained recruitment and population growth.”

Response 23: Amended as suggested.

 

Comment 24: Expand on Functional Redundancy: “Functional redundancy ensures that key ecosystem processes persist even if specific taxa decline, thereby enhancing resilience to environmental disturbances.”

Response 24: Amended as suggested.

 

Comment 25: Complement eDNA Limitations: “To mitigate detection biases, future studies should integrate eDNA with visual surveys, baited remote underwater video (BRUV), and substrate sampling to capture cryptic and sessile taxa.”

Response 25: Amended as suggested

 

Comment 26: Frame Ecological Succession Explicitly:
“The observed community differences likely reflect early successional stages, with ongoing recruitment and turnover expected to shape long-term assemblage trajectories.”

Response 26: Amended as suggested, with minor modification for clarity: “The observed communities – particularly those sampled from the restoration site – likely reflect early successional stages, with ongoing recruitment and turnover expected to shape long-term assemblage trajectories.”

 

Comment 27: Clarify Richness vs. Functional Convergence:
“While richness was higher at the restoration site, compositional differences suggest that the restored community is functionally distinct rather than converging toward the reference assemblage.”

Response 27: Amended as suggested.

 

Conclusion

The manuscript presents a scientifically sound, methodologically innovative, and ecologically relevant contribution to coral reef restoration science. With the above refinements, the manuscript will be well-positioned for publication and will offer valuable insights into scalable, ecosystem-based reef rehabilitation strategies.

Style suggestion:

Introduction

Comment 28: Text: In response, Artificial Reefs (ARs) have emerged as one of the most widely used interventions...Suggested: In response, artificial reefs (ARs) have emerged as widely adopted interventions...

Response 28: Amended as suggested

 

Comment 29: Text: Yet, the effectiveness of these design choices requires empirical validation. Suggested: However, the effectiveness of these design features requires empirical validation.

Response 29: Amended as suggested

 

Materials and Methods

Comment 30: Text: Lying just south of the Tropic of Cancer, Hong Kong supports a surprising level of coral diversity...Suggested: Located just south of the Tropic of Cancer, Hong Kong supports unexpectedly high coral diversity...

Response 30: Amended as suggested

 

Comment 31: Text: These dynamic conditions, combined with proximity to urbanized coastal zones, positions HHWMP...Suggested: These dynamic conditions, combined with the proximity to urbanized coastal zones, position HHWMP...

Response 31: Amended as suggested

 

Comment 32: Text: Each reef tile was seeded with six evenly spaced coral fragments... Suggested: Each reef tile was seeded with six coral fragments, evenly spaced to ensure uniform coverage...

Response 32: Amended as suggested

 

Results

Comment 33: Text: Survivorship of transplanted coral fragments remained high throughout the four-year monitoring period...Suggested: Transplanted coral fragments exhibited high survivorship throughout the four-year monitoring period...

Response 33: Amended as suggested

 

Comment 34: Text: Mortality and detachment rates remained low and stable over time, indicating sustained substrate stability and fragment viability...Suggested: Low and stable mortality and detachment rates suggest sustained substrate stability and long-term fragment viability...

Response 34: Amended as suggested

 

Comment 35: Text: Pavona had the lowest survivorship rate, with only 76% classified as healthy, 3% dead, and 21% detached...Suggested: Pavona exhibited the lowest survivorship, with 76% of fragments classified as healthy, 3% as dead, and 21% detached...

Response 35: Amended as suggested

 

Comment 36: Text: The Bray-Curtis distance analysis, which is calculated based on both occurrence and abundance...Suggested: Bray–Curtis dissimilarity, incorporating both occurrence and abundance data, revealed distinct site clusters with minimal overlap..

Response 36: Amended as suggested

 

Comment 37: Text: Bacillariophyta ASV richness was highest at the reference site (863 ASVs, 50.9% of site ASVs)…Suggested: Bacillariophyta ASV richness peaked at the reference site (863 ASVs, 50.9% of total site ASVs)...

Response 37: Amended as suggested

 

Discussion

Comment 38: Text: Coral restoration has two primary ecological goals, broadly: coral population enhancement, and the preservation of community biodiversity and ecosystem function. Suggested: Coral restoration aims to achieve two primary ecological goals: enhancing coral populations and preserving community biodiversity and ecosystem function.

Response 38: Amended as suggested

 

Comment 39: Text: This survivorship rate significantly exceeded the average of 66% reported for coral transplantation projects — a figure often considered an overestimate…Suggested: This rate significantly exceeds the reported average of 66% for coral transplantation projects, which may be inflated due to short monitoring durations...

Response 39: Amended as suggested

 

Comment 40: Text: Diver surveys using color indices found no signs of coral bleaching or disease. Suggested: Diver-based assessments using standardized color indices detected no signs of bleaching or disease.

Response 40: Amended as suggested

 

Comment 41: Text: Coral growth rates can vary considerably among species and morphologies...Suggested: Coral growth rates vary widely across species and morphologies...

Response 41: Amended as suggested

 

Comment 42: One of the principal considerations in designing the reef tiles was to create an artificial structure that would attract and support a diverse community of marine taxa.Suggested: A principal design goal of the reef tiles was to attract and support a diverse assemblage of marine taxa.

Response 42: Amended as suggested

 

Comment 43: Moderate densities of sea urchins and small herbivorous fishes on and near the tiles indicate promising progress...Suggested: The presence of moderate densities of sea urchins and small herbivorous fishes suggests early functional establishment of the restored reef.

Response 43: Amended as suggested

 

Comment 44: The increased abundance of fish and macroinvertebrates at the restoration site may reflect either true enhancement of local carrying capacity or merely an aggregation...Suggested: Elevated fish and macroinvertebrate abundances at the restoration site may reflect either genuine increases in local carrying capacity or aggregation effects...

Response 44: Amended as suggested

 

Comment 45: The rugose tile surfaces, combined with the growth of transplanted coral fragments, served to enhance the structural complexity...Suggested: Rugose tile surfaces, together with coral growth, enhanced structural complexity and resource availability...

Response 45: Amended as suggested

Reviewer 3 Report

Comments and Suggestions for Authors

The study on the efficiency of 3D-printed ceramic structures for coral restoration is presented in the publication in a thorough and well-designed manner. A key advantage is the use of both visual surveys and eDNA metabarcoding to evaluate coral growth and survival as well as more general cryptic biodiversity. Insights into the possibility of sophisticated restoration approaches are provided by the new findings. However, in order to increase the paper's overall effect, statistical rigour, and clarity, a number of crucial parts need significant improvement.

1. The abstract has to be updated to more effectively convey the key conclusions, especially the eDNA analysis's quantitative findings (such as the precise % increase in richness). Include some statistical and numerical results in the abstract. 
2. Figure 1, prepare a georeferenced study area map.
3. Most of the references are very old, better to update them with the recent studies. 
4. More justification and explanation are required for the statistical techniques used to analyse the eDNA data. Although QIIME2 and DADA2 are commonly used, the precise filtering criteria (such as minimum reads and prevalence) has to be specified in the methods section.
5. It is important to provide the statistical tests (like ANOVA and PERMANOVA) and their findings (such p-values and F-statistics) in a consistent and comprehensive manner. The text occasionally makes reference to significance without giving the required context, and the tables (such as Table 4) lack complete statistical statistics.
6. There needs to be a more thorough examination of eDNA's drawbacks, especially with regard to absolute abundance and DNA transport capability. In their study, the writers should additionally discuss how they accounted for this.
7. The absence of a critical pre-restoration baseline for the restored site in the paper's discussion makes it challenging to distinguish between true restoration and simple aggregation. This limitation should be discussed by the writers further. Direct comparisons between the restored and natural reference sites can be deceptive if other environmental factors are not taken into account, hence a more nuanced interpretation is needed.

Comments on the Quality of English Language

A comprehensive proofreading pass should resolve the manuscript's grammatical faults and formatting irregularities.

Author Response

Comment (Overall): The study on the efficiency of 3D-printed ceramic structures for coral restoration is presented in the publication in a thorough and well-designed manner. A key advantage is the use of both visual surveys and eDNA metabarcoding to evaluate coral growth and survival as well as more general cryptic biodiversity. Insights into the possibility of sophisticated restoration approaches are provided by the new findings. However, in order to increase the paper's overall effect, statistical rigour, and clarity, a number of crucial parts need significant improvement.

Response (Overall): Thank you very much for your thoughtful and constructive feedback on our manuscript. We appreciate your recognition of the study’s strengths, particularly our use of visual surveys and eDNA metabarcoding to assess both coral growth and broader biodiversity. In response to your suggestions regarding statistical rigor and clarity, we have made substantial revisions throughout the manuscript. Specifically, we have enhanced the statistical analyses by providing more detailed descriptions of our methods, consistently reporting test statistics in relevant contexts, and clarifying the presentation of our results. Additionally, we have addressed your comments regarding grammar and inconsistent formatting by carefully proofreading the entire manuscript. We hope these improvements address your concerns and contribute to the overall impact of the paper. Thank you again for helping us strengthen our work.

 

Comment 1: The abstract has to be updated to more effectively convey the key conclusions, especially the eDNA analysis's quantitative findings (such as the precise % increase in richness). Include some statistical and numerical results in the abstract. 

Response 1: The precise % difference in eDNA ASV richness (between sites) is included in the abstract text (lines 24 and 25). These numbers reflect only the difference in richness between sites, as there is no pre-restoration baseline sample to calculate an accurate % increase in richness from (this is addressed further in our response to comment 7). Visual count survey results have been added to the abstract as follows: “Visual surveys recorded seven-times more fish and almost 60% more invertebrates at the restoration site compared to a nearby unrestored area.”

 

Comment 2: Figure 1, prepare a georeferenced study area map.

Response 2: The figure has been updated to include reference to latitude and longitude. We have also redesigned the layout of this site map to make it more clear to readers where the study sites are positioned within Hong Kong (as the small rectangle was previously somewhat difficult to spot on its own.

 

 

 

Comment 3: Most of the references are very old, better to update them with the recent studies. 

Response 3:  We have endeavored to include more recent studies where possible, and the following references have been added:

• Ahmed, Q., Ali, Q., Bat, L., Öztekin, A., Ghory, F., Shaikh, I., Qazi, H., & Baloch, A. (2023). Gut Content Analysis in Holothuria leucospilota and Holothuria cinerascens(Echinodermata: Holothuroidea: Holothuriidae) From Karachi Coast. Mater. Environ. Sci., 14 (1), 31, 40.
• Clark, D. E., Pilditch, C. A., Pearman, J. K., Ellis, J. I., & Zaiko, A. (2020). Environmental DNA metabarcoding reveals estuarine benthic community response to nutrient enrichment – Evidence from an in-situ experiment. Environmental Pollution, 267, 115472. https://doi.org/10.1016/j.envpol.2020.115472
• Couton, M., Lévêque, L., Daguin-Thiébaut, C., Comtet, T., & Viard, F. (2022). Water eDNA metabarcoding is effective in detecting non-native species in marinas, but detection errors still hinder its use for passive monitoring. Biofouling, 38(4), 367–383. https://doi.org/10.1080/08927014.2022.2075739
• Geeraert, N., Archana, A., Xu, M. N., Kao, S.-J., Baker, D. M., & Thibodeau, B. (2021). Investigating the link between Pearl River-induced eutrophication and hypoxia in Hong Kong shallow coastal waters. Science of The Total Environment, 772, 145007. https://doi.org/10.1016/j.scitotenv.2021.145007
• Ip, Y. C. A., Chang, J. J. M., Tun, K. P. P., Meier, R., & Huang, D. (2023). Multispecies environmental DNA metabarcoding sheds light on annual coral spawning events. Molecular Ecology, 32(23), 6474–6488. https://doi.org/10.1111/mec.16621
• Jeunen, G.-J., Knapp, M., Spencer, H. G., Lamare, M. D., Taylor, H. R., Stat, M., Bunce, M., & Gemmell, N. J. (2019). Environmental DNA (eDNA) metabarcoding reveals strong discrimination among diverse marine habitats connected by water movement. Molecular Ecology Resources, 19(2), 426–438. https://doi.org/10.1111/1755-0998.12982
• Joseph, C., Faiq, M. E., Li, Z., & Chen, G. (2022). Persistence and degradation dynamics of eDNA affected by environmental factors in aquatic ecosystems. Hydrobiologia, 849(19), 4119–4133. https://doi.org/10.1007/s10750-022-04959-w
• Larson, E. R., Graham, B. M., Achury, R., Coon, J. J., Daniels, M. K., Gambrell, D. K., Jonasen, K. L., King, G. D., LaRacuente, N., Perrin-Stowe, T. I., Reed, E. M., Rice, C. J., Ruzi, S. A., Thairu, M. W., Wilson, J. C., & Suarez, A. V. (2020). From eDNA to citizen science: Emerging tools for the early detection of invasive species. Frontiers in Ecology and the Environment, 18(4), 194–202. https://doi.org/10.1002/fee.2162
• Liang, D., Xia, J., Song, J., Sun, H., & Xu, W. (2022). Using eDNA to identify the dynamic evolution of multi-trophic communities under the eco-hydrological changes in river. Frontiers in Environmental Science, 10, 929541.
• Liao, H., Hu, S., Yang, H., Wang, L., Tanaka, S., Takigawa, I., Li, W., Fan, H., & Gong, J. P. (2025). Data-driven de novo design of super-adhesive hydrogels. Nature, 644(8075), 89–95. https://doi.org/10.1038/s41586-025-09269-4
• Moore, A., Ndobe, S., Ambo-Rappe, R., Jompa, J., & Yasir, I. (2019). Dietary preference of key microhabitat Diadema setosum: A step towards holistic Banggai cardinalfish conservation. 235(1), 012054.
• Schenk, J., Geisen, S., Kleinboelting, N., & Traunspurger, W. (2019). Metabarcoding data allow for reliable biomass estimates in the most abundant animals on earth. Metabarcoding and Metagenomics, 3, e46704. https://doi.org/10.3897/mbmg.3.46704
• Shelton, A. O., Gold, Z. J., Jensen, A. J., D′Agnese, E., Andruszkiewicz Allan, E., Van Cise, A., Gallego, R., Ramón-Laca, A., Garber-Yonts, M., Parsons, K., & Kelly, R. P. (2023). Toward quantitative metabarcoding. Ecology, 104(2), e3906. https://doi.org/10.1002/ecy.3906
• Skelton, J., Cauvin, A., & Hunter, M. E. (2023). Environmental DNA metabarcoding read numbers and their variability predict species abundance, but weakly in non-dominant species. Environmental DNA, 5(5), 1092–1104. https://doi.org/10.1002/edn3.355
• URycki, D. R., Kirtane, A. A., Aronoff, R., Avila, C. C., Blackman, R. C., Carraro, L., Evrard, O., Good, S. P., Hoyos J., D. C., López-Rodríguez, N., Mora, D., Schadewell, Y., Schilling, O. S., & Ceperley, N. C. (2024). A new flow path: eDNA connecting hydrology and biology. WIREs Water, 11(6), e1749. https://doi.org/10.1002/wat2.1749
• Uthicke, S., Robson, B., Doyle, J. R., Logan, M., Pratchett, M. S., & Lamare, M. (2022). Developing an effective marine eDNA monitoring: eDNA detection at pre-outbreak densities of corallivorous seastar (Acanthaster cf. Solaris). Science of The Total Environment, 851, 158143. https://doi.org/10.1016/j.scitotenv.2022.158143
• Virta, L., Gammal, J., Järnström, M., Bernard, G., Soininen, J., Norkko, J., & Norkko, A. (2019). The diversity of benthic diatoms affects ecosystem productivity in heterogeneous coastal environments. Ecology, 100(9), e02765. https://doi.org/10.1002/ecy.2765
• Xie, R., Zhao, G., Yang, J., Wang, Z., Xu, Y., Zhang, X., & Wang, Z. (2021). eDNA metabarcoding revealed differential structures of aquatic communities in a dynamic freshwater ecosystem shaped by habitat heterogeneity. Environmental Research, 201, 111602. https://doi.org/10.1016/j.envres.2021.111602
• Xiong, J., MacCready, P., Brasseale, E., Andruszkiewicz Allan, E., Ramón-Laca, A., Parsons, K. M., Shaffer, M., & Kelly, R. P. (2025). Advective Transport Drives Environmental DNA Dispersal in an Estuary. Environmental Science & Technology, 59(15), 7506–7516. https://doi.org/10.1021/acs.est.5c01286
• Zinger, L., Bonin, A., Alsos, I. G., Bálint, M., Bik, H., Boyer, F., Chariton, A. A., Creer, S., Coissac, E., Deagle, B. E., De Barba, M., Dickie, I. A., Dumbrell, A. J., Ficetola, G. F., Fierer, N., Fumagalli, L., Gilbert, M. T. P., Jarman, S., Jumpponen, A., … Taberlet, P. (2019). DNA metabarcoding—Need for robust experimental designs to draw sound ecological conclusions. Molecular Ecology, 28(8), 1857–1862. https://doi.org/10.1111/mec.15060
  

Upon review of our cited references, we found that over half have been published within the past decade (53 out of 103 cited references in the original manuscript, and 78 out of 134 in the revised manuscript). We agree with the reviewer that it would be preferable to include even more recent papers where possible. However, as discussed in section 4.2 (Discussion: Measuring Coral Restoration Outcomes), the number of publications in the field of coral restoration ecology—and similarly, the ecology of artificial structures—remains relatively limited. In many cases, we have cited older references because they represent the only relevant studies available. For example, Connell et al., 2000 and Perkol-Finkel et al., 2007 are included as examples of work demonstrating that artificial structures can produce novel assemblages distinct from those found on natural substrates, as we could not identify more recent papers directly measuring and testing such differences in subtidal marine settings.

Some older publications were included because they are seminal works (e.g., Yachi and Loreau, 1999; McGrady Steed et al., 1997; Naeem and Li, 1997) that defined key concepts we refer to, such as (in this example): “Ecological theory generally posits that elevated taxonomic richness enhances ecosystem resilience and functional stability—even if much of the diversity is functionally redundant.” In some instances, we have cited these foundational papers alongside more recent studies on the same topic (for example, Gardner et al., 2003 is cited together with Whaley et al., 2023 in reference to widespread coral losses). Our intention in including both older and newer works is to acknowledge the original sources that established key concepts or identified important patterns, while also referencing recent literature that confirms or updates our understanding of these patterns and theories. A similar approach is taken when citing Reaka, 1997 and Knowlton et al., 2010 alongside Battaglia et al., 2023 and Edwards et al., 2024.

We hope this clarifies our rationale for the inclusion of both older and recent references in the manuscript.

 

 

Comment 4: More justification and explanation are required for the statistical techniques used to analyze the eDNA data. Although QIIME2 and DADA2 are commonly used, the precise filtering criteria (such as minimum reads and prevalence) has to be specified in the methods section.

Response 4: Thank you for drawing our attention to inconsistencies in how statistical methods were reported in our manuscript. We appreciate your thorough review and agree that further clarification and justification were needed regarding the eDNA data analysis. In response, we have made extensive changes to the text, improved the reporting of our statistical techniques, and added additional analyses where appropriate to ensure transparency and rigor. We trust these revisions address your concerns and strengthen the overall presentation of our methods

Regarding the DADA2 and QIIME2 methodology:

• We have added the following clarification to explain the de-multiplexing step: “Reads were demultiplexed to separate individual samples in QIIME2.”
• Additionally, we have clarified our use of the denoise-paired function with the following statement: “No modifications were made to the default parameters of this function.” As we did not alter any default parameters, we believe that providing detailed descriptions for each step within the main text would be unnecessarily cumbersome for readers. All relevant information regarding the function’s parameters is readily accessible in the function’s documentation (https://docs.qiime2.org/2024.10/plugins/available/dada2/denoise-paired/). This level of detail aligns with common practice in studies utilizing the DADA2 denoise-paired function, as demonstrated in the following examples:
• Manandhar P, Manandhar S, Sherchan AM, Joshi J, Chaudhary HK, et al. (2023) Environmental DNA (eDNA) based fish biodiversity assessment of two Himalayan rivers of Nepal reveals diversity differences and highlights new species distribution records. PLOS Water 2(6): e0000099. https://doi.org/10.1371/journal.pwat.0000099
• Dai, S., Bai, M., Jia, H., Xian, W., & Zhang, H. (2022). An Assessment of Seasonal Differences in Fish Populations in Laizhou Bay Using Environmental DNA and Conventional Resource Survey Techniques. Fishes, 7(5), 250. https://doi.org/10.3390/fishes7050250
• Palmer, B., S. Karačić, G. Bierbaum, and C. T. Gee. 2025. Microbial methods matter: Identifying discrepancies between microbiome denoising pipelines using a leaf biofilm taphonomic dataset. Applications in Plant Sciences 13(2): e11628. https://doi.org/10.1002/aps3.11628

Regarding the phyloseq portion of our pipeline (minimum reads/prevalence):

• We have clarified the issue of over-sequencing, which underlies our justification for the prevalence filtering steps used: “apparent over-sequencing (i.e., number of unique sequences exceeded realistic biological diversity in samples [53]; Figure A1)”.
• The minimum read threshold is specified in the first draft of the manuscript and remains unchanged to avoid redundancy: “ASVs were then filtered based on abundance, removing those with a cumulative read abundance of fewer than 50 total counts across all samples (493 ASVs)”.
• We initially omitted the specific prevalence threshold in the first draft due to concerns that it might be misleading, as prevalence thresholds are conventionally applied across all samples, whereas our approach was site-specific. For example, the standard phyloseq workflow with a prevalence threshold of 0.6 would retain only sequences present in 8 out of 15 samples across all sites, rather than 3 out of 5 samples from a single site, which was our approach. The amended text clarifies both this methodological detail and our rationale: “Only ASVs that appeared in at least three samples from each site were retained: samples were grouped by site using the subset_samples function, the phyloseq_filter_prevalence function (prevalence threshold = 0.6) was then applied to each of the site subsets, and the subsets subsequently combined back into one phyloseq object.”
• Our justification for this approach is included in the text: “Stringent prevalence and abundance filters were applied to eliminate spurious ASVs produced by background sequencing errors during apparent over-sequencing (Figure A1).”
• We have also added the following clarification to our eDNA bioinformatics methods: “The mock sample was then removed from the dataset: Five sample replicates for the Restoration and Reference sites and four replicates for the Unrestored sites thus remained for all downstream filtering and analysis.”

Regarding the statistical techniques used to analyze the eDNA data, we have revised the section describing the statistical methods for differences in phylum richness by site for greater clarity and detail. The revised text now states: “Following the detection of significant differences in phylum richness between sites, a post-hoc multiple comparisons test was performed with Holm-Bonferroni adjusted p-values, using the chisq.posthoc.test package. Amplicon sequence variant (ASV) richness for Bacillariophyta was markedly higher across all samples (1,612 ASVs) compared to the other 34 phyla, which were represented by 1 to 389 ASVs (median = 17, mean = 111 ASVs; Figure A2). Bacillariophyta are comprised primarily of diatoms, a particularly diverse group of organisms that exhibit high levels of both speciation and intraspecific variation. Although this high diversity is not unexpected biologically, it presents a potential source of statistical bias, potentially masking significant differences among less diverse phyla and increasing the likelihood of false negatives. To mitigate the bias resulting from uneven ASV richness, only phyla represented by at least 10 ASVs were included in the chi-squared analysis. Two versions of the test were performed: one including Bacillariophyta, and one excluding Bacillariophyta. For both tests, phylum richness for each site was normalized to the total ASV richness of that site (excluding Bacillariophyta from the total site richness for the latter test).”

We trust that these revisions provide the necessary justification and explanation for the statistical techniques used to analyze the eDNA data, and we hope this addresses the reviewer’s concerns.

 

Comment 5: It is important to provide the statistical tests (like ANOVA and PERMANOVA) and their findings (such p-values and F-statistics) in a consistent and comprehensive manner. The text occasionally makes reference to significance without giving the required context, and the tables (such as Table 4) lack complete statistical statistics.

Response 5: Thank you for highlighting the need for consistent and comprehensive reporting of statistical results. In response, we have carefully reviewed the manuscript to ensure that all statistical tests and findings are now reported completely and consistently throughout the text. We have also revised table captions to provide more detail and included a key for all abbreviations. In addition, we have standardized the presentation of common statistical features—such as degrees of freedom, p-values, and F-statistics—across all tables and within the main text. See below for point-by-point revisions:

3.3 Fish and Invertebrate Surveys

• Changed number of significant digits of reported p-values to be consistent with text
• Changed “Test Statistic” in Table 4 to “H” to more clearly identify which statistic is being reported
• Minor changes have been made throughout the text and table captions to better communicate what is which averages/totals are being reported
• Have added the number of observations for all reported averages/SDs
• Have added clarification that Holm-Bonferroni method was used to account for multiple comparisons Wilcoxon-Signed test statistics (W) for each pairwise comparison have been added to Table 4
  We also note that there have been some minor changes to Table 3 and Figure 6; As we were re-running these stats, we discovered some

3.4.3 Community Similarity

• Amended line 582 to include F value for Jaccard PERMANOVA
• Amended line 599 to include F value for Bray-Curtis PERMANOVA
• Line 584-586, has been amended to include full results for each pairwise comparison (Jaccard)
• Lines 600-602, have amended to include full results for each pairwise comparison (Bray)
• PERMDISP analyses have been added, given the significant results of the PERMANOVA

3.4.4 Taxonomic Composition by Site: Diversity and Abundance

• Holm-Bonferroni corrections have been applied to p-values to account for multiple comparisons
• Formatting revisions have been made to ensure chi-sq. test results are presented in a consistent manner with the ANOVA/PERMANOVA noted earlier (e.g., placing degrees of freedom within parentheses next to the statistic)
• Standard deviation is now reporded consistently within text

 

Comment 6: There needs to be a more thorough examination of eDNA's drawbacks, especially with regard to absolute abundance and DNA transport capability. In their study, the writers should additionally discuss how they accounted for this.

Response 6:  Thank you for this valuable suggestion. In response, we have added a few paragraphs and shorter sections to the Discussion that more thoroughly examine the limitations of eDNA that you have mentioned. We have also included discussion of how these factors may influence our results and outlined the steps we took to account for them in our study design and data interpretation. \

Regarding absolute abundance, we have added the following to the discussion:

“Similarly, it is important to recognize that DNA surveys do not provide direct measurements of absolute organismal abundance. Read counts are affected by variability in DNA shedding rates among taxa, PCR amplification bias, and sequencing depth [106–108]. The compositional abundance data from our study do not directly reflect the absolute abundance of different phyla at the sites. However, given broadly similar sample composition and consistent processing methods, differences in the relative read abundance of the same phylum between sites may credibly reflect relative differences in their abundance between sites. Such correlations, however, are likely limited to more dominant taxa [109,110].”

Regarding transport capability, we have added the following to the discussion:

“The interpretation of our eDNA results is further complicated by the physical transport of DNA within the marine environment. Water movement can introduce exogenous DNA from adjacent or even distant habitats, resulting in the detection of taxa that are not present at the actual sampling location. Recognizing the proximity between the restored and unrestored sites, we sampled sediments – which generally exhibit lower DNA mobility compared to water samples – to minimize the influence of DNA transport  [113,114]. However, even using sediment samples, some level of DNA movement through the water column prior to and following deposition cannot be fully excluded. Bioturbation also likely affected our results: Our visual surveys documented the presence of black sea cucumbers and sweetlips fish at the study sites, both of which are known to disturb sediments, potentially resuspending and moving eDNA [115]. Such transport dynamics would promote the homogenization of eDNA community profiles between nearby sites, potentially leading to an underestimation of true ecological differentiation between the restored and unrestored sites. Ideally, future studies in systems where control and restoration sites are connected should employ sampling designs that facilitate spatial analyses of eDNA community composition, and consider using an eDNA tracer [116,117]. While both approaches are currently novel in marine settings, implementing these methods would improve the ability to characterize transport dynamics within the system.”

 

Comment 7: The absence of a critical pre-restoration baseline for the restored site in the paper's discussion makes it challenging to distinguish between true restoration and simple aggregation. This limitation should be discussed by the writers further. Direct comparisons between the restored and natural reference sites can be deceptive if other environmental factors are not taken into account, hence a more nuanced interpretation is needed.

 

Response 7:

Thank you for raising this important point regarding the absence of a pre-restoration baseline, as well as the challenges of interpreting results without fully accounting for environmental variability between sites. We agree that having a pre-restoration baseline would have strengthened our ability to distinguish between true restoration effects and simple aggregation, and to more confidently attribute observed changes to the intervention rather than to other factors. Such samples would have also allowed us to report an ”increase in ASV richness” attributed to the restoration rather than simply the difference in richness between sites. Interest in and funding to support the eDNA analyses were only granted after the commencement of the deployment, at which point is was too late to collect pre-restoration baseline samples. We have amended the following text to section 4.2.2.1. to acknowledge this limitation:

“Crucially, without a pre-restoration baseline for our study sites, it is difficult to determine how much of the measured differences in richness and community composition between the restoration and control site can be directly attributed to the restoration project. Aggregation of taxa that were already present in the area, but attracted to a novel structure, likely contributes to the higher richness and distinct community detected at the restoration site. Such an effect does not necessarily indicate restoration success. Moreover, we have assumed that these two sites are comparable due to their close proximity (50 m) and similar substrate. However, even subtle differences in environmental conditions — such as hydrodynamics, water quality, and patterns of human activity — may drive variations in eDNA community structure that predate our deployment [111,112]. This concern is particularly relevant given the high heterogeneity of urbanized marine ecosystems [99]. Ultimately, the differences between the eDNA communities detected at the restoration site and the unrestored seabed likely reflect the impact of the artificial reef deployment in combination with aggregation effects, and extrinsic biotic and abiotic variation. Lacking the measurements needed to disentangle the relative influence of these factors, we caution against over-attributing differences in the eDNA communities of these two sites to the restoration project alone.”

We have also added the following to the Discussion, in the hope of guiding future work with eDNA towards better design:

“Although processing eDNA samples can be costly, sample collection itself is straightforward and inexpensive; therefore, we recommend that those considering using eDNA to monitor restoration work collect baseline samples prior to any intervention, even if resources to support their processing are not available at the start of the project. Ideally, this baseline sampling should be combined with annual monitoring to better capture the trajectory of succession and the persistence of these patterns in the sampled eDNA communities. Collecting and archiving baseline samples provides the opportunity to conduct retrospective analyses as funding or new methods become available, which is especially important for distinguishing restoration effects from natural or pre-existing variability.”

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

Revisions are satisfactory.

However, I am worried about the similarity report. Percent match: 71% iThenticate report 

If possible, enhance the figure quality. 

Figure 1 - Add a scale bar and global map. 

Figures 4-5-6-7-8-9: Darker the legends and axes.