A Bayesian Approach to Carrying Capacity Estimate: The Case of Greek Coastal Cage Aquaculture

Round 1

Reviewer 1 Report

Please see the attachment.

Comments for author File: Comments.pdf

Author Response

Response to reviewer No 1

based on the document jmse-1707966-review.pdf

1. All language comments were changed and corrected as stated by the evaluator.
2. For specific more complicated comments:

Comment 7. The paragraph was rewritten and 2 references added

Comment 15. A new map is added

Comment 26. The sentence was removed. The information provided in the paragraph is not changed with this removal

Comment 28. The sentences was rewritten in a simpler form.

3. Language was checked again and syntax/grammar corrected.

Dr Alexis Conides

Reviewer 2 Report

In general, the manuscript is very well written and very interesting. I also think that the manuscript is suitable for publication with some minor corrections. 

In the introduction, at Lines 48-50, the role of the anthropogenic effects on these habitats should be also mentioned. I recommend reading and citing the following article: https://onlinelibrary.wiley.com/doi/10.1111/azo.12310

Line 176: Authors should rephrase this sentence. They considered the correlations as significant according to the Spearman value. The signifance can be considered only according to the p-value. What the authors mean is "relevant correlation".

Also, authors why used Spearman's rank correlation without any normality test? Authors should perform before a normality test (e.g. Shapiro-Wilk test). If the data are normally distributed then they have to use Pearson correlation. 

Lines 207-213: Add the p-values too. 

Table 5: I recommend mentioning only Spearman values which were significant. Non-significant cases should be mentioned only as non-significant (n.s.)

Figure 4-5: Rephrase the tile of figures because these are not relationships. 

Line 423: Authors mentioned ANOVA, but they did not mention if any normality test was performed. Please perform a normality test on the data, write the results of it. And if they are not normally distributed then use Kruskal-Wallis test instead of ANOVA. 

Author Response

1. In the introduction, at Lines 48-50, the role of the anthropogenic effects on these habitats should be also mentioned. I recommend reading and citing the following article: https://onlinelibrary.wiley.com/doi/10.1111/azo.12310

We are not sure whether the recommended article is related to the issue:

Yancheva, V, Georgieva, E, Stoyanova, S, et al. A histopathological study on the Caucasian dwarf goby from an anthropogenically loaded site in Hungary using multiple tissues analyses. Acta Zool. 2020; 101: 431– 446. https://doi.org/10.1111/azo.12310

However, we added the lines 50-65 to cover this comment plus new references.

2. Line 176: Authors should rephrase this sentence. They considered the correlations as significant according to the Spearman value. The significance can be considered only according to the p-value. What the authors mean is "relevant correlation".

Also, authors why used Spearman's rank correlation without any normality test? Authors should perform before a normality test (e.g. Shapiro-Wilk test). If the data are normally distributed then they have to use Pearson correlation.

1. the term relevant was included in the text.
2. Shapiro Wilk test was used to test the normality of all parameters. Since almost half of them show p values less than 0.05, the hypothesis f normality is rejected and therefore, to allow the testing of all parameter pairs, the Spearman test was used. A clear description of the test selection was also added in Materials and Methods as it should be. Actually, in terms of normality tests, used 4 different methods: Shapiro-Wilk, Anderson-Darling; Martinez-Iglewicz, and Kolmogorov-Smirnov to ensure a proper analysis.

Lines 207-213: Add the p-values too.

Recalculated because we found mistakes in the analysis and corrected. p values are added

Table 5: I recommend mentioning only Spearman values which were significant. Non-significant cases should be mentioned only as non-significant (n.s.)

Done

Figure 4-5: Rephrase the tile of figures because these are not relationships.

Done

Line 423: Authors mentioned ANOVA, but they did not mention if any normality test was performed. Please perform a normality test on the data, write the results of it. And if they are not normally distributed then use Kruskal-Wallis test instead of ANOVA.

Shapiro Wilk was performed and found linearity; hence the ANOVA analysis is correct according to the reviewer. All information has been included in the parenthesis.

Author Response File: Author Response.pdf