Water Quality Indicators in Three Surface Hydraulic Connection Conditions in Tropical Floodplain Lakes

Round 1

Reviewer 1 Report

This is a meaningful study. However, the language expression of the manuscript should be further improved, especially the abstract of the manuscript. Abstract is a high generalization of the whole research, including but not limited to the research purpose, research methods and research results. In addition to the above problems, the manuscript also has the following problems that need to be significantly revised.

 

1. Line 9-14: The statements in this section are confusing and unreadable.

 

2. Abstract: When the indicator first appears, it must be the full name, not the abbreviation.

 

3. The "Results" section of the manuscript should include secondary headings to distinguish between the different results you obtained during the data analysis. Similarly, there is too much content in the "discussion" section, which needs to be cut down, and secondary headings should be added to this section as well.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

An interesting paper overall.

I felt the wording of the introduction could be improved.  I am not a domain expert - I've focussed on the statistical modelling/data analytics - but I thought there was an interesting and thorough literature review.  However, the presentation of the points made is a bit awkward at times and reads as though papers are being summarised but without all information.  For example, line 4 '...residence time were relevant'.  Here you're trying to introduce general points to the study and so I think it should be 'are relevant'. Similarly, in line 5 'which' needs removed.  These are just examples, this occurs throughout the introduction.

I think the final paragraph in the introduction with the hypothesis would benefit from re-writing/further explanation.  It's not clear to me what 'overlap' means here and the sentence is too long and so reads awkwardly - suggest breaking up a little and providing more detail.

Section 2.3

In line 190 remove 'was'

In line 193, you have 'test (p<0.05)' - I don't think this is clear do you mean that testing was done at 5% statistical significance?  I think it would be clearer to state that.

You need to explain why you have taken a log transform e.g. 'the original distributions of the variables were skewed and hence a natural log transform was applied which also resulted in linear relationships between variables'  The latter point is required for Pearson correlation to be appropriate.

In line 199 you have p>0.0531 and conclude statistically significant.  If we round to 2 decimal places we would have 0.05, which by your rule of >0.05 for sig would then be not statistically significant, and so it would be more accurate to say the p-value is borderline.

Section 2.4 - 'applied to the dataset..'  it would be helpful to be clearer here to make it clear which variables are in the PCA - it's not clear initially that SZ and ST are not in there.  The PCA statement would benefit from further explanation e.g. why log (x+1) - only natural log was mentioned previously and why standardise data when the correlation matrix is used?

In this section and the results, eigenvalues and eigenvectors are referred to.  I think the readers might find this more confusing.  I think it would be clearer to just refer to % variance explained and the PC loadings throughout methods and results.

The paragraph about the 'proportionality overlap' would benefit from further explanation and maybe some examples.  I didn't understand this I'm afraid.  

 

Author Response

Please see the attachment.

Author Response File: Author Response.docx