Evaluation of Spatiotemporal Patterns and Water Quality Conditions Using Multivariate Statistical Analysis in the Yangtze River, China
Round 1
Reviewer 1 Report
The article entitled “Evaluation of Spatiotemporal Patterns and Water Quality Conditions Using Multivariate Statistical Analysis in the Yangtze River, China” presents analyzes of 9 water quality parameters and WQI through PCA and cluster analysis, for 33 monitoring stations in the Yangtze River watershed. As an opinion, here are the main recommendations for the article:
- Keywords must be different from the words present in the title, to increase the search possibilities of the article;
- Figure 1 must be redone. What is the watershed location in China (not possible to see in the figure)? Submit an altitude map of the basin and a map of land use and occupation, with the location of stations, as these factors interfere with water quality. Furthermore, note that on the left of the image, there is a river section that is not part of the hydrographic basin, raising doubts about its delimitation. It is recommended that authors use the watershed delimitation and not the “province border”. The legend of the figure with the names of the stations can be converted into a table, which includes latitude, longitude, altitude, drainage area up to the sampling point and percentage of failures.
- What is the climatic seasonality in this hydrographic basin, mainly the rainfall levels in the four seasons of the year evaluated? Insert charts or maps that exhibit the same behavior
- The collections were monthly (Item 2.2) - is it possible to observe this only in the results? What are the laboratory procedures or methodologies used by MEE or RESDC to analyze the 9 evaluated parameters (cite methodologies, protocols – if any, and their references)?
- Present the methodologies for generating maps (Figure 3), interpolator, semivariogram and type of regression, among other details. Spatial distributions are representative when they present good distribution of points, which is not observed in the West and Southeast regions of the evaluated basin. How to justify that the maximum or minimum values always occur in these regions, and for some parameters wide ranges of variation are still observed in these regions without any sampling point. Present more details of these spatial distributions.
- Table 2 needs to be better formatted (the visualization of the values is confusing due to the alignments).
- Present graphs or a table, which indicate the annual seasonality of the 9 parameters and the IQA, with the lines of the Class I Standards (Figure 2 does not allow to verify differences). In addition, it is necessary to carry out a non-parametric statistical analysis (Kruskall Wallis test, for example) to verify the differences between the monthly averages per season and per month (as shown in Table 2 - TUR, there is a large variation between maximums and minimums for all parameters when grouped into a season)
- HCA presented distribution according to the regions of the basin (Figure 5)? The present discussion of this result (why did this happen? Is it expected for other places?)
- PCA to be representative when the reductions for PC1 and PC2, it is expected that accumulated Eigenvalue is greater than 75% (some authors indicate at least 80%). In this case, as there are only 9 parameters, it is observed that from PC3 to PC5, for most cases, each new component would be represented by 1 parameter. The authors must present these results/discussion, to justify the low Eigenvalue values obtained by PC.
- in the title of Figure 6, do A1, A2, A3 indicate the regions of the basin? This should be clear in the title. And in Figure 6d, which represents “annual”, is it observed that there are few points when compared to 6a and 6b? check these figures
- The conclusions must respond to the objective of the article. They must be clear, direct and objective (first paragraph, for example, can be disregarded).
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 2 Report
This paper summarises the results of water quality monitoring of the Yangtze River. The authors used a standard. statistical methods to arrive at an expected conclusion that the water quality of such a large freshwater system possesses spatiotemporal dimensions. This paper's value is that the study has a quantitative character. The problem with this study is the large scale of the study, which naturally dismisses local-scale pollution, yet still causes environmental damage.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
The article entitled “Evaluation of Spatiotemporal Patterns and Water Quality Conditions Using Multivariate Statistical Analysis in the Yangtze River, China” in its second revision, presents significant improvements and all suggestions/comments were accepted or duly answered. I recommend "accept" for publication.
