Research on the Impact of Landscape Pattern in Haikou City on Urban Water Body Quality
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThe paper addresses a relevant and timely topic, making a meaningful contribution by linking landscape pattern characteristics with urban water quality parameters. However, there are some methodological issues that limit the reproducibility, robustness, and interpretation of the results. Below are several observations and suggestions for improvement.
Data Source Description : The manuscript does not clearly indicate which land use dataset was used (e.g., its source, year, classification scheme, and spatial resolution).
Suggestion: It is important to explicitly state the database used (for example, GlobeLand30 or a national land use map), its spatial resolution (such as 30 meters), and the year of data acquisition. Additionally, it should be clarified whether the dataset underwent any preprocessing steps, such as reclassification or resampling. Furthermore, although landscape metrics are mentioned, the parameters used for their calculation—such as neighborhood radius, moving window size, or the definition of spatial units—are not detailed.
Suggestion: Specify these parameters (distances, window size) and justify their selection, referencing appropriate literature where possible.
Selection and Justification of Metrics: Several landscape indices are calculated, but it is unclear why these particular metrics were chosen or whether multicollinearity among them was assessed.
Suggestion: Include a brief explanation justifying that the selected indices adequately capture the relevant urban landscape structure for the study, and confirm that there is no strong correlation among them, for example, by conducting a variance analysis or presenting a Pearson correlation matrix.
Correlation and Regression Analysis: The authors apply correlation analyses and multiple regression models, but no correlation matrices or regression diagnostics are presented. This makes it difficult to assess the strength of associations and the reliability of the models.
Suggestion: Include a) a correlation matrix or at least a table with Pearson correlation coefficients; b) basic regression statistics such as R², adjusted R², p-values, and standard errors; and c) verification that model assumptions were checked (normality of residuals, heteroscedasticity, independence, etc.).
Also, it is unclear whether lag effects were considered, as landscape conditions may influence water quality with some temporal delay.
Suggestion: Include a brief justification for using synchronous data or, alternatively, discuss the potential impact of lagged effects in the limitations section.
Spatial Autocorrelation: Since both explanatory and response variables are spatially distributed, spatial autocorrelation is likely present and could bias correlation and regression results.
Suggestion: Test for spatial autocorrelation in model residuals (e.g., using Moran’s I), and if significant autocorrelation is found, consider using appropriate spatial regression models, such as spatial lag or spatial error models.
Results: In some parts of the discussion, objective results are mixed with speculative interpretations—for example, attributing changes in water quality to shifts in population density without supporting data.
Suggestion: Clearly separate the presentation of results from their interpretation, and ensure that interpretations are directly supported by the data or cited literature.
The authors frequently use the term “significant influence” without clarifying whether this refers to statistical significance (p < 0.05) or simply practical or qualitative importance.
Suggestion: Always clarify the type of significance being referred to and report the corresponding p-values.
Figures and Tables: The spatial figures provide valuable information, but the legends use small font sizes and low contrast, making them difficult to read.
Suggestion: Increase font sizes in legends and axes, and ensure sufficient contrast in color maps for better visibility.
Additionally, the results section would benefit from including a table summarizing regression coefficients alongside their significance levels.
Language and Terminology: There are inconsistencies in the use of terms, such as alternating between “green landscape” and “green infrastructure,” or between “gray infrastructure” and “impervious surfaces.”
Suggestion: Use consistent terminology throughout the manuscript.
Finally, there are some minor syntax and grammar issues. A thorough proofreading by a native English speaker is recommended to improve overall language quality.
Comments on the Quality of English LanguageThere are some minor syntax and grammar issues. A thorough proofreading by a native English speaker is recommended to improve overall language quality.
Author Response
please refer to the attachment
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsHello
I think the article covers a good topic, but it has some shortcomings that could be considered in the revision stage.
- The title is written very generally and does not specify the innovation or specific angle of the research, suggestion: the title should be more precise and limited to the location (Haikou City) and the method (landscape metrics + water quality monitoring).
- The objective, method, key results, and conclusion should be stated more clearly in the abstract section.
- The abstract should provide numerical results, for example, it says that COD and TN are high in a lake, but the exact amount or percentage of change is not mentioned.
- I think the Literature Review is lacking, it is recommended some literature in the introduction and to compare the results, especially in the context of landscape metric applications in urban green space.
- The purpose of the research is stated in general terms, but the research hypothesis or research question is not clear.
- In the research method section, GF-2 was used for the remote sensing data mentioned only and classification was performed with MLC, but the accuracy of the classification (Accuracy Assessment, Kappa, OA) was not reported.
- In the selection of landscape metrics, only (Patch Number, Fragmentation, Spread) was mentioned, but the reason for their selection and classification is not clear. Why were these 11 landscape level class indicators chosen? Could patch class indicators (e.g. for the Built-up Land class) provide more accurate information?
- The tool used in calculating landscape metrics should be mentioned. There are several articles in this field that can be used. For example: Spatial comparative analysis of landscape fragmentation metrics in a.
- The water sampling design is poorly explained. For example, why only 30 points? Is it statistically sufficient?
- Has regression or modeling of the relationship between landscape metrics and water quality parameters been performed?
- Why and how does landscape connectivity improve water quality? (e.g., by reducing runoff, increasing infiltration, creating habitat for purifying microorganisms).
- Why does built-up land in Jinniu Lake show positive and negative correlations in different buffers? This is a complex finding that requires deep ecological interpretation, not just reporting.
- Are there no sewage discharge points or point pollution inputs in these waters? The impact of these factors can be much stronger than the indirect impact of landscape pattern.
- The text mentions that these waters have undergone “ecological restoration” (e.g., Hongcheng Lake). How can the impact of these intensive engineering measures (dredging, circulation system, constructed wetlands) be separated from the passive impact of landscape pattern?
- Why are identical circular buffers used for a small lake (Jinniu Lake: 6.3 ha) and a long river (Meishe River: 23.8 km)? Wouldn’t it be more logical to have riparian buffers on both sides of the river?
- The number of sampling points for the Meishe River (only 5 points) seems too small for a long river. Can these samples really show its spatial variability? Please provide an explanation.
- Results should be segmented. For example, LULC analysis, landscape metrics analysis, water quality analysis, relationships. This study only shows correlation, not causation, and this is one of the main weaknesses of the paper.
- I think the discussion is more descriptive than analytical. For example, it is said that “vegetation improves water quality”, which is obvious, but the exact mechanism or new findings are not stated.
- No comparisons with other cities in the world are made, which needs to be expanded.
- Ecological meaning of the indicators: what does an increase in LSI or a decrease in SHDI actually mean for city managers on a larger scale? These results need to be translated into management practices.
- The conclusion is general, the final section of the article should state exactly which landscape metric had the highest correlation, what spatial scale was best, and what specific policies can be adopted for Haikou.
Author Response
please refer to the attachment
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Comments and Suggestions for AuthorsHello,
Dear Editor,
After a complete review, it can be said that the authors have adequately addressed the comments provided by the reviewers. Therefore, the manuscript can be accepted for publication.
Final Reviewer
