Investigating the Spatiotemporal Relationship between the Built Environment and COVID-19 Transmission
Round 1
Reviewer 1 Report
Dear authors,
You conducted a very interesting study incorporating the temporal dimension into spatial regression (GWR) and tackled a field of research which is still (to a certain extent) under-studied: the role of the built environment in infectious disease spread. However, there are still several questions related to your approach and the context. More specific comments can be found in the following:
Introduction
- page 2, lines 51-63: Evidence here almost exclusively based on studies from China or USA, suggest to widen focus towards other regions in the world, e.g., https://0-doi-org.brum.beds.ac.uk/10.1007/s11524-022-00708-5, https://0-doi-org.brum.beds.ac.uk/10.1016/j.scitotenv.2021.148626, https://0-doi-org.brum.beds.ac.uk/10.1017/S095026882000134X, https://0-doi-org.brum.beds.ac.uk/10.1136/jech-2020-216325
- page 2, line 51: Why are the transmission mechanisms "hidden"?
- page 2, line 59: Spread of disease and spread of pandemics used interchangeably in the text and also this paragraph. Suggest to avoid that as the focus is the disease.
- page 2, line 61: "Coronavirus occurrences"? So more than SARS-CoV-2? Please be specific here.
- page 2, line 64-65: Plural? So several pandemics. Which pandemics are you referring to?
- page 2, lines 68-69: How does OLS capture the spatial relationship?
- page 2, lines 82-85: I cannot follow your argument placing the meso-scale as the best approach here between macro- and micro-scale. Please further elaborate to underpin your point.
Data and methods
- page 3, figure 1: What is "Lianjia & Anjuke"? What is "amap"?
- page 3, lines 102-103: Please elaborate. What were selection criteria? Based on which approach have you reduced the number of variables?
- page 3, lines 104-105: In the intro you praised GTWR as your method of choice. Why would you compare the statistical models here in the manuscript if you focus is on the association of BE and COVID and not the methodical approach per se.
- page 4, line 111: Why is the reference to the size of US countries relevant?
- page 4, line 113: COVID-19 is the disease. Omicron is a variant of SARS-CoV-2.
- page 4, line 116: "positive cases" - within which time frame?
- page 4, lines 116-117: Unclear. What does the level of management mean?
- page 4, lines 117-118: How were you able to detect and follow asymptomatic cases?
- page 4, figure 3: How do you know "asymptomatic infected cases" if they were not confirmed (i.e. orange colour)?
- page 5, line 128: Which year?
- page 5, line 134: The variables mentioned below are demographic and economic (rents and prices). Suggest to remove "social" here.
- page 5, table 1: Calculation process needs to be better aligned with corresponding variables on the left hand side. Currently very diffiult to read. In addition, variables would benefit from a definition. Could be added as an additional column to the table.
- page 5, table 1: What are 12 categories of POI datasets?
- page 6, line 154: Why?
- page 6, lines 156-157: Please reflect that stepwise regression is a purely data-driven process without no theoretical underpinning related to your data. What impact may have this on your results?
- page 7, figure 4: What are the units of the respective legends and why does the granularity of data vary (i.e. different spatial scales, for example, subdistrict population very detailed and metro line length rather coarse)?
- page 7, lines 165-168: If you are aware of the shortcomings of OLS by proposing an alternative method, why include OLS here in the manuscript?
Results
- page 8, line 185: The following paragraph is difficult to follow.
- page 8, line 187: What was the outcome under investigation here? Cumulative cases? At what point in time?
- page 8, line 188: What is the average infected population?
- page 8, lines 191-193: I do not understand. "Remained stable" in relation to what?
- page 8, line 195: "Remained unchanged" in relation to what?
- page 8, figure 5: Does not add important information here.
- page 9, lines 208-209: I cannot follow. Please elaborate why.
- page 9, table 2: Perhaps I am lost here, but Moran's I ranges from -1 to 1. The differences you show here seem negligible in this context, no?
- page 9, lines 214-216: This sentence would be well suited in the methods sections.
- page 10, line 221: What is "inn"?
- page 10, line 221: Did you do a correlation analysis? If so, where? The beta indicates a strong positive relationship or association.
- page 10, line 227: What is the reference case area?
- page 10, line 229: R²
- page 10, line 231-232: Interesting that the spatial non-stationarity apparently had basically no influence, if one compares the AICs of the two models. Do you have an explanation for that?
- page 11, lines 238-245: This part is not a result but belongs to the method section.
- page 11, linw 251: How can the metro lines have a negative impact? On what? Formulation not clear.
- page 11, line 256: Positive impact meaning higher or lower number of infected people?
- page 12, figure 7: Maps not readable, please increase the font size of the text inserted (specifically of the legend).
- page 13, figures 8.1 and 8.2: Maps not readable, please increase the font size of the text inserted (specifically of the legend).
- page 14, figures 8.3 and 8.4: Maps not readable, please increase the font size of the text inserted (specifically of the legend).
- page 14, lines 312-313: No need to repeat it several times in sub-sections.
- page 14: lines 313-314: In how far vary population, rents and BE variables used by you over time in such a short period of only a few months?
- page 14, line 316: metro line length also positively
- page 14, lines 318-319: Why mention first positive, then negative, and now again positive assosications?
- page 14, line 319: "Hotel and inn density" not included in table.
- page 14, line 321: Higher compared to what? Almost all estimates show higher SDs than their averages.
- page 15, lines 323-330: What is this supposed to tell me? Not sure it adds valuable information here.
Discussion
- page 16, line 335: Please be explicit and refer to the disease.
- page 16, lines 351-353: How do you assess the effectiveness of the "dynamic zero COVID-19 strategy" here in your manuscript. Please also reference this statement.
- page 16, line 358: How did you determine it as the most influental aspect?
- page 16, line 366: How did you assess "risk of infection". If I understood correctly, you only know where a person, who got infected, lives and not where the actual infection/transmission has been taking place, no?
- page 16, line 372: See comment above on "pandemic".
- page 16, lines 373-376: Following your statement: the higher the walking accessibility, the more infections in the city center. Is that plausible and if so, why?
- page 17, lines 416-417: I do not understand this sentence. What does "majority of incubation period instances" refer to?
- page 17, lines 418-419: "Earlier migration data", what is that?
- page 18, line 464: "Ignored" sounds it like it was excluded on purpose. Was this the case?
- page 18, lines 478-481: This would be indeed very interesting. Good point.
Conclusions
- page 19, line 488: Prevalence, no?
- page 19, lines 495-496: Not sure about this. Sounds like working on the "symptons" rather than the underlying "disease". What about promoting other modes of transport?
Dear authors, some inaccuracies in the language made it difficult to follow at times. Please avoid using pandemic, epidemic and disease interchangeably, e.g., "spread of pandemic" as the pandemic does not spread in Shanghai but the disease. There are a couple of other similar linguistics questions I inserted in my comments.
Author Response
Dear Reviewer,
Thank you for your feedback on our manuscript. We have taken your concerns seriously and have taken steps to address them. We hope that the paper now meets the standards of publication. Your contributions will be duly acknowledged in our published paper. Please kindly find the response letter at attached word file.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
ijgi-2495320
Under Siege: Investigating the Spatiotemporal Relationship Between the Built Environment and COVID-19 Transmission
Comments to Authors:
1. In the title, the word “Under Siege” looks inappropriate for a research article. Authors are advised to avoid using such violent and aggressive words in their research.
2. Very long sentences are used in the manuscript. See Abstract Lines 26-30. Hence, authors should revisit their writing styles and recheck their whole manuscript for such redundancies and grammatical errors.
3. Revisit keywords. Authors are suggested to review the rules of highlighting keywords in the research article.
4. In the Introduction, the authors tried to clarify the situation of a pandemic by referring to the two renowned cities of Europe, i.e., London and New York. The claim is made without any source information or real-time data on these both cities. The authors claimed that both cities have higher densities, but not clarified the density information. Large-scale population mobility stats are also not mentioned for both cities.
5. Lines 42-43. Authors claimed that both cities, i.e., London and New York are more likely susceptible to high infection of COVID-19. But why? No data are available to support this claim. My questions are as follows:
a. If population and its density are the culprits of higher infection rates, then why authors have not chosen the highest populated countries of the Globe, including China, India, and Pakistan? India and Pakistan are comparatively less developed countries compared to China, these both countries have the highest densities of metropolitan compared to New York and London.
Therefore, my advice is to rethink the idea and be careful in selecting case studies or literature. Authors are advised to rewrite the Introduction Section and show some facts and figures to validate arguments.
6. Lines 55-56, show facts as claimed.
7. Line 74. Revisit citation rules. [35] can’t conduct macroscale investigations, but Authors. This is inappropriate. Please correct the whole manuscript and its citations with author names and years.
8. Put related case studies of “Geographical and Temporal Weighted Regression (GTWR)” to support your methodology.
9. Redraw Figure 1. See related published articles.
10. Figure 2 is inappropriate and not readable. See related published articles.
11. What is the source of Figure 3? If it is part of the literature, then move it to its appropriate place. Authors are advised to show all respective years of infection data.
12. Table 1. Looks again part of the Literature. Link it with literature. It is missing in the manuscript.
13. What is the purpose of Figure 4? Where did it come from? No link is found.
14. All mathematical expressions should first be discussed in literature then methodology. No relationship was found between the methods and literature. Therefore, the methodology is termed Vague. Authors are suggested to correct their manuscript. This is a massive correction. Rewrite the whole section.
15. Results should be discussed for study objectives.
16. I can see many outliers in Figure 5 (a). What are x and y axis in Figure 5 (a) and (b)? Rewrite the explanation of Figure 5.
17. Figure 5 (b): What is Morans 1:0.48? Why the curve is left skewed?
18. Why Global Moran's I analysis was executed? Is this method cited in the literature? I don’t see any related literature. Justify and correct the manuscript.
19. Figure 6 is not clear. Redraw. See the related articles of MDPI.
20. How you obtained the values as explained in Table 2? Have you discussed the patterns in the manuscript as given in Table 2? Justify and Correct.
21. All the models should be calibrated and validated as described in the manuscript. See Table 3. How these are related to the study’s scope?
22. Many loopholes were found in the results section. Authors are advised to rewrite the whole section.
23. Figure 7 is not readable. Break it into multiple Figures. Explain this figure very well.
24. Figure 8 is not mentioned, but 8.1 to 8.4. All the figures are not readable. Correction is required.
25. What is the utility, and purpose of Figure 9?
26. Summarize every section of the manuscript.
27. Put policy implications, significance, and contribution of the study just before the conclusion section.
28. Rewrite the conclusion section.
29. Proofread manuscript. Many errors were found. Show similarity index.
30. Show valid literature to support the methodology. Justification is needed concerning the result section.
31. Perform all suggested corrections to make this research publishable.
English needs massive improvement.
Author Response
Dear Reviewer,
Thank you for your feedback on our manuscript. We have carefully reviewed your comments and have taken them into consideration. In our revised manuscript, we will thoroughly address each of the corrections you have suggested. This includes clarifying the methodology, improving the organization of the sections, enhancing the readability of the figures, providing detailed explanations for the results, and ensuring a comprehensive discussion of the findings.
We are grateful for your valuable feedback, as it plays a crucial role in strengthening the quality and impact of our research. We will dedicate the necessary time and effort to implement the necessary changes and make sure that our manuscript is publishable. Thank you for your thorough review and constructive feedback.
For the detailed response letter, please see find the attachment.
Kind Regards
Author Response File: Author Response.pdf
Reviewer 3 Report
The authors of the submitted manuscript# ijgi-2495320 have investigated the different urban aspects of urban built environment for their association with the COVID-19 Omicron outbreak. The authors have used time-series data during the Omicron outbreak in the Shanghai area in China and have used OLS, Geographically Weighted Regression, and Geographically and Temporally Weighted Regression to examine the spatial as well as temporal patterns of association between the built area characteristics and Omicron cases. After reading their manuscript, this reviewer’s opinion is that the paper does a nice job of explaining the association between the observable urban variables to the Omicron cases over space and time. The paper may be accepted for publication after the authors have addressed this reviewer’s comments and that of the other reviewers and editors.
Please find the comments below:
- I feel that the abstract section can be edited to highlight that this paper highlights the spatial and temporal variation in the association between the BE variables and Omicron cases. (Lines 23 – 25).
- Line 70: The term “dynamically” may be better written as “spatially”?
- Line 138: Data “purification” is not commonly used; please use an alternative such as correction or cleaning.
- Table 1: The way this table has been presented needs to be clarified. Please use vertical and horizontal lines so the readers can understand which item relates to what across the columns. Also, for the variables that used an equation under the “calculation process” column, it is suggested that the authors use separate equations for each such variable and not combine them as has been done in, e.g., equation 1 (green space density and road/bus length/metro line length densities)
- Line 155: What undesirable effects of sample data do the authors indicate here? Please include one or two examples.
- Section 2.4: Please also provide the equation for OLS and GWR here. Similarly, please provide the equation to show how Global Moran’s I is calculated.
- Line 242: “…spatial variations in the effects of BE variables on infections.”- suggested that the authors use the term association instead of effects as regression does not indicate causality but rather an association.
- Line 373: “… as shown in previous studies.” Please indicate the references here.
- Section 4.1: This section lacks reference to the regression results as the authors discuss the spatial patterns of these variables. Please either provide the regression coefficients for these variables here or refer to the tables and/or maps in the reference section.
- Line 408: From the description of the data used in this study, it does not seem like “Big Data” is a proper way to identify it. Rather the term multi-source data or something similar may be a proper way to say this.
- Lines 494 – 496: The reviewer does not feel that the authors can claim metro line optimization or improved ventilation systems based on the study results, as these variables are not included. These can be claimed generally but not from the study results. Similarly, the study variables did not include equity aspects.
Author Response
Dear Reviewer,
Thank you for your feedback on our manuscript. We have taken your concerns seriously and have taken steps to address them. We hope that the paper now meets the standards of publication. Your contributions will be duly acknowledged in our published paper. We have revised the manuscript based on your comments. The specific revisions made are summarized at attached word file. For the detailed response letter, please see the attachment. Many thanks.
Kind Regards
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Comments to Authors:
I must appreciate the efforts of the authors. They polished their manuscript well. Meanwhile, a few minor corrections are still required as follows:
1. Show research contribution, significance, potential impacts, and the way forward in factors. These factors should also be elaborated in different parts of the manuscript, including the abstract, review, and conclusion.
2. It is not suitable to start a new section without putting some introductory text. Therefore, at the start of section 2, authors should first briefly describe the data and methods as text. After that authors may put Figure 1.
3. Figure 1 should also include a conclusive explanation of the methodology as the last step.
4. In many articles, it is observed that study area (s) are explained first then data and methods. See your manuscript and shuffle the related stuff.
5. See Table 2 and related literature. For the accessibility-related portion, authors should also cite the following articles: DOI: 10.3390/su15076228; for accessibility and regression analysis authors should cite DOI: 10.1007/s10668-022-02288-5.
6. In the methods and results section, authors should discuss the related factors keeping in view the aim and objectives of the city. This thing is missing.
7. Show results as per study objectives and relate the same with SDGs. Clarify that your study is linked with the SDGs. Inserts this in various sections of the manuscript.
8. The conclusion section should also clarify the results and novelty of the research. It is comparatively a detailed section of the Abstract. Therefore, show problems, methods, and results in this section. Authors are also suggested to clarify the research gap, significance, and contribution of their study in this section.
9. Check the similarity index and show it to the editor.
10. Proofread for minor grammatical errors.
English requires minor proofreading.
Author Response
Dear Reviewer,
Thank you for your valuable feedback on our manuscript. We appreciate your acknowledgment of our efforts in polishing the manuscript. We have carefully reviewed your comments and have made the following minor corrections.
For the detailed response letter, please see the attachment.
We would like to express our gratitude for your valuable suggestions, which have significantly improved the quality of our manuscript. Your feedback has helped us enhance the clarity and effectiveness of our research. Thank you for your time and consideration.
Kind Regards
Author Response File: Author Response.pdf
