The Effect of the COVID-19 Pandemic on the Distribution of Traffic Accident Hotspots in New York City

Round 1

Reviewer 1 Report

The manuscript presents the spatial analysis of accidents in New York City before and during the mobility restrictions that followed the COVID-19 outbreak.

In general, the readability of the manuscript is acceptable.

Literature review barely lists a bunch of papers related to the impact of COVID-19 mobility restrictions to safety. To improve it, the papers cited should be grouped by specific subject matter or by common aspects of the research. For example, urban or intercity mobility; focus on vehicles, roadway features or human factors; statistical toolset used, etc. Although I missed many studies, 53 documents cited are more than enough.

Getis-Ord analysis is a spatial statistic used in geography and spatial analysis to identify clusters of high or low values in a spatial dataset. When two or more different spatial datasets are to be compared using these techniques, a quantitative comparison is expected. The G*-statistics, which indicate the level of clustering, have not been showed nor compared. By comparing these values the researchers could have observed which dataset has a more pronounced spatial pattern and draw conclusions in this respect.

Z-scores and p-scores have not been examined. A high positive z-score and a low p-value indicate significant clustering of high values, while a low negative z-score and a low p-value indicate significant clustering of low values. The differences in these indicators could have led the authors to interesting conclusions.

The authors have neither examined rigorously the spatial overlap and differences in the identified clusters. Please, consider using statistical tests to formally compare the results. This could include hypothesis testing or other statistical methods to determine if there are significant differences in the spatial patterns identified by the two Getis-Ord analyses.

The discussion section contains general statements about mobility pattern changes during the restrictions without empirical evidence in the results. As a result, this section does not contain conclusions nor contributions to the state of knowledge.

A list with other minor issues can be found below:

Literature review. If the author of a document is mentioned, citation number must appear immediately after.

Line 59. “different time granularities”. Meaning not clear

Line 64. “Influence of Covid-19 pandemic” should read “influence of mobility restrictions during Covid-19 pandemic”

Line 164. What does “process the missing” data mean?

Figures 3 to 9 should be enlarged. The reader can hardly read the legend.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

The authors put a lot of effort in producing this paper but I do no think this paper presents something new to the literature. Similar paper is published recently discussing the same topic using almost the same methods.

How COVID-19 impacted the temporal and spatial distribution of collision hotspots

I see no value of repeating the same method again in a different City.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

This is an interesting article summarizing the phenomena taking place in road traffic engineering in recent years.

The structure of the article is correct. Research methods and tools are selected appropriately.

1. Please complete the bibliography referencing the main R language packages used. Information on how to cite a given package is available using the citation("package_name") function.

2. Please insert differential maps showing the increase or decrease in the number of accidents in specific areas. You can make such maps using raster map algebra.

3. I propose to address, at least in the discussion, whether the temporal/spatial patterns are identical for all types of accidents. For example, in the article: https://www.mdpi.com/2071-1050/14/3/1253 you can see that the time pattern for accidents with animals has not changed (Fig. 4 and Fig. 5). What about the data used for the article?

4. Please also refer to the severity of accidents. In my country, the share of fatal accidents increased significantly during the lockdown period - this was described in the presentation: https://media-prof.pl/web/files/24448/PKD_MBajor_Analizy-BRD_2021_05_18.pdf

5. Please enlarge the maps and improve their resolution.

6. Place a space before square brackets - e.g., lines 81 and 85.

Good luck! I hope to read the revised article soon.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

I am really impressed with the authors' work. They have been able to address my sctrict comments and follow my suggestions to improve the quality of the research and therefore that of the manuscript. It deserves publication now.

Author Response

Thank you very much. Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

The response is not satisfactory enough. I do not see differences between this paper and the other published paper that I mentioned in the first round

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 3

Reviewer 2 Report

.