Hotspot Detection and Spatiotemporal Evolution of Catering Service Grade in Mountainous Cities from the Perspective of Geo-Information Tupu

Round 1

Reviewer 1 Report

This study analyzed spatial structure of catering services over time in a mountainous city. It uses an interesting approach to characterize the distribution of catering services. The manuscript is generally well written and clearly presents the methodology. Nevertheless, there are some points that could be strengthened.

While the introduction would benefit from a clearer structure, some more information should be added to the methods section. More substantial revisions are required for the results and discussion section. The results section contains many details and a more concise presentation would increase clarity.

The discussion is in my opinion still the weakest part, as I miss the placement of your findings in a broader context, which would make it more interesting for an international readership (e.g. also referring to other approaches in other countries and discussing the innovation of the proposed method). Although some limitations are mentioned, I think that methodological issues should also be more critically reflected such as disregarding influencing drivers for the development of catering services. It would also useful to discuss the applicability of the method in other socio-ecological contexts. Moreover, please consider inserting subheadings to better structure this section.

Specific comments:

L33-97: The introduction needs to be structured. Please separate different ideas by inserting paragraphs.

L50-60: Listing examples is useful to understand the application possibilities of such data or methods used, but this part may be more particularized by summarizing key messages instead of providing a serious of single sentences, each presenting another case.

L90-92: Please indicate also the period of the analysis (i.e. 2015 – 2020).

L108-110: Please add demographic data, including changes over the past years.

L116-124: The division in groups in the two master plans is not clear for international readers and maybe not of relevance for the study. Please consider removing or providing further information.

Fig. 1 (L100): As mentioned before, I am not sure about the added value of indicating the groups. I suggest strengthening the map by indicating population density, which is more relevant for evaluating the spatiotemporal development of catering services.

L131: Please explain in more detail the classification, to make clear what the difference is between first-class and second-class catering.

L137: A reference would be useful to support the theory.

L151: Please indicate the software used.

L217-231: I suggest moving this paragraph to the methods section, as it actually describes what has been done.

L216-407: The results section is very long and somehow difficult to follow due to many details. Moreover, some statements are already an interpretation of the results, which would be part of the discussion (e.g. L329, L337-339, L404-407). To increase readability, I therefore suggest to limit the description to the main results and to remove all less relevant information as well as statements that are rather a discussion of the results.

L421-431: This information would be better placed in the introduction. Please consider adapting and moving.

L474: As you mention, this analysis focused on structure characteristics without further exploring the influencing factors. However, such understanding is highly important to develop management strategies at different levels, e.g. for spatial planning. I therefore suggest deepening the discussion here on possibilities to integrate the analysis of underlying mechanisms in your approach.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

This paper is devoted to the analysis density estimation of catering services. This topic seems to be interesting.

Although I do not feel qualified to judge the English language and style, it seems that the paper's problem is a complex and confusing structure of the text. Several enumerations are used in the text, which would be more appropriate to present as bulleted lists. It would be more appropriate to divide long paragraphs into smaller ones containing individual coherent ideas.

Some figures also look confusing (Fig. 5-8). For more information about proper cartographic visualization of kernel density outputs, see Netek et al. (2019; https://doi.org/10.1515/geo-2018-0029).

See the following major comments:

Introduction

I recommend dividing the section into partial paragraphs; or dividing it additionally into two parts: introduction (where the motivation and goals of the article will be described) and related work.

Rows 69-71: I would reconsider the statement about 'Geo-information Tupu.' All the following mentioned studies are from China. I do not agree that 5 or 6 case studies can be marked as 'many scholars'. 

Row 92: It is not clear to me how the third dimension was used in the analyses.

It would be appropriate to present the aim of the paper in a clearer form, e.g., in the form of a bullet list.

Study Area

The information in this section is usually provided in the 'Data Sources and Methods' section. 

Data Sources and Methods

Data source 'AMap' should be explained (and cited). 

How is the information in Table 1 used in the paper (e.g., type - first-class and second-class)?

Figure 2: Term' Extremum of neighborhood pixels' is the same as 'maximum surface' used in paragraph 3.2.2? 

Information on how the 'Geo-information Tupu' calculation is implemented is missing. At least roughly, it would like to state this.

Results 

Based on what (what attribute and what values) were the hotspots divided into five categories?

Figure 5-8: In figure captions, it should be mentioned that the letters, as well as Roman numerals in all these figures, corresponding to those used in Figure 1 (if so).

Figure 6: The topographic base map (rivers, roads) used in the map window for 2015 does not correspond to the map window for 2020. The terms used for the individual types of symmetry in schemas (a.-e.) do not fully correspond to those used in Figure 4.

Discussion

The sources cited in the discussion [42-51] should be mentioned beforehand in the introduction or literature review.

Conclusions

In conclusion, there is no outline, how these results are applicable in the future.

Other minor comments:

Figure 4: The authors created this figure? If not, a citation should be provided.

Figure 5-8: Labels in maps would generally be read from the right.

Row 412: Elsewhere in the text, the term 'Geoinformation Tupu' is written with a capital first letter.

Check for typographical errors (e.g., rows 35, 41-42, 72-73, 512).

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

Introduction

The article outlines very important and useful context of presenting data connected with urban life and its spatio-temporal context. In my opinion, this issue is by all means important.

 Minor remarks

• The introduction section is very long. It is worth emphasizing what is most important. I suggest to put the rest in a separate paragraph in this section.
• It could be interested to prepare a map showing the density of catering POI changes between 2015 and 2020.

·       It would be also useful and interesting for readers to get some remarks about the choice of catographic presentation methods.·       It would be worth adding some information about the software that was used.  Technical remarks·  

Figure 1 – please change the scale bar into the simplest form (just one or two sections with the distance marked), it is not necessary to describe the 00' and the geographical direction for each coordinate (each meridian and paralel),

Figure 2 – it is very usefil and clear,

Figure 5 – the background content of maps should be generalized, the blac letters (A, B, C,…) are really invisible, it is not necessary to describe the 00' and the geographical direction for each coordinate (each meridian and paralel),

Figure 6 – please, correct the linear scale (scale bar): 2 or 4 km should be the base segment (section), not 3,5 km;

Figure 5, 7, 8 – Please corrrect a legend of hotspots: big values (large catering service hotspots) should be at the top, ans the small one in the bottom.

Although the paper is well structured, minor edits are required. I recommend the paper to be considered for publication.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Most of my suggestions have been responded to in the text of the paper. The clarity of the paper has increased. I have only a few following comments now.

Points 3 & 19: Some figures still need improvement. Figures 5 and 6 should have one legend; the map is degraded by low graphics quality (resolution) or lossy compression. Fig. 6 additionally contains contour lines in strange colors. The layer of water areas (rivers) should be placed above the contour lines. I do not see you adjusting the orientation of the map labels when I meant the horizontal ones. See, for example, https://desktop.arcgis.com/en/arcmap/10.3/map/working-with-text/general-label-settings-about-the-general-label-set.htm (section Orientation of vertical labels).

Point 13: If you have used Jenks natural breaks as a classification method, it will help list a specific extent of all five categories. Because category ranges can vary a lot. Jenks' natural breaks should be cited.

Point 17: I had in mind a few sentences or paragraphs that would indicate "Future work."

Note: I thought that the instructions for submitting modified versions of the manuscript include information that the modifications should be highlighted.

Author Response

Please see the attachment.

Author Response File: Author Response.docx