Quantifying Urban Expansion from the Perspective of Geographic Data: A Case Study of Guangzhou, China

Round 1

Reviewer 1 Report

Review:

This manuscript develops an important topic of interest for the scientific community, a new method for quantifying urban expansion. However, the structure and the depth of treatment of the themes in some of the points demonstrate the need to refine the text. Therefore, here are some optional suggestions to improve the manuscript:

1.

The authors are encouraged to update and engage the literature review with current innovative studies in the planning and design literature that explore the link between rapid urbanization and the built environment. It is also important to bring non-technical aspects of urban planning and design, namely measurable human aspects. For example, the following papers:

in

Zhang, J., Tan, P. Y., Zeng, H., & Zhang, Y. (2019). Walkability assessment in a rapidly urbanizing city and its relationship with residential estate value. Sustainability, 11(8), 2205.

Larsen, L., Yeshitela, K., Mulatu, T., Seifu, S., & Desta, H. (2019). The impact of rapid urbanization and public housing development on urban form and density in Addis Ababa, Ethiopia. Land, 8(4), 66.

Wendel, H. E. W., Zarger, R. K., & Mihelcic, J. R. (2012). Accessibility and usability: Green space preferences, perceptions, and barriers in a rapidly urbanizing city in Latin America. Landscape and urban planning, 107(3), 272-282.

2.

Frenkel, A., & Ashkenazi, M. (2008). Measuring urban sprawl: how can we deal with it?. Environment and Planning B: Planning and Design, 35(1), 56-79.

3.

The indicator of 200m is important for urban planning as a walkable index. Good urban blocks are based on this measure. The authors are encouraged to refer to the delineated grid of 200m×200m determined by the BUAs. This most likely will lead to important observations of the research.

4.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper is interesting but the topic of the paper is not well presented. The methodology part needs to be significantly improved. The part related to the verification of the model needs to be also improved. Model verification has not been done in an appropriate way and presented in an adequate way to see the real advantages and characteristics of the model. The verification of the model should be explained in more detail and the advantages of the proposed model should be confirmed. The shortcomings of the models are too general and their solution is emphasized as the subject of new research

TRANSLATE with x English

TRANSLATE with COPY THE URL BELOW Back EMBED THE SNIPPET BELOW IN YOUR SITE Enable collaborative features and customize widget: Bing Webmaster Portal Back

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

The manuscript is well-written and provides a clear view of the goals of the research.

However, it presents major flaws. 

The literature review on the algorithms to delineate built-up areas (BUAs) is very limited (it is somehow included in section 1) and should be expanded. There should be a section for literature review, where it should be clear to the reader what is the state of the art, and understand how this article provides an advance. A particular emphasis should be done on transformations of POI data into BUAs, as this is the focus of the paper.

In the section that presents the method, to delineate the BUAs from the points of interest (POI), the description is very basic and the proposed algorithm is not compared to other methodologies, i.e., its advantages and disadvantages, when compared to other geometry-based techniques for the extraction of areas from massive sets of points, are not described. 
Moreover, it is not explained what is the criteria to select triangles from the TIN that constitute the Natural City (NC) polygon. Why are some triangles selected and others are not? Is there any geometrical criteria? How are the TIN polygons delimitated? It seems that convex hull methods were not used, but the reader has no idea how to replicate the method as no details are provided.

In the application to the Guangzhou case study, details on the data (number of POI, database structure, POI typologies, POI locations) are absent. It is necessary to understand if a building can have multiple POIs; if every building has a POI; if POIs are related only to businesses and public buildings; etc.

Later, it is not clear how the expansion to the eight spatial orientations was calculated. Is the central point used in the graphs in Figure 8 the centroid of each subdivision of the Guangzhou district?

Finally, the conclusions are not adequate, as the authors should focus on analysing their algorithm's accuracy and ability to approximate the "real BUAs", and not specifically on Guangzhou's expansion dynamics.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 4 Report

The manuscript "Quantifying urban expansion from the perspective of geo- graphic data: A case study of Guangzhou, China" is, in my opinion, very interesting and with minor revisions should be published. Both the purpose, method and results are very interesting. The manuscript should pay a little more attention to the source data. For foreigners, POI data are an interesting but unknown database. Who is the administrator of this data, how is it collected, how often is it updated? Such information should be described in more detail. The reliability of the results depends on this information. The time of data updating can be a significant disadvantage in comparison to remote sensing data.
I also don't quite understand the described concept of Natural cities (NCs) - line 68 and onwards. I do not know what exactly the author meant by this phrase nor how he identifies NCs. Does this concept appear in other literature?
The differences between Figures 5 and 6 are not apparent. Neither between them nor within each of these drawings. As such, they add nothing to the paper, I would advise that the information the authors wanted to show in them be put in a table (e.g. surface differences) and cartographically show examples of close-ups of where the changes are most apparent.
I have doubts about the conclusions drawn from BUA expansion rates in multiple spatial orientations - Table 3 and 8. Based on only one time frame (2014 - 2020) we can talk about change but I don't know if it is expansion. We are not sure if the trends are constant. It is worth noting in conclusion that this type of research should be cyclical to talk about a trend. How many years can such studies be repeated based on POI data?
These remarks do not significantly affect the overall quality of the manuscript, which in my opinion is high.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors have fully addressed all the comments and I recommend the paper for publication as it is.

Comments for author File: Comments.pdf

Author Response

Thank you very much for your recognition of this research.

Reviewer 2 Report

The work has been improved. I suggest it be accepted

Author Response

Thank you very much for your recognition of this research.

Reviewer 3 Report

Comparing with the previous version, In the new revised version most of the problems were corrected. There was a minimal effort to describe the process of generating NC polygons from POI data, and describe the source of data, which greatly improves the clarity of the manuscript.

However, from the scientific point of view, it is not yet clear how the applied methodology constitutes a step in the path to improve the process of delineation of urban expansion areas based on POI data, within the body of knowledge of spatial data processing.

Authors have based their argumentation on the fact that their POI data processing method is "more" feasible to delineate built-up areas (BUAs) than satellite imagery-based methodologies. However, this should not be the central argument, as no comparison or discussion with satellite imagery-based processing results is done.

A more complete literature review must be presented, where references that guide the reader to understand what is the state of the art (in generating BUAs from POI points) can be found, and explaining why the work described in the paper is giving a contribution. For example, the passage of points to areas where the points are concentrated/clustering can be made using heatmaps, density maps from kernel density functions, zonal raster analysis functions, and many other algorithms and resources available in GIS software. Perhaps these methodologies require too much parametrization and are dependent on the knowledge of the analyst? and perhaps your method is better/quicker/more consistent? These would be some valid reasons to use it.

A good example of such presentation, in which authors are invited to reflect, is given in section 2 of the following paper: Zhou, Na (2022). "Research on urban spatial structure based on the dual constraints of geographic environment and POI big data". Journal of King Saud University - Science, 34(3), 101887. doi: 10.1016/j.jksus.2022.101887

Another example is the description of alternative generations of areas from POI points presented by Wang Z, Ma D, Sun D, Zhang J (2021) Identification and analysis of urban functional area in Hangzhou based on OSM and POI data. PLoS ONE 16(5): e0251988. https://doi.org/10.1371/journal.pone.0251988

Authors did almost nothing in respect to this, except change references. Please give the readers the opportunity to understand what contribution does the method you presented provide, when compared to other similar research.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf