Large-Scale Automatic Identification of Industrial Vacant Land
Round 1
Reviewer 1 Report
This manuscript proposes a method for large-scale automatic identification of IVL. The framework uses deep learning to train remote sensing images of potential industrial vacant land to generate a semantic segmentation model, further use population density and surface temperature data to filter model predictions.
I think this paper is worth publishing. But before the paper can be published, the following key scientific questions must be clarified:
Resource-based cities should not be considered only in terms of growing, mature, declining, and regenerating. The main mineral resources of Tangshan, Daye and Xinyu are iron ore, iron ore and gold ore, respectively. The number of coal-based cities is the largest in China. Shanxi, Shaanxi and Inner Mongolia in China have a large number of resource-based cities represented by coal mining.
On the one hand, the three resource-based cities selected in the study are located in the eastern and central parts of China, but why not consider the western part? Because the surface texture characteristics and spatial distribution characteristics of mining areas in different regions are different.
The above problems will lead to the representative, typical, and practical application value of the selected cases in this study is limited.
None.
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 2 Report
This study presents a novel method for large-scale automatic identification of industrial vacant land by integrating remote sensing images with deep learning techniques. The paper is well-written and meets the standards for acceptance in its current form.
The quality of English in this paper is satisfactory.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 3 Report
Dear Authors,
The manuscript: "Large-scale automatic identification of industrial vacant land: An open-source approach using semantic segmentation of remote sensing images and multi-source data" presents a relevant topic with Environmental and social scope by addressing the identification of industrial areas in disuse. Regarding the writing of the article, please review the use of punctuation, especially the comma. Avoid long paragraphs as they tend to confuse the reader. Chapter 1 is well-written and adequately addresses the problem to be solved. The detailing of datasets is confusing. I also realized that some concepts related to GIS are not very clear: geographic data format for example. I have pointed out minor comments that can be consulted in the digital archive. I ask for your attention to the following suggestions:
1) Title: it is long. Suggestion: Large-scale automatic identification of industrial vacant land;
2) Abstract: present the study site and highlight the accuracy obtained;
3) In Chapter 1: the LVI issue is more urgent in large urban centers. Please highlight this issue.
4) It would be interesting to present the objectives of the study in Chapter 1.
5) Line 63: in the years 2000 and 2005 there were already images with a spatial resolution of 30 cm. Please review.
6) Line 85: when presenting an acronym (HRRS) for the first time, please mention its meaning.
7) Line 155: "Temperature to ensure the exclusion of industrial sites that are still in operation using surface temperature data". In this case, you used bands 11 and 12 (SWIR). Please mention it as it seems that only bands with 10 m spatial resolution were used.
8) Line 159: "City, the validation target, and merged them to achieve a pixel resolution of 15 m, which was used to calculate the land surface temperature inversion". In this case, did the pansharpening process occur? Please detail this fusion procedure with the thermal bands of the TIRS sensor. Remember that Tirs data do not provide the surface temperature. Detail (line 307) the steps to obtain the temperatures.
9) Table 1: I did not understand the use of the letters a,b,c,d, and e in the table legend. The images presented in the table are not from Sentinel 2 and Landsat 8. Did you use data from other sources at this stage (Google Earth Pro, for example)?
10) Line 192: "We segmented the urban RS images in ArcGIS into 256 × 256-pixel rasters based". Did you use the mean sift segmenter? How was the distribution of the 256 × 256-pixel grid defined?
11) Line 227: the raw images obtained from the Copernicus (Sentinel 2) and Earth Explorer (Landsat 8) platforms are ready for use and properly referenced to a geodetic system. The conversion of the images does not imply the loss of the reference system.
12) Line 239: "Delete polygons deemed as noise after area calculation". What is the threshold of the area value used? In this case is it a table operation?
13) Line 288: what is the criterion used to select the spatial resolution of 100 m? In this case, the standard unit for density is people/km².
14) Figure 8: the contours of the buildings are not consistent with the segmentation of Sentinel 2 or Landsat 8 images.
15) Chapter 4.3: Ablation Study? An ablation study usually compares the results that are obtained progressively with the elements of the network. In your case, it would be a comparison study.
16) Chapter 5: separate the conclusions from the discussion. It would be interesting to discuss the results in Chapter 4.
17) Are the results obtained in your study comparable to other works cited in the references? Are there studies considering traditional methods? It would be interesting to present them and compare them with the results obtained by your study.
18) In chapter 2 three cities are mentioned (line 134): "We therefore selected potential LVI from two declining cities (Xinyu and Huangshi) and one regeneration city (Tangshan)". You presented the results only for Tangshan.
I hope that my suggestions will be useful for the improvement of the manuscript and subsequently enable its publication.
Respectfully,
Comments for author File: 
Comments.pdf
The article needs a moderate review of the English language: punctuation, verbal and nominal agreement. Confusingly worded passages were marked in the comments of the digital file.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
The author's manuscript has been greatly improved from the previous version. This paper is therefore recommended for publication.
However, there is still a minor issue that the authors are asked to consider before publication:
In the Section 4 (Result), the authors mention some formulas such as Equation 3—Equation 6. Could the editor and the authors please reconsider whether it would be more appropriate to move the relevant parts of the formulas to the Section 3 (Methods)?
None.
Author Response
The authors thank you for spending your time in careful evaluation and providing positive comment of our manuscript. Presented below is a point-by-point response to your suggestion.
Point 1: In the Section 4 (Result), the authors mention some formulas such as Equation 3—Equation 6. Could the editor and the authors please reconsider whether it would be more appropriate to move the relevant parts of the formulas to the Section 3 (Methods)?
Response 1: We thank you for the constructive suggestions. In our manuscript, Section 3 primarily focuses on introducing the relevant methods and implementation details of the experimental design to the readers. On the other hand, Section 4, including Equation 3 - Equation 6 and the associated explanations, serves as a description of the methods used to validate the experimental results.
We are concerned that moving this content to Section 3 may confuse readers regarding the differentiation between the experimental methods and the methods used to validate the results. Simultaneously, keeping these equations in Section 4 ensures the completeness of the description of the experimental results.
Therefore, upon our reconsideration, we would like to retain Equation 3- Equation 6 in their current position in Section 4. We hope that you can understand our reasoning behind this decision.
Reviewer 3 Report
Dear Authors,
The topic addressed by your article and the methodology proposed to solve the problem with industrial vacant land is very interesting and could probably become a bibliographic reference for other works with the same approach.
I have carefully analyzed the second version of your manuscript and verified that all my suggestions have been incorporated. Your cover letter helped me conduct the analysis.
I would like to congratulate you on your efforts to improve the manuscript.
Respectfully,
- Revise the use of commas.
Author Response
The authors thank you for recognition of this article and the modifications we have made.
We have carefully reviewed and revised the use of commas in the manuscript in accordance with Reviewer 2 comment regarding the quality of English language. At the same time, we have reduced the use of lengthy complex sentences by incorporating more connective adverbs, and improved accuracy in word choice to enhance the overall readability of the article. The all revisions are labeled in red in the revised manuscript.
