Ha Long—Cam Pha Cities Evolution Analysis Utilizing Remote Sensing Data
Round 1
Reviewer 1 Report
This paper uses three Landsat images acquired at different times to evaluate the land use and land cover changes in Ha Long and Cam Pha cities. Overall, the paper is well organized. However, it may be better to take in-depth analysis of the results in your paper. I have following concerns and suggestions for improving the paper.
- Since your results mainly focus on the generation of the land use maps in the study area, the title may not match the content of the paper.
- Lines 68-71: It should be note that the Landsat 6 didn’t achieve orbit.
- Figure 1: It’s better to reorganize the figure to make it clearer and easier to digest. For example, you can change the color mathching of this figure; you should also ensure that the text in the figure is clear. The same problems need to solved in other figures in your paper.
- Figure 3: What about the administrative boundary modification of the Cam Pha city? Explain that how the Figure 2 and 3 contribute to the main objective of your paper. Or maybe they are redundant?
- Line 205: The Table 2 is redundant, since it is well known to us.
- Figure 7: It seems that the sizes of land cover classe samples for supervised classification are too large, thus the samples may contain many land cover classes rather than one class. This can have important impacts on the accuracy of classification results.
- Lines 330-344: How many samples did you use for the classification of each land cover? Make sure that the samples are representative and sufficient.
- Figure 11: Put the corresponding Landsat images rather than the boundary of pixels here to better compare your classification results.
- Lines 365-375: It’s better to provide the original land cover maps generated by this paper, too.
- Lines 378-404: Section 5.1. Change detection: Further analysis is needed. Maybe you can provide the transfer matrix to demonstrate the land cover changes in more detail.
- Table 7: It’s redundant because there are too many missing data.
- Lines 451-485: What’s the relationship between this section and the previous results? More detailed descriptions are needed to be added.
- What is your suggestion for improving the land use planning in the study area?
Comments for author File: 
Comments.pdf
Author Response
We would like to thank the reviewer for careful and thorough reading of this manuscript and for the thoughtful comments and constructive suggestions, which help to improve the quality of this manuscript. Our response follows:
Point 1: Since your results mainly focus on the generation of the land use maps in the study area, the title may not match the content of the paper.
Answer 1: We appreciate the reviewer’s insightful suggestion and agree that it would be useful to demonstrate that change detection is a main part of this study. But the urban evolution is more critical purpose. Urban evolution is the adaptation of organisms to heavily populated urban areas. Anthropogenic environmental and geographical changes brought to habitats when urbanization takes place are known to have a significant evolutionary impact on organisms inhabiting these city areas [1]. This research aims to characterize urban dynamics at the coastal area of Ha Long – Cam Pha cities over this period and highlighting the different tendences of urban evolution between these two cities through the generation of land use/land cover maps. This is the reason we would like to conserve the title as the previous one.
Point 2 Lines 68-71: It should be note that the Landsat 6 didn’t achieve orbit.
Answer 2: We agree with the reviewer. We added into the text: (except Landsat 6 due to the failure at launch after not reaching the velocity necessary) at lines 70-71.
Point 3: Figure 1: It’s better to reorganize the figure to make it clearer and easier to digest. For example, you can change the color matching of this figure; you should also ensure that the text in the figure is clear. The same problems need to be solved in other figures in your paper.
Answer 3: We apologize if our original Figure 1 is not clear and easy to digest. We have modified the figure and checked to be sure that it is now clearer: We add water class into the legend, and we remove road network from the map to make urban sections appear clearly and solve the color unmatching.
Font size in all maps was not large enough in the original version. We increased the font size of all Figures to make them clearer and easier for readers. However, toponyms are conserved to avoid the overlapping with other elements on the maps. Otherwise, this article will be published on the electronic version, and we expect that zoom in or zoom out is easy for reading.
Point 4: Figure 3: What about the administrative boundary modification of the Cam Pha city? Explain that how the Figure 2 and 3 contribute to the main objective of your paper. Or may be they are redundant?
Answer 4: We appreciate the reviewer’s insightful suggestion and agree that it would be helpful to demonstrate the size modification of Cam Pha city. However, the boundaries of Cam Pha city have not been changed in comparison to the initial period. By this revision, we add the boundary of Cam Pha city in Figure 3 at the same time, we combine maps that show the size modification of Ha Long city into one map.
We believe that Figure 2 and Figure 3 are necessary, Figure 2 provides the change in terms of policy, a factor chat can affect the urban evolution. In fact, Ha Long city and Cam Pha city have strongly changed following the policy. Figure 3 shows spatial expansion by the policy. These Figures provide information that we believe is necessary for readers.
Point 5: Line 205: The Table 2 is redundant, since it is well known to us
Answer 5: We agree with the suggestion of the reviewer to remove Table 2 because it is well known.
Point 6: Figure 7: It seems that the sizes of land cover class samples for supervised classification are too large, thus the samples may contain many land cover classes rather than one class. This can have important impacts on the accuracy of classification results.
Answer 6: There was a mistake in the caption of Figure 7 that makes readers confused. The samples for supervised classification have been taken more precise and selected for homogeneous land cover. We change the caption of Figure 7: Illustration of land cover categories from colour composite (b4-b5-b7) of Landsat image (28 October 1991): (a) Forest; (b) Rice field; (c) Coal field; (d) Residence; (e) Sea water; (f) Logging land; (g) Mangrove; (h) Rock dump.
Point 7: Lines 330-344: How many samples did you use for the classification of each land cover? Make sure that the samples are representative and sufficient.
Answer 7: We thank the reviewer for pointing this out. The number of samples for SVM classification has been presented at lines 292-293: “Each sample set corresponds to 10 samples with the average of 200 pixels/sample for each land cover category (Figure 7)”. The number of samples for accuracy assessment have already been mentioned at lines 347-348: “Each set consists of 10 samples with the average of 200 pixels/sample for each land cover category”.
Point 8: Figure 11: Put the corresponding Landsat images rather than the boundary of pixels here to better compare your classification results.
Answer 8: We agree with the reviewer, and we put the Landsat image in the background of column II for comparing with the SVM classification results issued from Landsat image (column I). Its outline is overlaid on Google Earth image with a finer spatial resolution (column III). Figure 11 can give an idea about the accuracy of classification owing to the geometric and thematic match of corresponding features.
Point 9: Lines 365-375: It’s better to provide the original land cover maps generated by this paper, too.
Answer 9: We believe that adding a new figure, as the reviewer suggested, would be necessary to present original and combined land covers maps. We did not use the original land cover maps generated by this paper, but the simplified (some land cover classes are combined) maps for the analysis. Also, the original land cover map is too complicated and may be hard to read. Accordingly, we add a new figure (Figure 13) to illustrate the original and combined land cover maps. The explanation of combining classes is added at line 391.
Point 10: Lines 378-404: Section 5.1. Change detection: Further analysis is needed. Maybe you can provide the transfer matrix to demonstrate the land cover changes in more detail.
Answer 10: We appreciate the reviewer’s insightful suggestion and agree; For change detection, we have thoroughly analysed at lines (395-397). As mentioned above, we are using the simplified land cover map rather than the original land cover maps in this paper. The change detection matrix of simplified land cover classes (deduced from 3 classified images) are displayed in Table 4 and Table 5 to show the changes. We think that it is redundant to add a transfer matrix in this section, but we add several paragraphs at lines (395-397) and change the captions of Table 4 and Table 5 for clarifying the recommendation of reviewer.
Point 11: Table 7: It’s redundant because there are too many missing data.
Answer 11: We agree with the reviewer that it is redundant if too many missing data. When we prepared the manuscript in the middle of 2021, the statistic data of 2020 were unavailable due to the covid-19 pandemic. We have just found new data for missing elements, and we have added them into the Table 6 at line 469.
Point 12: Lines 451-485: What’s the relationship between this section and the previous results? More detailed descriptions are needed to be added.
Answer 12: In order to link this section with the urban evolution of the previous results and recommend for improving the land use planning of the study area, we change the title of section 5.3 Analysis of the urban development, and we add several new paragraphs in this section (lines 470-489) for detailed descriptions.
Point 13: What is your suggestion for improving the land use planning in the study area?
Answer 13: We appreciate the reviewer’s insightful suggestion and agree. We added our suggestions for improving the land use planning at lines (501-515).
Reference
- Rivkin, L.R.; Santangelo, J.S.; Alberti, M.; Aronson, M.F.J.; Keyzer, C.W.d.; Diamond, S.E.; Fortin, M.J.; Frazee, L.J.; Gorton, A.J.; Hendry, A.P.; et al. A roadmap for urban evolutionary ecology. Evolutionary applications. 2019, 12, 384-398, doi: https://doi.org/10.1111/eva.12734.
Author Response File: 
Author Response.docx
Reviewer 2 Report
This article, seems to me, be correctly arranged, and clearly shows the applied methods of analyzing different spatial forms of urban land expansion. I suggest only a few corrections in the text:
in line194 I would add: parameters from Landsat images are...,
perhaps Figure 4 should be moved to Chapter 3.1.1. where the image is described,
I think that with the cited authors it is not necessary to write in parentheses the year when the number of the cited work was entered,
I would rather write 248 hectares per year than year-1 in Chapter 5.1. if it has that meaning.
Author Response
We would like to thank the reviewer for careful and thorough reading of this manuscript and for the thoughtful comments and constructive suggestions, which help to improve the quality of this manuscript. Our response follows:
Point 1: in line 194 I would add: parameters from Landsat images are...,
Answer 1: We agree with reviewer to add this part into paragraph for clarifying the sentence at lines (199-200).
Point 2: perhaps Figure 4 should be moved to Chapter 3.1.1. where the image is described,
Answer 2: We agree with reviewer to move this Figure 4 to the page that is located after section 3.1.1
Point 3: I think that with the cited authors it is not necessary to write in parentheses the year when the number of the cited work was entered,
Answer: We think that there are different citation methods depending on the regulation of each journal. But we agree with reviewer to remove the year in parentheses for Remote Sensing journal.
Point 4: I would rather write 248 hectares per year than year-1 in Chapter 5.1. if it has that meaning.
Answer 4: We think that there are different unit presentations depending on the regulation of each journal. For Remote Sensing journal, ha/year is accepted in published article (https://www.mdpi.com/2072-4292/14/4/844). We change all mentioned issues to ha/year.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
I have no more comments.