Exploring the Spatial–Temporal Analysis of Coastline Changes Using Place Name Information on Hainan Island, China
, Fei Zhao 1,2,*, Zhiyan Peng 1,2, Lu Song 1,2, Shucheng Tan 1,2 and Zhifang Zhao 1,2Round 1
Reviewer 1 Report
The manuscript Exploring the spatial-temporal analysis of coastline changes using place name information on Hainan Island, China presents a very interesting assessment to coastline studies, However Still needs to improve the current state before its publication.
The Figures, graphics and tables should be presented with higher resolution, with same size words, including missing information for each of the maps showed (coordinates, legends). The captions for each figure are poorly described. Each figure must have a caption with all the details presented, which are missing.
Fig 3, 4, 5 captionmust be rewritten and revise the resolution and figures size letters.
The English text still needs a moderate revision to be suitable for publication.
Fig. 8 coordinates are poorly visible.
The 2.2.4. Coastline semantic generation section must be rewritten, as important aspects are missing to understand how coastline will be automatically classified according to the type of place name points.
The discussion should compare the results of other previous studies on coastline extraction for other areas worldwide, globally, not local results.
The manuscript still show comparison with 2 local research at Hainan Island:
Li, G.; Wang, R.; Chen, H.; Tang, J.; Li, B. Changes and causes of coastline in southern Hainan Island. Marine 532geological frontier 2018, 34, 48-54.53337.
Bao, M. Remote Sensing Monitoring and Change Analysis of Hainan Island Coastline in Recent 40 Years. 534Master's, Inner Mongolia normal university, 2014.
Conclusion must be rewritten highlighting the fundamental findings of the work.
Author Response
Responses to Reviewer 1’s Comments:
Comments and Suggestions for Authors:
The manuscript Exploring the spatial-temporal analysis of coastline changes using place name information on Hainan Island, China presents a very interesting assessment to coastline studies, However Still needs to improve the current state before its publication.
Response: Thank you for your suggestions and comments on previous manuscripts, which greatly improved my writing ability. In this revised manuscript, I have revised it according to your comments. The following is the specific revised content.
Comment 1: The Figures, graphics and tables should be presented with higher resolution, with same size words, including missing information for each of the maps showed (coordinates, legends). The captions for each figure are poorly described. Each figure must have a caption with all the details presented, which are missing.
Response 1: Thank you very much for your comment. I'm sorry for these mistakes. In the revised manuscript, I rewrote the caption of each figure and table, and added the detailed description of the figure and table. The resolution of the figure is improved, and the font and size in the figure are equal.
Comment 2: Fig 3, 4, 5 caption must be rewritten and revise the resolution and figures size letters.
Response 2: Thank you very much for your comment. In the revised manuscript, I rewrote the captions of Figures 3, 4 and 5, and redrawn the figures, which improved the resolution of the captions and ensured the consistency of font sizes in the captions.
Comment 3: The English text still needs a moderate revision to be suitable for publication.
Response 3: Thank you very much for your comment. I am sorry that there are some problems in my English expression. In the revised manuscript, we reorganized the English grammar and expression of the article, and invited English editors to polish the article.
Comment 4: Fig. 8 coordinates are poorly visible.
Response 4: Thank you very much for your comment. I am sorry for this problem. In the revised manuscript, Figure 8 has been redrawn, and the contents in the picture have been rearranged, which improves the resolution of the figure, makes the figure more beautiful and makes the coordinates clearer.
Comment 5: The 2.2.4. Coastline semantic generation section must be rewritten, as important aspects are missing to understand how coastline will be automatically classified according to the type of place name points.
Response 5: Thank you very much for your comment. In the simulated coastline preliminarily generated by TIN in 2.2.2., the two endpoints of each line segment are assigned with the corresponding toponymic point type, which is also an important guarantee for optimizing each line segment by using buffer. After optimizing the simulated coastline, these attributes still exist. Based on these attributes, Python program is used to analyze the generated simulated coastline line by line, and coastline semantics are generated according to the principle described in 2.2.4. In the revised manuscript, this part has been supplemented.
Comment 6: The discussion should compare the results of other previous studies on coastline extraction for other areas worldwide, globally, not local results. The manuscript still show comparison with 2 local research at Hainan Island.
Response 6: Thank you very much for your comment. I'm sorry for such a problem. In the revised manuscript, in the discussion section, I added a comparison between this study and the relevant coastline extraction research done globally, instead of being limited to local areas, and described the differences between this study and other research methods and data used.
Comment 7: Conclusion must be rewritten highlighting the fundamental findings of the work.
Response 7: Thank you very much for your comment. The most outstanding contribution of this study is that in the absence of regional remote sensing images, the proposed method can simulate the basic trend of coastline through place names. Moreover, in areas with remote sensing images, the method proposed in this study can provide a simpler and more effective method to extract coastline with certain accuracy. In the revised manuscript, the highlights of this study have been re-described in the last part of the conclusion.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The current version of the paper is ameliorated.
General remarks:
figures 8, 9 and 10 should be enlarged.
English editing is needed.
Author Response
Responses to Reviewer 2’s Comments:
Comments and Suggestions for Authors:
The current version of the paper is ameliorated.
Response: Thank you for your suggestions and comments on previous manuscripts, which greatly improved my writing ability. In this revised manuscript, the captions of the figures and tables in the text were revised to make the descriptions of the figures and tables more detailed, and the figures 2, 3, 4, 5, 6, 8, 9, 10 and 11 were redrawn to ensure the consistency of fonts and sizes in the figures and improve readability. Besides, we reorganized the English grammar and expression of the article, and invited English editors to polish the article. Finally, thank you again for your valuable comments.
Comment 1: Figures 8, 9 and 10 should be enlarged.
Response 1: Thank you very much for your comment. In the revised manuscript, I redraw Figures 8, 9 and 10, re-typeset the maps in the pictures, and unified the font size, which improved the readability and clarity of the pictures.
Comment 2: English editing is needed.
Response 2: Thank you very much for your comment. I am sorry that there are some problems in my English expression. In the revised manuscript, we reorganized the English grammar and expression of the article, and invited English editors to polish the article.
Reviewer 3 Report
This latest version has improved a lot over the original version.
Author Response
Responses to Reviewer 3’s Comments:
Comments and Suggestions for Authors:
This latest version has improved a lot over the original version.
Response: Thank you for your suggestions and comments on previous manuscripts, which greatly improved my writing ability. In this revised manuscript, we revised the conclusion and highlighted the highlights of this study. The most outstanding contribution of this study is that in the absence of regional re-mote sensing images, the proposed method can simulate the basic trend of coastline through place names. Moreover, in areas with remote sensing images, the method proposed in this study can provide a simpler and more effective method to extract coastline with certain accuracy. Besides, we reorganized the English grammar and expression of the article, and invited English editors to polish the article. Finally, thank you again for your valuable comments.
Author Response File: Author Response.pdf
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
Round 1
Reviewer 1 Report
This manuscript presents a very interesting assessment for coastline extraction, but from my point of view, this work needs a complete revision before its publication.
The Figures, graphics, and tables should be presented with better quality, with visible letters, including missing information for each of the maps, showed (coordinates, legends). The captions for each figure are poorly described. Each figure must have a caption with all the details presented.
Data and methods used in the research are very poorly described in both sections 2.2.2. as 2.3 should be rewritten with more scientific details to explain the algorithm analysis and parameters.
Results are not well described. The figures and tables presented should be described with more detail in the text overall for section 3-Results.
The discussion should compare the results of other previous studies on coastline extraction for other areas worldwide, globally, not local results.
The conclusion must be written highlighting the fundamental findings of the work, but not by comparing with other studies (e.g. Mandelbrot, 1967).
Reviewer 2 Report
The paper deals with the use of TIN and python Programming for the reconstruction of the coastline in areas where remote sensing data are missing.
The idea is promising but the paper presents serious drawbacks.
- Title:
The title is confusing. Exploring the coastline changes using place name information and remote sensing data….. This is not the case. In the paper the place names are mainly used, TIN and pythons are the methods and remote sensing data are used for the accuracy control.
- Abstract:
The abstract is too general. It should be rewritten.
- Introduction:
It should be also rewritten providing the necessary references for the shoreline extraction. There are hundreds of studies about the use of remote sensing data for shoreline extraction. Furthermore, there are traditional methods and data like topographic maps of in situ surveys. I suggest to the authors to read some related review papers.
- Methodology:
The methodology section should be expanded and figure 2 has to be redraw from the beginning as it gives no useful info.
- Section 2.3.2 Coastline Retrieve:
It is not clear to the reader the way that the TIN worked out. How the buffer zone was selected? Why in all the examples the authors refer to a settlement (point 1 in figures 4) and to an enterprise (point 2 in figures 4)? Why they didn’t test any other combination (for example settlements to settlements)
- Results:
Figures 5 and 6 have no scale. The reader cannot understand anything about the accuracy of the simulated coastline.
- Discussion:
The results are validated in four different periods 1980-1989, 1990-1999, 2000-2009, 2010-2018. But for these periods there are free satellite data (Landsat program started back to the seventies). Why the authors didn’t use satellite data for the calibration? Which is the usefulness of the proposed methodology?
Reviewer 3 Report
The most important questions of the article are:
- Where do you get TIN points?.
- Where do you get points 1 and 2 in figure 4?.
- What are place names used for?.
The article has formatting errors, some are indicated in the review.
Abstract:
Line 12: “The temporal and spatial resolution is high, but the coastline before the advent of remote sensing technology cannot be obtained”. What is the reason why it cannot be used?
Line 17: “pre-founding(before”. There is a space before the (
Line 19: “This method realizes the coastline construction under the condition of remote sensing image 19 missing”. So is not going to use any satellite image ?.
Keywords:
It is correct.
- Introduction
Related to the first paragraph of the introduction, a figure should be made indicating the geographical location of the place of study. Is the study all over the island or in some areas of the island? The people who read this article, we must get an idea from the beginning of what the place or places of the investigation are like. Possibly in this section figure 1 (line 102) could be included. In this figure, the place where the investigations are to be carried out should be indicated, indicating it with a simple point.
Figure 1. Line 102. When putting the map of China I would advise putting only the map with its land territory, this is how all maps are made. Maritime limits are never indicated.
I don't understand why lines 43 to 52 are included, speaking of place names. What sense does it have with the environment of the article? With the information indicated (place names) it refers to what is indicated in tables 5, 6, 7, 8, but is this information obtained from the urban information of the city?.
- Study Data nd Methods
Line 95: missing an "a" in and: Study Data and Methods
2.1 Overview of the Study Area
I would delete this section and include it in the Introduction.
Removing the 2.1 Overview of the Study Area, which can be included in the introduction, section 2, could look like, which is the title of section 2. Study Data and Methods:
2.1. Study Data
2.2. Methodology
Line 110: lon and lat, put it as longitude "lon" and latitude "lat"
2.2.1. Hainan Place Names Information Data
The article does not mention any place name obtained before 1949 (the founding of the People's Republic of China), and from where this information was obtained: Cartography, books, talking about older people of the place, ... Or this refers to the information obtained in the Ministry of Civil Affairs of China during 2014-2018 (line 108), but is there information prior to that period, since it is data from 2014-2018, which has nothing to do with what it could be 60 years ago , and that is the objective of the article?.
In table 1, when it indicates: Place history (1950), what does it refer to?
2.2.2 Coastline Data of Hainan Island
Line 121: Where do you get this information: satellite images, cartography, ...? Modern coastline: 1987-2017: 30 years?.
Visual interpretation and digital vectorization: Cartography, satellite image? What scale of cartography, what satellite image, year, ...?
2.3 Methodology
Line 126: For now I still don't understand what place names are for, for determining the coastline. I ask: Give me some place names. What is the relationship between toponyms and the TIN?.
2.3.1 Place Name Data Clustering
What year is taken as the origin for the study: 1949 or 1980? What documentation (cartography, toponymy, ...) is available from 1949?
Line 141: “there are some obvious place types that did not exist until after the founding of the People's Republic of China, such as park, port, hotel, etc. Question: hotels in 1949 or 1980?
2.3.2 Coastline Retrieve
This question seems very important to me: What is the origin of the coordinates of the vertices of the TIN?.
What is the need to do the entire TIN of the island if only the coastline is to be analyzed?
Figure 4: Important question: What is the origin of points 1 and 2 in the first figure (a)?
Line 161-177: How do you divide the coast according to the place name? Is this work done in the field or looking at old photographs, cartography, satellite images? The abstract indicates that this article seeks a methodology where satellite images are not used, but there are figures where these images do exist. If not, tell me: how was the true coastline of figure 5 obtained? Is it the digitization of a satellite image, indicate the year?.
Line 184: Formula 1. I don't understand what B, B1 and B2 are. Are B1 and B2 the distances from point 1 and 2 to the coast? B is the mean of these distances. What is the objective of calculating B, since it does not have to coincide with the tangency to the coast as pointed out in figure 4?.
2.3.3 Accuracy Verification
The interpretation of this paragraph could be indicated by a figure, so that the reader can interpret the explanation by an image.
2.3.4 Coastline Semantic Generation
2.3.5 Coastline Spatial-Temporal Analysis
Line 222: Remove the space at the end, before the period: causes.
- Results
3.1 Place Names Generate Shoreline Availability
Line 231: Figure 5: The
Line 233: Table 3: The
In line 193 it is indicated: "The coastline was obtained by remote sensing interpretation using 30m resolution Landsat TM / OLI images.", Therefore the intention should look for values in the setback of the coastline similar to 30 m. But in Table 3 it is shown that with a precision of 1500 meters 91.7% has been detected, therefore there is another 8.3% that has accuracy values higher than 1500 m.
These values are given by the low density of points of the TIN, as can be seen in figure 7. May I comment that I do not know where these vertices of the TIN are obtained from?.
3.2 Coastline Change Result
3.2.1 Cities’ Change Area
Line 240: You are missing the end point.
In figure 6 it should be indicated that Haikou City is the top image and Sanya City is the bottom image.
If images are being compared, then the dates of the images being compared should be indicated and if they are not images, then what is being compared must be indicated between which dates the comparison is. In any case, you must put the dates.
In the figure it says: Morden and it's Modern. A scale should be set.
Change area: It is not necessary to put 6 decimal places to km2. Are the areas of profit or loss?.
For example, in Google Earth if you select a satellite image from 1985 and another satellite image from 2021 there are no great variations in the coastline. I do not understand such large variations, the blue line: pre-founding Where is it obtained from? That's where such large variations come from.
3.2.2 Coastline Type Change
When it comes to comparing with "the founding of the People's Republic of China", is there graphic information: satellite image of that time (1949)? There were no satellites in those years, until 1957 (sputnik).
Table 5 appears in the middle, it is cut on the right.
Table 6 goes far beyond the margins of the article.
Neither ratio adds up to 100%, but the Haikou (modern) ratio adds up to only 76.4%, where is the rest up to 100%?.
Table 6: Is it logical that in pre-founding there is 16.3% of entertainment-port coastline and currently it does not exist ?, and like this, other denominations or areas.
- Discussion
Figure 7: The TIN should have triangles with smaller sides to better fit reality. This figure shows that the adjustment between the TIN (I don't know where it is obtained from) and the real digitization of the coast on a satellite image? it is not correct.
Lines: 278-280: Where do you get this information?.
Line 289: Continuously throughout the article there is talk of: “before the founding of the People's Republic of China”, which is the year 1949, but then all the information is referred to not before 1980 (table 7).
Line 291: “From the 1880 s to the present, vigorously developing tourism”. ¿1880? it won't be 1980.
As an example, a map should be indicated where the percentages for each area studied should be indicated. For example, what was once a rural coastline has now been transformed into a port.
- Conclusions
If you want to compare the coastal retreat, then you must select elements of the same coast or comparable elements in the two times that you want to compare. It is not feasible to select different items at different times.
