Assessing Land Subsidence-Inducing Factors in the Shandong Province, China, by Using PS-InSAR Measurements

Round 1

Reviewer 1 Report

Dear Authors,

I reviewed the present manuscript recently. I attempted to provide you with some suggestions to improve the strengths of the manuscript. I recommended major revision twice before considering the present manuscript suitable for publication in Remote Sensing. I remember that the first revised version did not integrate reviewers’ suggestions.

 

For what concerns this “second” revised version of the manuscript, I have to say that the text has been significantly improved, both from the contents and the scientific soundness of your findings.

 

Therefore, now I recommend your paper for consideration for publication pending minor revisions. They are as follows:

0) I do not agree with the manuscript title. You should include the keyword “inducing factors”. Perhaps you may use “Assessing land subsidence inducing factors in the Shandong Province, China, by using PS-InSAR measurements”.

 

1) line 32: use “Moreover” instead of “However”;

 

2) line 34: insert “may” between “findings” and “provide”;

 

3) line 41: insert ref. 36 in addition to [8-10] (before reference 8 or after reference 10).

 

4) line 42: insert ref. 36 in addition to [11] (before or after).

 

5) Equation 3: introduce P before introducing W_exploitation and W_storage.

 

6) Line 406: Add the following statement:

“Nonetheless, specific infrastructural surveys should be carried out on the bridge to investigate whether the deformations observed are due to an effective lowering of the bridge deck or to a displacement of the layers of the road pavement. Accordingly, it is not possible to know with certainty the causes of the bridge movements without further in-depth analysis, such as by surveying the bridge deck with Ground Penetrating Radar. Indeed, as recently demonstrated by Fiorentini et al. [37], the two surveys (Ground Penetrating Radar and PS-InSAR) can be integrated. However, using only one does not exclude the use of the other, as infrastructural distresses are not completely detectable by the PS-InSAR survey, especially by exploiting the resolution of Sentinel-1."

 

7) line 421: insert ref. 36 at the end of the statement;

 

8) line 429: insert ref. 37 at the end of the statement;

Author Response

Detailed answers and responses can be found in the following word file.

Author Response File: Author Response.doc

Reviewer 2 Report

Comments for author File: Comments.pdf

Author Response

Detailed answers and responses can be found in the following word file

Author Response File: Author Response.doc

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

Review:

This paper uses InSAR technology to monitor and analyze the surface subsidence in Shandong Province, which has certain research significance. It is difficult to process Sentinel-1 data in a large range, but there are still some problems in this paper, as follows:

• In this paper, SRTM DEM with a resolution of 90m is used to remove the terrain phase. To improve the accuracy, it is suggested to use DEM with a resolution of 30m or even higher.
• PS-InSAR technology is really effective for monitoring land subsidence in urban areas, but the density of coherent target points extracted by PS-InSAR technology is low in non-urban areas, which leads to large errors and even wrong unwrapping. In this paper, the surface subsidence in Shandong Province is monitored and analyzed, and most of the areas are non-urban areas. Therefore, only PS-InSAR technology has serious defects. Please increase the monitoring results of other sequential InSAR methods, such as SBAS-InSAR or DS-InSAR.
• Only 9 GPS monitoring points were selected for verification and analysis in Shandong Province, which can't well show that the results of this paper are reliable. It is suggested to select more GPS monitoring points for verification and analysis.
• The maximum settlement rate in Figure 3 is -296.8mm/year. Why is the annual maximum settlement rate in Figure 4 greater than -296.8mm/year? And the position of the maximum settlement rate in the two figures is different, please explain the reason.
• Are there any problems in the two graphs corresponding to Tuck 69 and Tuck 40 in Figure 7? Why does one point of one track correspond to multiple points of the other track? Please explain why.
• The existing Figure 9 and Figure 11 have little significance. In order to show the deformation area and degree more clearly, please superimpose and display the InSAR deformation monitoring results.
• The analysis of the factors causing land subsidence in the study is insufficient, which is not enough to show that groundwater and mineral resources development are the main factors leading to land subsidence in Shandong Province. Please select typical areas to analyze the relationship between groundwater level change and land subsidence, and the relationship between mining face and land deformation.
• There is a large uplift area in the InSAR monitoring results in this paper. Please explain the reasons.
• Whether "50mm/year" in lines 348 and 378 should be "-50mm/year", please check.

Author Response

Responds to the reviewers’ comments can be found in the following word file

Author Response File: Author Response.doc

Reviewer 2 Report

The manuscript “Land Subsidence in the Shandong Province (SDP), China, Determined by Sentinel-1 Datasets” presents land subsidence in Shandong using PS-InSAR technique with Sentinel-1 images from 2017 to 2020. The structure and logic are clear. However, throughout the paper, it is a simple application of the exited mature method without innovations.

The method briefly introduced a simple merging solution of different frames, which has little research significance. After that, the relationship between subsidence and possible triggering factors is discussed, mainly including groundwater exploitation and underground mining, but the analysis is simple and qualitative, and it is difficult to be convincing based on the urban optical images and plain text descriptions. Besides, the subsidence contour maps (in Fig.8 & Fig. 10) seem to be different from a surface subsidence contour in common sense. Some post-processing needs to be done to enhance this result and make it look more like a contour map. Finally, the settlement of Jiaozhou Bay Bridge is analyzed, which is also plain and shallow.

Therefore, it is suggested to improve the method and make the analysis more quantitative. What’s more, the quality of English needs to improve.

Author Response

Responds to the reviewers’ comments can be found in the following word file

Author Response File: Author Response.doc

Reviewer 3 Report

Land Subsidence in the Shandong Province (SDP), China, Determined by Sentinel-1 Datasets

By Fengkai Li, Guolin Liu, Huili Gong, Beibei Chen and Chaofan Zhou

 

The manuscript discusses the analysis of land subsidence in the Shandong Province. Moreover, the authors attempted to find the reason for such subsidence occurrences.

The paper is decently written (even if there are some typos) and well structured. Figures and Tables are pretty straightforward; generally, they are not discussed in detail. Instead, authors should put much effort into presenting their graphical outcomes.

 

Furthermore, in my opinion, the present paper does not introduce sufficient novelty in the field. We can find tens of papers related to PS-InSAR processing and how subsidence effects are related to overexploitation of groundwater. The author should better explain the research gap that the present paper attempts to fill out.

 

Moreover, I have some minor and two major concerns about the paper. Please read the following comments.

• L20-21: "Finally, the response relationships between the land subsidence and its inducing factors in the SDP were analysis". Do you mean "analyzed"?

 

• L36: "… has occurred in many countries on the earth [1,2]". You should add more than two references if you stated "many countries on the earth". Please add some references concerning Europe (e.g., you can find significant surface motion in the Tuscany Region, central Italy), America, and Africa.

 

• L130-132: "Five hundred and fifty-eight Sentinel-1 images collected from the ascending orbit between March 2017 and December 2020 were used to derive land subsidence information in the SDP.":

1. Do you think that adding descending orbit information to your analysis may improve your outcomes?
2. Did you have the availability of such data?
3. Do you think this may be future research?
4. What are the weaknesses of your method that do not consider such data?
5. You are observing the displacements in the LOS direction. Subsidence and/or uplift are occurrences that manifest in a vertical direction concerning the earth's surface. How can you derive that such movements are related to subsidence and not, for instance, to landslides? Equation 1 relies on Vlos concerning ascending orbit. If you consider both ascending and descending orbit, I would expect that you observe different V applying Equation 1 to the same pixel.

You should acknowledge this issue and introduce a paragraph with the answers to all the questions above, especially questions "4" and "5".

 

• L152 "3.1. PS‐InSAR Processing for the Sentinel-1 Datasests". You should include information on the software you exploited for PS-InSAR processing.

 

• L155: "Frist". Do you mean "firstly"?

 

• L167: "Previous studies have shown that the displacement along horizontal direction throughout the SDP occurs within a velocity of 2 mm/year [15].". What does this statement add to your discussion?

 

• Figure 4 is not well introduced, and I found it hard to read. You should add an appropriate legend to the Figure (Figure 4a=…, Figure 4b=…, Figure 4c =…, and Figure 4d=…) and describe it.

 

• L348-350: "This was mainly caused by the static loads brought by the weight of the JBB and the dynamic loads brought by vehicles on the bridge." I would pay attention to inserting the present statement in affirmative form. Indeed, why does static load of the weight of JBB occur just in the middle of the bridge? This is not a beam with both ends resting! I believe that most of its weight is not concentrated in the middle. Moreover, dynamic loads related to traffic flow should not impact just in the middle. Indeed, it impacts all the bridge. Therefore, you should deeply reconsider your statement and propose some different reasons related to the displacements of JBB.

Author Response

Responds to the reviewers’ comments can be found in the following word file

Author Response File: Author Response.doc

Round 2

Reviewer 1 Report

Review:

Thank you very much for your reply to my review comments.

Although some modifications have been made in this paper according to the suggestions put forward, only some minor problems have been modified, while important problems have not been solved, as follows:

(1) In the reply of opinion modification, the author said that the DEM of 90m has been changed to 30m, which is equivalent to the calculation of PS-InSAR again. However, I think it is impossible to complete the calculation of InSAR in Shandong Province in such a short time. The author just changed "90m" to "30m". If the author really used 30m DEM to complete the calculation of InSAR, please compare the results of 30m and 90m, and the InSAR results obtained from 30m DEM and 90m DEM are compared with GPS or leveling data.

(2) In clause (2) of my opinion, I don't mean that the results of PS-InSAAR technology are unreliable. I mean that PS-InSAR technology can obtain more coherent target points in urban areas, but the coherent target points extracted in non-urban areas is less. In this paper, only PS-InSAAR technology is used for deformation monitoring, which leads to the inability to obtain deformation information in non-urban areas, and the obtained results cannot fully reflect the deformation situation in the study area. Therefore, it is necessary to combine other sequential InSAR technologies for monitoring and analysis, and the author has not modified it.

(3) I still think there is a problem with the second and third graphs in Figure 7. I understand what the author is trying to say, but I think the correlation graph should have a "one-to-one" relationship, while the second and third graphs in Figure 7 obviously have a "one-to-many" relationship.

(4) Figure 9 and Figure 11 still haven't been modified as required. Using isoline means that all deformation values within isoline are equal? This representation method is not feasible, so it is strongly recommended to superimpose InSAR points.

(5) The author says that the main purpose of this study is to analyze the relationship between land subsidence and its influencing factors in Shandong Province, but there are not enough arguments in this paper to support that groundwater and mineral resources exploitation are the main factors that lead to land subsidence in the study area. This conclusion cannot be reached only by groundwater and mining location. Please further collect groundwater level data and mining face for analysis.

If the above problems are not solved, I think it will not be published in this journal.

Reviewer 3 Report

Dear authors,

Thanks for considering my comments.

As far as I am concerned, the paper has improved from solely the point of view of the English style.

Unfortunately, the authors attempted to address my concerns but did not report the answers in the manuscript. This makes the manuscript very similar to the previous one, except for the English style.

Below, I have reported some points of fundamental importance that must necessarily be considered by the authors revising the manuscript’s text.

 

First essential issue:

For example, in my previous review, I advised authors to reconsider their statement in lines 372-374: “This is mainly due to the soil compression caused by the static loads brought about by the weight of the JBB and the dynamic loads brought about by vehicles on the bridge “.

Unfortunately, the authors left the statement unchanged. Therefore, the authors cannot be sure of these reasons. The authors can only presume that these aspects may cause the major deformations observed over the bridge.

Specific infrastructural surveys should be carried out on the bridge to investigate whether the deformations observed are due to an effective lowering of the bridge deck or to a displacement of the layers of the road pavement. Accordingly, it is not possible to know with certainty the causes of the bridge movements without further in-depth analysis.

Recent research that integrates specific infrastructural analyzes (Ground Penetrating Radar and PS-InSAR) is: https://doi.org/10.3390/s21103377

In such a study, the authors recognize that the two surveys (Ground Penetrating Radar and PS-InSAR) can be integrated. Nonetheless, using only one does not exclude the use of the other, as infrastructural distresses are not completely detectable by the PS-InSAR survey, especially if you are using a low-resolution satellite such as the Sentinel satellite. Accordingly, you cannot derive your conclusions (and state that they are truthful) by analyzing just PS-InSAR measures.

The authors must recognize (and report in the text) the issues mentioned above and acknowledge the possible analyses I proposed (and report them in the text) to recognize the real causes of the detected damage.

 

Second important aspect:

In the previous review, I asked the authors to specify well research gaps that this manuscript attempts to overcome. The authors replied but did not integrate the text.

 

Third important aspect:

In one of their responses to my concerns, the authors stated: “The main purpose of this study is to analyze the relationships between land subsidence and its influencing factors across the whole Shandong Province (SDP)”.

 

They actually processed data with the PS-InSAR algorithm and qualitatively correlated the deformations to possible causes through results obtained by others in the literature.

Indeed, a relationship has not been identified (a mathematical formula, a model, an algorithm) that links observed subsidence with hydrological parameters (groundwater exploitation), land use parameters (mining underground activities), and infrastructural ones (traffic flow of heavy vehicles).

Authors must recognize these shortcomings and propose a new subsection at the end of the discussion chapter (called “Future works”) to include what has not been done and what can be implemented later.

As supporting references, some recent articles show that the causes of subsidence can be identified and linked to PS-InSAR measures through machine learning models. Please refer to:

• https://doi.org/10.3390/rs12233976

In such a study, by implementing some machine learning algorithms, the authors find a good correlation between subsidence and uplift effects by linking PS-InSAR measures and several environmental factors, such as topography, hydrography, geomorphological features, land use information, etc.

 

Moreover, in the “Future works” subsection, the authors should report shortcomings expressed in my comments from the first review, such as:

1) adding descending orbit information to the analysis (and in which way it may improve the outcomes);

2) the author should report in the “Future Work” also the following statement:

“The spatial resolution of Sentinel-1 images is too low, which limits the density of PS points detected in study area. We will obtain the land subsidence InSAR monitoring results covering the whole SDP by using SAR images with higher spatial resolution in our future study. Previous studies have shown that land subsidence lags behind the exploitation of groundwater. Detailed information can be found in the paper entitled “Land subsidence lagging quantification in the main exploration aquifer layers in Beijing plain, China”. The change of groundwater level in SDP was small form March 2017 and December 2020. More information can be found in “Shandong Province Bureau of Statistics (http://tjj.shandong.gov.cn/col/col6279/index.html)”. The short span of study period limits the analysis of the relationship between groundwater and land subsidence. We will quantitatively analyze the relationship between groundwater level change and land subsidence in our subsequent study. This scenario also applies to the relationship between land subsidence and underground resource exploitation. “

 

They reported this statement as an answer to one of my concerns. But, once again, they did not rearrange their text.

 

Therefore, in order to consider this manuscript for consideration for publication in Remote Sensing, the authors should put much more effort into their revisions, not just answering reviewers ’concerns.

 

Kind Regards