Understanding the Influence of Building Loads on Surface Settlement: A Case Study in the Central Business District of Beijing Combining Multi-Source Data

Round 1

Reviewer 1 Report

1. This resubmitted manuscript looks like just some minor changes based on their previous reversion. 

2. The authors still need to define different settlements well in this paper. Land subsidence is related to regional land settlement (check latest publications: https://0-doi-org.brum.beds.ac.uk/10.1038/s41598-020-74696-4 and Galloway, D. L., Jones, D. R. & Ingebritsen, S. E. Land subsidence in the United States. (U.S. Geological Survey, Circular 1182, 1999).); foundation settlement is clear in this papers; surface settlement is not defined well in this paper. It looks surface settlement would be that settlement of land surface around building impacted by foundation settlement. But surface settlement also could be "total settlement". 

3. The settlement detected by InSAR in Fig. 4 would be total settlement which includes regional land subsidence, foundation settlement under building, and surface settlement around building. The authors need to clarify and quantify each type of settlement.

Author Response

Dear  Reviewer 1：

Detailed revisions can be found in the word file

Author Response File: Author Response.docx

Reviewer 2 Report

Dear Authors,

The revised version of the manuscript answered part of the questions that I've raised. These regarded mostly the critical ones. Despite this, I still believe that more work has to be done in order to provide tangible support based on solid geotechnical analyses (experimental/theoretical). Although I consider this as the major limitation of this manuscript; the approach authors used seems interesting if it is applied on single buildings. Accordingly, I recommend its publication.

Regards,

Reviewer

Author Response

Dear  Reviewer 2:

Detailed revisions can be found in the word file

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Line 21: the surface settlement of Beijing. The settlement in this paper is only for one district of Beijing. In other districts, land subsidence due to groundwater withdrawal is significant. It could be changed to "the surface settlement in the study area" or others.

Line 56: which include should be which includes.

Lines 55 to 61: The reason for why surface settlement is used in this paper instead of land subsidence is not justified well here. Both surface settlement and land subsidence are total land elevation change in quantity. Surface settlement would be more appropriate for land subsidence due to foundation settlement impacts in this study area than for land subsidence due to regional tectonics and groundwater withdrawal in other regions. To add a good justification is suggested.

Lines 45 and 47: The subsidence citied from references may not be surface settlement as defined in this paper. A check for more accurate citation through the whole paper is suggested.

Equation (1): Equation (1) doesn't work well for the area without building where surface settlement is impacted by FS of surrounding building(s). Modification sounds necessary.

Line 239: 4.1. Surface settlement Measured by PS-InSAR. Very good.

Author Response

Detailed revisions can be found in the word file.

Author Response File: Author Response.docx

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 is a revision of “Understanding the influence of building loads on land subsidence: A Case Study in Central Business District (CBD) of Beijing by combining multi-source data”. In this revision, surface settlement replaces land subsidence in the previous version. Foundation settlement is added. A regional hydrogeological profile (Gig. 2) and two geotechnical borings (supplementary tables 1 and 2) are added. There are major weaknesses as follows:

1. The PS-InSAR detected deformation still needs a further verification. Compared to the land rebound of 1 cm/year in Haidian District was identified by Leying Hu et al. (2019) with Sentinel-1 InSAR, the rebound of 2.0 to 2.4 cm/year in Fig. 4a is too big. Hu et al. employed GPS points. The authors should use GPS measurements for the further verification other than on the benchmark observations in Figs. 1a and 5. At same time the regional rebound in Fig. 4a needs a good interpretation: why does the so significant rebound occur in such areas with compressible aquifer systems? The tectonic plate is moving up in this region?
2. The authors need to review concepts of land subsidence, surface settlement and foundation settlement well. Surface settlement is caused by engineering construction, which includes foundation settlement. Land subsidence would be the sum of all components of land surface election change, which means it can include surface settlement other than tectonic subsidence and aquifer system compaction. A better reference review would help the authors to understand their results and define them well in their manuscript. For example, the surface settlement in the study area should be derived by removing tectonic subsidence and aquifer system compaction. The PS-InSAR detected deformation in this manuscript would be land subsidence but surface settlement would be a major part. The term of land subsidence is still used in Fig. 3. The surface settlement would much larger than the current values in Fig 4a if region land rebound exists in Beijing.
3. Hu et al. (2019) and other references showed groundwater withdrawal is the major cause of land subsidence in Beijing. The authors may collect ground level fluctuation data in the study area to verify if aquifer system compaction due to groundwater withdrawal occurred in the study area other than regional groundwater level change in Fig. 1a. Surface settlement in the study area would be more reliable if considering this aquifer system compaction and land rebound well.
4. The use of aquitards in Fig. 2 is not correct because each aquifer includes aquitards. The aquitard layer in Fig. 2 would be confining bed(s) in a hydrogeological profile. The authors may need to check the original profile. The source of the original profile should be indicated with citation.
5. It looks in this manuscript that there exists self-plagiarism and citation is not rigorous. The author published a similar paper in Remote Sensing in 2020. Data source for Supplementary Tables 1 and 2 is not showed well besides Fig.2. There are no citations for the corresponding engineering construction standards on line 235. The citation on the end of this sentence is the author’s paper in which a citation still is the author’s conference paper. 5 is very similar with Fig. 7 in their paper published in 2020 but there is no citation for Fig. 5. Fig. 3 is one copy of Fig. 2 of their previous work without citation. The methodology (equations (1) and (2) without citations) is almost the same as that (equations (4) and (1)) published. The authors should follow the plagiarism and citation policies of the journal. The original reference Boussinesq (1885) for equation (1) is not cited. From respect to the original author (Boussinesq) P_i should be point load of a building on foundation instead of “gravity of building”.
6. The validity of equation (1) in the study area is not justified in this manuscript. The assumptions of Boussinesq equation are not remined in this manuscript and No justification are given to see if the assumption work for the soil conditions in the study area. The estimates of stress increase at depth caused by building loads are not compared to any results assessed by other method(s), which is included in discussion section. Detailed pile foundation data is not given and analyzed for justification of equation (1) used in this manuscript. Probably Equation (1) only works for 6 story buildings. Shallow foundation stress influence depth is about twice of foundation width. Pile bearing layer is 12 m to 20 m deep. Pile foundation stress influence depth would be about 8 to 50 m, which is why the geotechnical boring depth is 50 m in the two supplementary tables. Compressibility modulus values of soils below 50 m deep are available for settlement calculation with equation (4)?
7. The length of discussion section is about 10 out or 20 pages with 7 figures (Figs. 7 to 13) is really inappropriate in scale. The variation of surface settlement from consolidation estimated from compressibility modulus with time is not reliable. The TerraSAR-X radar satellite sensor launched in June 2007 revisits every 11 days (lines 138 to 139). So annual images could be employed to detect annual land deformation which is more scientific than the deformation variation estimated in this manuscript. Reorganization is needed.
8. Settlement value is negative in Figs. 4a and 4b doesn’t make sense. Rebound value would be negative. This is not revised well. The negative values for depth in Fig. 9 also doesn’t make sense.

The detected land deformation would be valuable with further validation and justification with GPS measurements and other findings. Annual land deformation is expected to be detected with the PS-InSAR data. A significantly enhanced manuscript in quality would be expected with above detailed comments. Since so many critical flaws above still exit in the current version, I have to suggest the journal to decline this manuscript again and encourage the authors to resubmit a quality manuscript.

Reviewer 2 Report

The authors have made an effort to improve the earlier version of their paper and to address all remarks that have been made during the former submission of this manuscript. Nevertheless, some major issues still exist that need to be adressed before this paper can be considered for publication.

1) The authors should pay more attention to the specific differences between both sets of characteristic points. They shortly indicate that these two sets have different geological settings but do not further explain these somewhere in the text and only provide some borehole data as supplementary material. They should also explain within the manuscript the relevant differences in the soil strata of the two borehole datasets that differentiate the location of both point sets. In particular, as they conclude that differences in the physical properties of the soils are mainly responsible for the observed differences in some of their results (L. 314-315).

2) Furthermore, a detailed discussion on how the specific geological properties are responsible for differences in the results of these two point sets is almost completely missing in the Discussion section. Moreover, the majority of the Discussion section is still just a description of results and how they were calculated and not a sufficient discussion of the presented results. For example:

- In L. 354-368 the authors only describe the results from figure 14 but do not draw any conclusions from these results. Why are the additional stress gradients for points with high additional stress values have their peak at 20m depth, whereas figure 12 indicates that the generally highest stress gradients are at 10m depth?

- In L. 387-390, it is nowhere discussed what are the reasons for the different behavior of points A and B in the second group compared to the points of the other group?

In my opinion, the Discussion section is still a mixture of things that should stand in the Methodology and Results sections. All descriptions and formulas of the calculations in the Discussion section should stand in the Methodology section and all results should already have been presented in the Results section. A detailed discussion of these results is still almost not there.

3) Figures 6,7 and 8 show that the additional stress is decreasing with increasing depth. In figure 9 and in L. 266-278 the authors describe an increasing additional stress for all 6 Points with increasing depth? That’s a contradiction and does not really make sense to me.

4) In figure 1 the timespan of the displayed groundwater level change should be added. Additionally, the red color of the characteristic points and boreholes are hard to read. Probably a different color would be better here.

5) In L. 199 the mentioned Figure 4b does no longer show the average surface settlement rate. I suggest to additionally add the values of the cumulative settlement rate in the text here.

6) The legends of figures 6, 8 and 12 are still of poor quality and the stress values are hardly readable. The font size of these legends should be increased.

7) What are these nine categories that the authors describe in L. 338-339? It should be further described here how these categories were selected.

8) L. 424-427 only describe what has been done. This is not a conclusion. The authors should better state their findings from the described calculations here.

Reviewer 3 Report

Dear Authors,
I read the updated version of the manuscript finding you solely answered to all my concerns. Moreover, I also checked the suggestions and modifications requested from the others Reviewers and it is my opinion you satisfied all such criticisms as well. 
Therefore, I consider worthwhile the manuscript is suitable for acceptance and piublication in its present form.