Identifying Causes of Urban Differential Subsidence in the Vietnamese Mekong Delta by Combining InSAR and Field Observations

Round 1

Reviewer 1 Report

The manuscript presents an analysis to identify and quantify drivers of processes causing differential subsidence within three major cities in the Vietnamese Mekong delta using satellite data and field observations. Specifically, the authors report only in the result section the correlation between urban differential subsidence and the differences in piled foundation depths. Although the object is clear, it is difficult to appreciate whether it is novelty or scientific advance in this research.

 

Major points:

First, this work doesn’t do any processing related to satellite remote sensing PSI signals. The PSI dataset in the paper is from other sources (i.e., from Copernicus Emergency Management Service - EMS). PSI processing for large scale such as Mekong Delta is really challenge in term of quality. Particularly, it is very difficult in tropic areas where atmosphere is dominated InSAR signals. From EMS, I found there is no validation for such dataset N57 and N62, leading to unreliable dataset. So, it is not the right way to use this one for scientific research. I recommend the authors take time to process PSI from the SLC data and quantify their uncertainty relative to atmosphere propagation before moving forward.

 

Second, the quality of field data in this work is not sufficient to use as a ground truth. As in line 258 Figure 6, the authors can only measure the difference between the building and its land surface surround. This is difficult to choose where to measure because it is not fixed. But after all, the change is the local land surface and not the building. To measure the change of the building, we need to find a stable point (ground control point) and do measure the building with respect to that fixed point using the levelling with high precision. By using this technique, buildings can be monitored. In Vietnam, there is a requirement to monitor the movement until few years as a standard in building construction. I am surprised such data monitoring not available in this work. Another issue for measuring the change local land surfaces is that there is a physical difference between them and PSI scatterers. To summary, using such field data measurement, there is a strong bias with respect to PSI.

 

The quality of section 3 Results is poor. The whole section is only from 292 – 320 line with one figure 10. The objective is to study the correlation between the differential subsidence occurring and the existence of a piled foundation underneath a building. The author reported a strong correlation on the piling depth but I can not find its coefficient in the result report. When we report correlation, at least we show scatter plot with coefficient and p.

 

The Section Discussion is confused and uncorrelated to the section results. In Section 4.1 Piled foundation depths and building sizes, line 361-384: uncorrelated to this work – it doesn’t link to the results part. Section 4.2 Lithology, Line 386, there is no map horizontal spatial variability in lithology and depth-dependent variability. How can the author analyze and discuss on it ?  Section 4.3 Land use and groundwater extraction, there is only examples on land use discussed. No analysis on land use is reported in results.  There is no groundwater extraction information in result also. In Section 4.4 Recommendations for future studies to unravel urban differential subsidence, this is not correlated to the result and this should move to conclusion.

To conclude, it is really confused on the methodology and the contribution of this paper.

 

 

 

 

 

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

Over all a good research work................

Author Response

Please see the attachment

Reviewer 3 Report

de Wit et al. assembled three InSAR-derived surface displacement velocity datasets and field survey data (i.e., field observations and site-specific information of the studied buildings) to statically investigate the causes of the differential subsiding velocities between the individual buildings and their surrounding areas in three cities in the Vietnamese Mekong delta. They found that only the piling depth, out of multiple building characteristics, shows a strong correlation with the urban differential subsidence. Finally, they discussed the contribution of some other possible factors on the differential subsidence.

 

This is an interesting, well-written study of using satellite-based observations and field surveys to identify and quantify the causality of urban differential subsidence. The data collection and analysis are somewhat reasonable, and the discussions and implications are extensively useful. I enjoyed reviewing the paper and think it is very suitable for Remote Sensing publication. I have some detailed comments would like the authors to address before the publication and hope to help them improve the paper.  

 

Line 100: Figure 1b is very useful for understanding the relation between piling depth and differential subsidence by schematically showing the depths of different sediment layers. This explains the different correlations in the three cities (Figure 10). I therefore recommend the authors to add one figure of such stratigraphic column with the thicknesses of different layers for the three cities. A rough or schematic plot will work.

 

Line 159: It may be useful to add the Mekong river in the inset plot. Please use the decimal point in English (i.e., “,” - “.”) in the Figure 2 and all the other following figures.

 

Line 162: Data collection -> InSAR data collection.

 

Line 172: It will be helpful to add some sentences describing the InSAR data processing (e.g., what software were used respectively, how the atmosphere effects were corrected).

 

Line 176: Please add an accessible link for the KIT dataset.

 

Line 183: In the first column of the Table, the links need to be revised “EMSN57” as “EMSN057” and “EMSN62” as “EMSN062”.

 

Line 202: “surroundings e.g. year” -> “surroundings, e.g. year”

 

Line 225: The corresponding value interval for each color in the legend is very confusing. For example, from the top to the bottom, it should be something like “x > 10.0”, “5.0 < x ≤ 10.0”, “0.0 < x ≤ 5.0” and so on. Please revise them in this figure and all the other following figures.

 

Lines 306-316: Using a same color bar in Figure 10 may help the readers understand these results.

 

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 4 Report

I carefully read the manuscript remotesensing-1003660, untitled '' Identifying causes of urban differential subsidence in the Vietnamese Mekong delta by combining InSAR and field observations''. The paper shows a systematic presentation of identifying and quantifying drivers of processes that caused the observed differential subsidence over Can Tho, Ca Mau and Long Xuyen cities in the Vietnamese Mekong delta by utilizing Sentinel-1 C-Band SAR data during 2015 and 2020. The data are solid and conclusions are supported and they present a nice interpretation of several reasons that could be responsible for the differential deformation occurred between buildings and their surroundings through investigating specific building information. They revealed this urban differential subsidence that superimposed on the delta-wide subsidence patterns mainly occurred at buildings with piled foundation. And the piling depths, previous land use, loading of structures without a piled foundation and variation in piling depth are also proposed as important factors determining urban differential subsidence. Moreover, this paper is well structured and written and should be published by this journal. I have only a few comments and recommendations that I would like to address before publication.

1. Please adjust the length of scale of Figure 1 to reach an integer number, such as 100, instead of 98 kilometers.
2. Figure 3: please narrow width of label.
3. Line 269: I should note that InSAR is not a ‘absolute velocity’, it has its own reference point at each imagery.
4. Figure 9&10: How do you determine the sequence of buildings? i.e., why name building A as ‘A’? I s there any rules? If not, why do not you use the building depth as indicator? And utilize the sorted building depth as horizontal axis, which may deliver a better visual effect of the correlation.
5. Figure 10: Its hard to read and distinguish the mean velocities from InSAR (light blue colors). Please use various shades of blue, just like what you used for the piling depth.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 5 Report

This is a well-written and highly-focused study using a InSAR-derived method that exploits small-scale features that allows recognition of patterns of differential subsidence in the Mekong River delta. Local measurements of relative subsidence in the selected or 'anomalous' sites (buildings) is a good complement to explore the basal causes for these subsidence patterns. 

I have no detailed comments along the text because my impression is that the text is clear and easy to follow. The only suggestion would be to reorganize the discussion chapter including some figures now in the appendix. In fact, despite the conclusions could be mostly as expected or no conclusive in terms of piling structures or land uses, it could be better presented with some figures showing the correlations in a more visual way.

 

Author Response

Please see the attachment

Author Response File: Author Response.docx