A Research on Cohesion Hyperspectral Detection Model of Fine-Grained Sediments in Beichuan Debris Flow, Sichuan Province, China

Round 1

Reviewer 1 Report (Previous Reviewer 1)

 

General Comments

 The authors have improved the manuscript by their revision. However, a bit more work is needed for the reader to easily and fully understand the research. For example, I found more information on the Internet about CHEOS than is in this paper. A reader should not have to search for this information. A simple summary of CHEOS and Gaofen 2 would be helpful. See specific comments for suggestions.

 

Specific comments

 Line 26: Sensitive bands would seem to indicate red and near infrared portions of the electromagnetic spectrum. Why not specifically mention those terms in the abstract and elsewhere in the paper?

 

Line 54: Change “analyze” to “analyzed”.

 

Line 68: Change “;” to “.” and end sentence.

 

Line 70: Change “;” to “.” and end sentence.

 

Lines 75 and 83: Change “inter-friction” to “internal friction”.

 

Lines 76-82: The authors mention numerous studies by various researchers. A semi-colon (;) as part of this list is not an appropriate grammar.  The authors could change all semi-colons to periods or use a colon (:) and then number the studies. A suggestion follows:

 

Various researchers have studies debris flows and factors influencing initiation and movement: (1) Cunmei Zhou et al. studied influencing factors of shear strength 76
parameters of compacted loess in 2018; (2) Mingkang Yan studied the effect of water content on apparent sand cohesion in 2018; (3)

 

The suggested paragraph shown above is the appropriate use of a semi-colon in a list of authors.

 

Line 77: Change “;” to “.” and end sentence or use the grammatical change suggested above.

 

Line 86: Please explain why “Hyperion” is mentioned. Also, there is not a complete sentence in lines 86-89.

 

Lines 113-118: Online news mentions the high-resolution optical as well as radar in the China High-resolution Earth Observation System (CHEOS) series of satellites. Did you use any radar as well as optical imagery in this research?

 

Line 122: Change “none” to “no”.

 

Line 123: When you mention “brunches” do you mean “bushes”?

 

Lines 137-155: Nice addition to paper. It helps the reader understand the research.

 

Lines 200-207: Also, good addition to paper.

 

Line 272: Change “firstly” to “first”.  The word “first” is the appropriate grammar.

 

Lines 285-293: Sensitive bands would seem to indicate red (~ 700 nm) and near infrared (beyond 1000 nm) portions of the electromagnetic spectrum. Why not specifically mention those terms here? That way the reader gets a clear understanding of the sensitive bands and relationship to minerology as well as particle size.

 

Line 293: Change “we select” to “we utilize”.

 

Lines 312-314: These lines are not clear. Please reword.

 

Line 365: Please clarify what “chain action” refers to. Do you mean “chain reaction”?

 

 Line 394: Add “apart” after “farther” and change “decreasing” to “decreases”.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report (Previous Reviewer 2)

Dear authors,

This manuscript is an updated and resubmitted version of previously rejected paper. The manuscript has been significantly updated, with majority of the remarks from the initial review incorporated into the updated manuscript.

 

However, in my opinion, the manuscript is still lacking significant improvements in some aspects – methodology is too general and itemized, characteristics of the study area are not properly outlined, and the basis of their work is not properly addressed. I believe that this manuscript needs significant improvement which requires extensive time and effort to properly address and present the methodology and reasoning behind the conducted research.

Therefore, the revised manuscript isn’t improved enough over the initial version to be suitable for publishing.

 

Please find below the main remarks that remained after introducing the changes from the initial review:

 

General remarks:

In the introduction the references are still grouped together - 8 references [4-11] that were earlier used for simple definition of cohesion now are used to reference early warning system. Please address relevant references individually and remove dummy references from the text.

 

Description of the study area is still too short, with only general information given, without information about site-specific characteristics. Please provide background data about your case study that introduce site-specific characteristics to the reader. Reflect on the existing geological profiling, or lack of it, etc…

 

Presented methodology only depicts the process flow chart and short outline of each component. Methodology doesn’t cover the methods used or detail approach necessary to undertake the research. Please expand the methodology section to accurately describe the steps undertaken for the research which allow for the replicability of the research.

 

Authors use several references in the methodology, which are neither clearly explained and it is not clear why are they relevant. E.g. authors use their earlier work [26] to describe overtaken methodology or data?

 

Authors introduced Figure 3 as “Particle size histogram of fine-grained sediments”, while the standard for representing grain size distribution is granulometric curve. Why did you use this unconventional and inappropriate graph?

 

Scatter plot of the factors influencing cohesion show very low coefficients of regression – all below 0.45 and two below 0.15 (density and moisture). Authors need to show that these correlations are statistically significant (which is almost impossible if R2 is <0.3), in order to justify their use. Otherwise, they can be omitted without influencing the results. Alternatively, they can compare calculated and measured cohesion with and without one or both of these variables to show their relevance.

 

Discussion and conclusion have remained the same as before.

Discussion is pure readout of the results, without contribution of the authors to provide insight into their findings. Please rewrite the discussion to provide in-depth evaluation of your findings relevant for the discipline.

 

Conclusion should reflect on the specific contribution of the presented research to the state of the art.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report (Previous Reviewer 3)

August 10, 2022

Dear Authors

·        Line 96 change Strata with Lithology

·        Fig. 3: Can you provide a classic Particle Size Distribution Curve?

·        Be sure that all the obtained results are described in result section and not in the method section.

·        In the reference list the great majority of the references comes from China. This is related certainly to location of the case study but could be better balanced by citing works from other countries. Please make an effort in reducing the situations above. 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report (Previous Reviewer 2)

Dear authors,

I believe that you have made sufficient changes to the manuscript and it can be published as is.

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

 

General comments

The authors have selected a topic worthy of study. Debris flows are highly destructive world-wide and difficult to predict or monitor. However, the authors have not discussed clearly the remote sensing used in their study. A summary of the satellite, imagery, and bands would help the interested reader. It might be useful to include a figure showing a band or combination of useful infrared bands. Additionally, the discussion of important factors in debris flows needs improvement.  Both of these limitations make it difficult for the reader to assess the model’s usefulness. Some specific suggestions are included in the specific comments.

 

Specific comments:

 

Line 17: Delete “a”.

 

Lines 17-19: It is unclear how a cohesion model affects debris flow stability. Please clarify.

 

Line 27: Why is the near-infrared 750 nm band most useful for a hyperspectral cohesion model?

 

Lines 28-29: Why are bands longer than 750 nm also effective? Is there something in the infrared characteristics of the earth materials that makes those bands useful? Please explain how water, moisture, particle size or other factors are at work.

 

Lines 38-39: Suggest changing sentence as follows: “A debris flow is a deposit of rock, mud, debris and water often triggered by intense rain or rapidly melting snow”.

 

Line 42: Change”446000” to “446,000” and “with 26 died” to “with 26 dead”.

 

Lines 44-45: Change “42 people dead” to “42 deaths” and “4000” to “4,000”.

 

Lines 50-52: Change sentence as follows: Particle sizes less than 2 mm easily are transported by water, especially during high-intensity rainfall or rapid snow melt”. “Thus, these earth materials play an important role in a debris flow early warning model.”

 

Lines 62-66: This paragraph is crucial in convincing the reader about the usefulness of a hyperspectral debris flow model for rapid detection. Fine-grained deposits can be mapped. Slope can be calculated. Rainfall intensity can be measured and monitored. It is unclear how hyperspectral bands alone can be useful.

 

Line 73: Change “10000” to “10,000”.

 

Line 75: Figure 1 needs coordinates on the x and y axes.

 

Line 84: Please explain “Gaofen” for the reader with an interest in debris flows and a model.

 

Line 88: Change “Firstly” to “First” and clearly summarize what was done. The reference is good but not a sufficient explanation for the interested reader.

 

Line 92: Change “Secondly” to “Second” and change “In which,” to “The”.

 

Line 120: Figure 3 need x and y geographic coordinates.

 

Line 132: Unsure what is meant by “biting” force. Do you mean binding or bonding? Please clarify.

 

Line 146: Change “silty soil” to “silt size particles”.

 

Line 150: Change “silty soil” to “silt size particles”.

 

Line 170: It is unclear why short-wave (Near infrared-NIR) is eliminated. In the abstract, the NIR (750 nm) was considered the best band. Please clarify.

 

Line 208: It is interesting that % moisture and density are correlated in figure 4 and this should be discussed in text.

 

Lines 248-249: This is not a complete sentence. Please adjust text.

 

Lines 261-269: It is not clear how these discussions of permeability and density relate to the figure 4 correlated curves of density and % moisture.

 

Lines 315-317: I agree that a debris flow early warning/monitoring system would be useful but the manuscript does not make a convincing case with the data presented.

Reviewer 2 Report

Dear authors,

The research presented in this paper, cohesion hyperspectral detection model and factors influencing its performance, is very interesting and has the potential to be published.

 

However, manuscript is lacking in all crucial aspects – state of the art review is non-existent, methodology is too general and itemized, characteristics of the study area are not properly outlined, results and discussion contain data readout without proper discussion and elaboration. Therefore, the paper in current form is not suitable for publishing. Please rewrite the paper to properly reflect the effort behind the research and resubmit the paper.

 

Please find below few crucial remarks that need to be addressed in order to improve the manuscript.

 

 

General remarks:

 

State of the art is too generic – the references are grouped together and their relevance to the presented topic remains unclear. E.g., authors have used 8 References for simple definition of cohesion. Please address relevant references individually and explain briefly main finding for each of them to provide context for your research.

 

Aim of the paper is not clearly defined – instead, the aim is dropped and in its place authors have placed introduction to remote sensing.

 

English phrasing is confusing at times, e.g. „For remote sensing has the characteristics of macro“, “In this paper, fine-grained deposits refer to Quaternary sediments with particle size less than 2mm, which are mainly composed of sandy soils, with loose structure and large porosity.” These sentences, among many others make no sense. Please have someone proofread the manuscript.

 

Many parts of the text are written in the imperative mood, which presents paper as the list of commands, e.g. “Firstly, make the samples…”, “Finally, carry out soil hyperspectral..”, etc.

 

Description of the study area is too short, with only information given is brief geographical position.

 

Presented methodology only depicts the process flow chart and short outline of each component. Methodology doesn’t cover the methods used or detail approach necessary to undertake the research. Please Expand the methodology section to accurately describes the steps undertaken for the research.

 

Authors use several references in the methodology, which are neither clearly explained ans it is not clear why are they relevant.

 

Figure 3 is almost carbon copy of the Figure 2, making it redundant. Additionally, both figures are lacking descriptive characteristics of the study area.

 

Determination of the factors influencing cohesion is not clear – scatter plot of results would here be beneficial to supplement data from the table. Also, authors should provide references backing up these relation to confirm that they are not working on site-specific data.

 

 

From the figure it can be seen that for several parameters shorter wavelengths have similar correlation coefficient across the larger band. Without in-depth description of their methodology, readers might conclude that the selection of the sensitive bands was done arbitrarily. Please provide numerical background for your data to provide insight into the selection.

 

Authors describe that the high correlation zones of the band wavelength and relevant parameters were determined by multi-variate regression analysis. What is the difference between the Selection of sensitive bands and Multi-variate regression analysis of cohesion and sensitive bands, and why was the discussion continued only for the second method?

 

Selection of sensitive bands shows that effective internal friction angle has similar corelation coefficient regardless of the band, which raises question if it contributes to the model at all. On the other hand, corelation coefficient drops significantly from higher zones to zone 1 for density and moisture. Therefore, other question is if zone 1 is relevant. These parameters should be separately analysed in the results to additionally try to reduce scatter between predicted and measured cohesion.

 

Discussion is pure readout of the results, without contribution of the authors to provide insight into their findings. Please rewrite the discussion to provide in-depth evaluation of your findings relevant for the discipline.

 

Apart from the above, discussion contains some introductory data, e.g. definition od cohesion.

 

Conclusion should reflect on the specific contribution of the presented research to the state of the art.

 

 

Specific remarks

 

Kewyords:

„stability“ is too general. Please use „stability of fine-grained deposits“ or something as specific

Along the same lines, instead of „hyperspectral remote sensing“ use „hyperspectral detection model“

 

geological coordinates?

 

 

Number of data used is constant (121) and it is redundant to burden all of the tables with the same number over and over.

Reviewer 3 Report

July 10, 2022

Dear Authors:

This manuscript entitled “A research on cohesion hyperspectral detection model of fine-grained sediments in Beichuan debris flow, Sichuan province, China” presents a new method developed to estimate the cohesion hyperspectral detection model in fine-grained soils in order to have a useful tool to incorporate in an early warning system. The soil considered was involved in debris flow landslides occurred after the earthquake taken place in 2008. This soil was investigated in terms of shear strenght parameters, density, pearmeability, particle size and moisture. Although the topic is really interesting and fits the scope of the journal, in its actual form, the manuscript is not ready for publication and needs to be improved before to be reconsidered for publication in Land - MDPI.

In addition to the comments included in the pfd file, other comments are reported below:

1)    The structure of the manuscript needs to be improved. The manuscript is set up in many points, that, however, all of these are poorly described. There are a lot of information that can be insert, for example about the study area, but also about the measurement experiments. The introduction can be improved with information about previously studies on the role of cohesion and its estimation with remote sensing methods. I recommend a more continuous text instead of a series of points.

2)    I cannot understand if the soil involved is a sandy or a silty soil, or do you have considered only the fine-grained part for the cohesion estimation? Really, it is not clear. Generally, debris flows involve gravel and sandy sediments with a low amount of clay or more generally of fine-grained sediments. But you have fine-grained soils, and the results of measured parameters confirm values typical of fine-grained soils. It sounds strange that debris flows are in this material, please verify the classification of these landslides (maybe they can be earthflow), or provide more specific information about: i) the material involved in the landslide; ii) the geological context in which the triggering occur, iii) form and character of landslides (and pictures, if avaiable).

3)    The description of the study area is very short and, in my opinion should provide much data about past landslide events, also in terms of triggering, and about the earthquakes that involve the region.

4)    Finally, I am curious to ask you how the early warning system could use the variation in cohesion derived by remote sensing data, in a earthquake triggered landslide. I think for rainfall triggered landslide it can be reasonable, but in this case the variation due to wave could take a rapid change in the fabric of soil and consequently, a rapid variation in soil water pressure that can strongly reduce soil cohesion.

 

I think there are many heavy structural problems that affect the manuscript. Once addressed these problems, the manuscript can be reproposed for publication in Land MDPI.

Comments for author File: Comments.pdf