Prediction of Landslide Deformation Region Based on the Improved S-Growth Curve Model
Round 1
Reviewer 1 Report
The authors present a very interesting study on prediction of landslide deformation areas, based on experimental results. Although some of their findings only confirm what was very well known before (e.g. that the increase of the slope angle will accelerate instability process), they present physical model experiments and they also propose a new S-growth curve model that is shown to be successful in predicting of landslide volumes. Overall, I think the presented study is interesting to read, the paper is well organized and structured, and I recommend it for publication after only a minor revision.
Comments:
1. Could figures 3, 4, 5, and 7 be provided with a higher resolution?
2. Can the authors provide examples or explain in more detail the types of the vegetation cover (what is low density – certain type of grass or some kind of trees?, general density – bushes?, high density – grass?). An example of each vegetation cover type would be very helpful for visual estimation of the landslide areas.
3. Do the authors plan to include seismic effects in their future studies of the landslide deformation areas/volumes?
Author Response
请参阅附件。
Author Response File: 
Author Response.doc
Reviewer 2 Report
Dear authors,
The research you have presented in this paper, focusing on landslide deformation under different rainfall, slope and soil conditions is very interesting and deserves to be published. Paper is well structured and concise, but is lacking in clarity and is poorly written (in regard that many phrases used are uncommon for the research field). Unfortunately, the manuscript quality doesn’t match the inevitable quality of the research, lacking in the state-of-the-art review, methodology and discussion which affects the overall quality of the paper.
Therefore, the paper requires major revision before publishing. Please find specific and technical remarks below.
General remarks
Major shortcomings of the paper are lack of state-of-the-art review that would provide context for your research, uncoherent description of the experimental setup that is hard to follow, vague description of the methodology that doesn’t explain the specifics of the methods used for analysis, and presentation of the results is cluttered and doesn’t offer the answer to research questions and as a result the aim of the paper cannot be considered as achieved.
The following aims of the paper are not achieved (or rather their achievement is not evident from the contents of the paper):
· “find the commonalities and differences in the slope instability process based on experimental phenomena”. There is no discussion present that would allow to determine commonalities and differences in the slope instability process. Authors have only presented results of different experiments without identifying similarities between them.
· “reveal the evolution mechanism of rainfall-type landslides”. This is not addressed at all.
· “an improved S-growth curve model is proposed to predict the landslide volume by taking the landslide volume values at two adjacent times and the factors affecting the landslide volume as the input values”. This is presented, but authors do not explain how is this useful since continuous monitoring of landslide is required and result is achieved during post-processing. It seems redundant since after the landslide the final volume can be determined?
Specific remarks
The introduction is not adequately referenced – General statements (such as “rainfall is one of the main factors inducing landslides”) are heavily referenced, while the contribution of landslide disasters overtaken from National Geological Hazard Bulletin is not referenced at all, as well as rainfall distribution (“rainfall is mainly concentrated in May to September”). Remove unnecessary references and properly reference the sources from which the data was overtaken.
The state-of-the-art review doesn’t offer sufficient overview of the research in the field, failing to provide context for the presented research. Out of the 15 papers ([10]-[25]) referenced in the state-of-the-art review only 9 are newer than 5 years, and 4 of them are referencing one of the co-authors. Please address the main findings of the relevant and current papers other than your own.
In the aim of the paper authors state that “this paper collects the data of indoor rainfall model tests completed by the team in recent years [26-28]”, but they do not offer main characteristics of the data overtaken. I suggest that you add the paragraph describing the data overtaken from your previous work in the Methodology section.
In the aim of the paper authors state that “S-growth curve model is proposed to predict the landslide volume by taking the landslide volume values at two adjacent times”. Instead of two adjacent times you should specify the exact time step, as the current phrasing does not sufficiently describe what “adjacent” is in the landslide time scale.
There are several instances of redundancy in the manuscript, e.g.
· subsidence in the world in the world (ln37)
· rainfall is mainly concentrated in May to September, which leads to landslides also mainly occurring from May to September (ln42)
The phrasing used is not conventional and makes reading strainful, e.g.
· In the time dimension of the year (ln43)
· …shows a normal distribution of the law (ln44)
· "western group east single, south more north less, central and southwest frequent" (ln46)
· “explore the change law of slope failure degree” (ln97)
Please have someone proofread the manuscript.
The experimental setup is not explained clearly (see also remark above). Please explain what are the specific conditions for each experiment – in current state it is not distinguishable how are the experiments related or differentiated. Table 1 is not readable, and basically only serves as the tabular representation of text contained in the ln125-134. Please rephrase the text to fully describe the main factors of each experiment. Also, please adjust the Table 1:
· Remove “Fujian Province Coastal Slope” column since it doesn’t change across the experiments.
· Redistribute the table in a way that for each experiment all relevant variables are included (rainfall intensity, slope angle, slope crest, compaction of subsoil, vegetation cover density, wind speed).
Quantitative analysis methods are vaguely presented – methods used to process the recorded images are not described. They are only presented as a concept, without the methods or tools used for actual processing.
Experimental setup is lacking detail:
· What vegetation was used for cover and what was the root zone extent and canopy cover?
· How are rainfall intensities related to storms in the Fujian Province?
· How are wind speeds related to Fujian Province?
· What were the rainfall intensities for S7-S16?
Please describe the methodology in more detail.
Presentation of the results is cluttered:
· On Figure 3 same experiments are represented with different color, making the comparison less straightforward (e.g. S2 is red and black, S2, S6, S10, and S12 are all red, etc.)
· On Figure 4 and Figure 5 colors for observed and predicted values are reversed
· Haw are experiments presented on Figure 6 representative?
The results are a pure readout of the graphs, without proper explanation by the authors:
· What has caused the peaks in the area and volume change? Is it uncertainty associated with image capturing/processing or natural phenomenon?
· Why were only 6 experiments fitted with S-curve? How are they representative?
Some of the conclusions are not supported by the data presented, e.g. “(2) slope angle … were negatively correlated with slope stability”. If we recall the data from the Table 2 where slope is increased from S4 to S6, one cannot observe the corresponding increase in the landslide area.
Technical remarks:
In sentence starting on ln108, Figure 1 is referenced twice.
Is “model box” == “experimental tank”?
Figure 1 is almost unreadable: The rainfall system is shown in small photos, not showing the detail. The legend should be provided in the figure caption to free more space for the photos. There is no need to show the picture of a computer in the manuscript since it doesn’t provide any important information of the setup.
Reference to the layout of the Figure 3 is provided on ln142, where it is out of context. Please rephrase the ln142-144 and add the current text to the caption of Figure 3.
The paragraph starting on ln149 only has one sentence. Please merge it with other related content.
Is “broken surface” == “zone of depletion”.
Description of the results starting on ln206 should be separated into smaller paragraphs, each dedicated to one line of graphs on Figure3, to make the results more readable.
Author Response
Please see the attachment.
Author Response File: Author Response.doc
Round 2
Reviewer 2 Report
Dear authors,
The revised manuscript presents significant upgrade in comparison with the original submission.
Since you haven’t highlighted the introduced changes, and some references to manuscript lines are not correct it is hard to estimate the degree of introduced changes.
Therefore, please find below a list of your revisions I think haven’t been addressed in the revised version:
- “The latest research in this field has been added in lines 68-80 and 103-105, and the cited articles have been revised.” – these lines are the same as in the original version
- “On lines 217-221 of the article, we added a paragraph describing the data obtained from previous work.” – these lines are not added, they are the same as in the original version, although slightly modified
- “On lines 119-120 and 439-440 of the article, we modified the problem by defining the selection of adjacent 5-minute landslide volume values as inputs.” – this was not described in these lines
Author Response
Please see the attachment.
Author Response File: Author Response.doc
