Research on the Prediction Model of Loess Collapsibility in Xinyuan County, Ili River Valley Area
Round 1
Reviewer 1 Report (New Reviewer)
Comments and Suggestions for AuthorsI read the paper very carefully. The authors used a multiple linear regression model and the RBF-based neural network model for loess collapsibility prediction. I congratulate the authors for making such an attempt. However, the presentation of the paper is highly good, and significant novelty is found.
1. The author must constructively change the abstract by adding more numerical value in terms of error to the result [Especially PBIAS].
2. Please add more recent literature (2021, 2022, and 2023) in terms of loess collapsibility prediction modelling for better understanding.
3. Please modify the objective section for a clear understanding i.e first input selection, then modelling by hybrid NN that is RBF.
4. A description of the RBF model must be required with architecture and pseudo-code.
5. There are so many optimization techniques in the recent world; why does the author use any optimization technique for optimization? Is there any specific reason for this?
6. Author citation: My suggestion please add some Water, Nature, and Springer journal
7. Author used Xinyuan County's Yili Valley area for research purposes, are there any scientific reasons for this? The figures do not match the MDPI journal standard. Improve the figure quality and writing.
8. Study area figure must be changed; Request to draw study area figure using GIS
9. The author must add statistical components/parameters of collected data in the study area section in a tabular format.
10. Equations for WI, PBIAS, and NSE must be there; please add a citation for reference purposes also
11. A comparison statement (compare with other research articles) must be added in the result and discussion section to better visualize the proposed research. The author must add future scope in the last portion of the manuscript.
12. The advantages and limitations of the RBF model must be added.
13. For better analysis of the result author must add a Box plot, Histogram, and Taylor Diagram
14. Fig 6-8: No critical discussion is provided, critical discussion is required.
15. The author must add a study flow chart with the methodology for a better understanding
Author Response
Dear profess,
Thank you very much for your comments and suggestions.
Those comments are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our researches. We have studied comments carefully and have made correction which we hope meet with approval. Revised portion are marked in red in the paper. The main corrections in the paper and the responds to the reviewer’s comments are as flowing:
Point 1:The author must constructively change the abstract by adding more numerical value in terms of error to the result [Especially PBIAS].
Response 1: Thank you very much for your suggestion, we have added the discussion of the model evaluation index in the summary section.Including root mean squared error (RMSE), correlation coefficient (CC), Nash–Sutcliffe efficiency coefficient (NSE), and percent bias (PBIAS)
Point 2: Please add more recent literature (2021, 2022, and 2023) in terms of loess collapsibility prediction modelling for better understanding.
Response 2: Papers on loess collapsibility prediction models published in recent years have been added.
Point 3:Please modify the objective section for a clear understanding i.e first input selection, then modelling by hybrid NN that is RBF.
Response 3: The target part has been modified, and the modeling process of RBF neural network model is briefly introduced.
Point 4:A description of the RBF model must be required with architecture and pseudo-code.
Response 4: The RBF neural network prediction model in this study is established by using the RBF neural network model modeling tool built in the ‘IBM SPSS Statistics’ software toolbox.
Point 5:There are so many optimization techniques in the recent world; why does the author use any optimization technique for optimization? Is there any specific reason for this?
Response 5: In this paper, the stepwise regression method is used to optimize the regression model. Through the modeling results of the initial regression model, it can be found that the reason for the low validity of the model is that the significance of some variables is too low, and the stepwise regression method can just solve the problem that the significance of some variables in the model is too low. Stepwise regression is a method of selecting independent variables in linear regression model. The basic idea is to introduce variables one by one, and the condition of introduction is that the partial regression square and experience are significant. At the same time, after introducing a new variable, the old variables that have been selected into the regression model are tested one by one, and the variables that are considered insignificant are deleted to ensure that each variable in the independent variable subset is significant. This process goes through several steps until no new variables can be introduced. At this time, all the variables in the regression model are significant to the dependent variables.
Point 6:Author citation: My suggestion please add some Water, Nature, and Springer journal.
Response 6: A part of the journal from Water, Nature, and Springer has been cited in the article.
Point 7:Author used Xinyuan County's Yili Valley area for research purposes, are there any scientific reasons for this? The figures do not match the MDPI journal standard. Improve the figure quality and writing.
Response 7: Xinyuan County of Yili River Valley is located in the inland area of China. The loess in this area shows different properties from the loess in other areas due to its special formation. At the same time, due to the rapid development of tourism in Xinyuan County this year, human engineering activities have gradually intensified, and engineering problems caused by collapsible loess in engineering construction have gradually increased. However, there are relatively few studies on loess collapsibility in this area. Therefore, this paper selects Xinyuan County in the Ili River Valley as the research target to provide a scientific basis for the evaluation and prediction of loess collapsibility in this area. In the future, the collapsible loess in the Ili River Valley will be further studied.
Point 8:Study area figure must be changed; Request to draw study area figure using GIS.
Response 8: We used ‘ArcGIS’ software to redraw the location of the study area, and superimposed the satellite image map of the study area on the map. Finally, we improved the resolution of the map and solved the problem of unclear legends and texts in the map.
Point 9:The author must add statistical components/parameters of collected data in the study area section in a tabular format.
Response 9: We appreciate your suggestion; but, we feel that including the study's data as a table in the article might make it too long. Therefore, the study's data will be submitted as an attachment instead. Hope to get your understanding.
Point 10:Equations for WI, PBIAS, and NSE must be there; please add a citation for reference purposes also.
Response 10: We much appreciate your recommendations, which enable us to make this content even better. In the article's section 4.3, we include the model's evaluation index analysis.
Point 11: A comparison statement (compare with other research articles) must be added in the result and discussion section to better visualize the proposed research. The author must add future scope in the last portion of the manuscript.
Response 11: We outline the distinctions between this study and the studies of other researchers in the article's introduction:Due to its special geological conditions and topographic and geomorphic characteristics, the loess in the Ili River Valley area are distinct from that in other regions of China.However, at present, the evaluation of loess collapsibility in China has mainly focused on the loess in northeast, central and east China, and few scholars have evaluated the loess collapsibility in Xinjiang. Therefore, this paper collected various physical, hydraulic, and mechanical parameters of loess in Xinyuan County, and analysed the correlation of various soil property indictors of collapsible loess by means of mathematical statistics based on observed engineering cases.In addition, a prediction model for loess collapsibility in this area was established using the multiple linear regression theory and the neural network method. Finally, the rationality, effectiveness, and accuracy of the established prediction model were verified through observed engineering in the area.
We add the scope of future research in the last part of the article.
Point 12: The advantages and limitations of the RBF model must be added.
Response 12: The advantages and limitations of the RBF neural network model have been supplemented in the 5.2 part of the article.The advantages of the RBF neural network model are also mentioned in the 4.2 section of the article.
Point 13:For better analysis of the result author must add a Box plot, Histogram, and Taylor Diagram.
Response 13: In order to better analyze the results, we added a box plot to the article as a comparison of the models. However, for the sake of the length of the article, we did not add histograms and Taylor diagrams to the article.Hope to get your understanding.
Point 14:Fig 6-8: No critical discussion is provided, critical discussion is required.
Response 14: In this part of Fig.6-Fig.8, we mainly show the prediction results of the prediction model. In the following part, we supplement the comparison of the model.
Point 15:The author must add a study flow chart with the methodology for a better understanding.
Response 15: Thank you very much for your opinion. We supplemented the research flow chart in the data source section of the article.
Best wishes,
LIFENG CHEN
Author Response File: 
Author Response.pdf
Reviewer 2 Report (Previous Reviewer 2)
Comments and Suggestions for AuthorsThe paper approaches an interesting research regarding prediction model of loess collapsibility in Xinyuan County, Ili River Valley Area. However, the paper requires major revision to be understood internationally, to increase its accuracy and reproducibility.
I believe that the following aspects must be improved:
- it is necessary for the authors to specify in the hypothesis of the research, what is the novelty brought by their research in comparison with other studies that have been conducted on loess collapsibility in Xinyuan County, China
- present references and explanations related to (1) Particle size composition analysis, what methodology did you use to present the statements: particle gradation in the study area is good?
- the research methods must be clearly presented and the corresponding references.
- the discussions have no references, you only make statements, it is necessary to approach the results in relation to other similar studies.
- the references must be improved.
Comments on the Quality of English LanguageModerate editing of English language required.
Author Response
Response to Reviewer 2 Comments
Dear profess,
Thank you very much for your comments and suggestions.
Those comments are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our researches. We have studied comments carefully and have made correction which we hope meet with approval. Revised portion are marked in red in the paper. The main corrections in the paper and the responds to the reviewer’s comments are as flowing:
Point 1: it is necessary for the authors to specify in the hypothesis of the research, what is the novelty brought by their research in comparison with other studies that have been conducted on loess collapsibility in Xinyuan County, China.
Response 1: We outline the distinctions between this study and the studies of other researchers in the article's introduction:Due to its special geological conditions and topographic and geomorphic characteristics, the loess in the Ili River Valley area are distinct from that in other regions of China.However, at present, the evaluation of loess collapsibility in China has mainly focused on the loess in northeast, central and east China, and few scholars have evaluated the loess collapsibility in Xinjiang. Therefore, this paper collected various physical, hydraulic, and mechanical parameters of loess in Xinyuan County, and analysed the correlation of various soil property indictors of collapsible loess by means of mathematical statistics based on observed engineering cases.In addition, a prediction model for loess collapsibility in this area was established using the multiple linear regression theory and the neural network method. Finally, the rationality, effectiveness, and accuracy of the established prediction model were verified through observed engineering in the area.
Point 2: present references and explanations related to (1) Particle size composition analysis, what methodology did you use to present the statements: particle gradation in the study area is good?
Response 2: By collecting collapsible loess soil samples in the study area, we obtained the non-uniform coefficient(Cu) and curvature coefficient of the loess in the study area through the particle analysis test.According to “Soil Properties and Soil Mechanics (Fifth Edition) ”, when the cumulative curve of the soil concaves upward, the slope is gentle, and the non-uniformity coefficient (Cu) is greater than 5 and the curvature coefficient Cs = 1 - 3, it indicates that the particle size grading of the soil is good. In this study, the Cu and Cs of the loess in the study area are 6.12 and 1.47, respectively, indicating a good particle size grading.
Point 3:the research methods must be clearly presented and the corresponding references.
Response 3: In this paper, the research methods used are introduced, and the references are supplemented.
Point 4: the discussions have no references, you only make statements, it is necessary to approach the results in relation to other similar studies.
Response 4: The differences between this paper and other similar studies are described in the Introduction.
Point 5: the references must be improved.
Response 5: The references have been improved.
Best wishes,
LIFENG CHEN
Author Response File: Author Response.pdf
Reviewer 3 Report (New Reviewer)
Comments and Suggestions for AuthorsThe presented work deals with the problem of predicting the destructibility of loess. From the point of view of geotechnics, loess is a problematic soil, which significantly complicates its use as a base for structures. As a result, the cost of construction works on such soils increases to a great extent due to the need to take measures to prevent deformations due to subsidence of the foundations of structures. In connection with the above, the work can be considered relevant.
In this paper a statistical analysis of the relationship between the physical parameters of soils and compressibility coefficients is carried out. As a result of the presented analysis, the authors proposed a regression equation, which, in their opinion, allows predicting the behaviour of loess soils.
There are a number of comments to the paper:
1. Figure 1 and 2 should be improved. The lettering and legend are poorly legible.
2. It is not clear why the material composition analysis ((3) Material composition analysis) was carried out, as these parameters are not taken into account and not used in the future when creating the predictive model.
3 The authors talk about the development of a predictive model of loess fractureability for the soils of the Ili River Valley. However, it is not quite clear from the text of the article whether this model can be used to predict the behaviour of loess in other conditions. It may be worthwhile to analyse in more detail the relationship between the considered physical parameters of soils and compressibility coefficient not only from the point of view of statistics, but also from the point of view of soil mechanics.
4. Like any predictive model, the model proposed by the authors should have limitations. It is worth specifying initial and boundary conditions.
Author Response
Dear profess,
Thank you very much for your comments and suggestions.
Those comments are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our researches. We have studied comments carefully and have made correction which we hope meet with approval. Revised portion are marked in red in the paper. The main corrections in the paper and the responds to the reviewer’s comments are as flowing:
Point 1:Figure 1 and 2 should be improved. The lettering and legend are poorly legible.
Response 1: We redraw Figure 1 and Figure 2, and modify the unreasonable part of the graph.
Point 2: It is not clear why the material composition analysis ((3) Material composition analysis) was carried out, as these parameters are not taken into account and not used in the future when creating the predictive model.
Response 2: Loess belongs to special soil. There are differences between the causes of loess in western China and those in other regions. At the same time, according to the literature, the research on loess in China is mostly concentrated in Northeast China, Central China and parts of East China. The research results of loess in Xinyuan County of Yili River Valley mainly focus on the causes and disaster-causing mechanisms, and there are relatively few research results in physical characteristics. Therefore, based on the collection of data and field sampling, this paper analyzes the physical characteristics of loess in Xinyuan County.
Point 3:The authors talk about the development of a predictive model of loess fractureability for the soils of the Ili River Valley. However, it is not quite clear from the text of the article whether this model can be used to predict the behaviour of loess in other conditions. It may be worthwhile to analyse in more detail the relationship between the considered physical parameters of soils and compressibility coefficient not only from the point of view of statistics, but also from the point of view of soil mechanics.
Response 3: Thank you very much for your suggestions. The focus of this study is to predict the collapsibility of loess in this area. In future studies, we will further study and predict the behavior of loess under other conditions.
Point 4:Like any predictive model, the model proposed by the authors should have limitations. It is worth specifying initial and boundary conditions.
Response 4: We supplement the limitations of the model in the 5.2 part of the article. The main problem of the model is its over-fitting phenomenon and sensitivity to abnormal data. Therefore, when using the model, the amount of data used cannot be too small, and abnormal data needs to be eliminated.
Best wishes,
LIFENG CHEN
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report (New Reviewer)
Comments and Suggestions for AuthorsThank you for the revision.
Reviewer 2 Report (Previous Reviewer 2)
Comments and Suggestions for AuthorsThe authors answered my questions. The paper is improved.
I recommend accepting the paper for publication.
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
Comments and Suggestions for AuthorsThe specific opinions and suggestions are as follows:
(1) The format of the references is very non-standard.
(2) Due to the location of the research area in the valley terrace, the sampling point should be located in the low or high terrace of the valley. The authors selected landslide sampling points, especially disturbed soil samples, which cannot reflect the collapsible characteristics of undisturbed loess.
(3) The particle size curve is incorrect, and the corresponding evaluation indicators obtained based on the particle size curve are also unreliable.
(4) When analyzing material composition, the authors should pay more attention to the content of soluble salts, clay minerals, and calcium nodules.
(5) From the perspective of engineering applications, the self-weight and non self-weight collapse coefficients, as well as the level and degree of collapse, should be considered.
(6) The dimension of the ordinate unit in Figure 5 is incorrect.
(7) The density, dry density, water content, saturation, porosity ratio, and porosity boundary indicators provided in 3.2 Results of correlation analysis are worth discussing.
(8) In the conclusion, the author believes that the Quaternary loess in the study area has significant collapsibility and self-weight collapsibility. This conclusion clearly lacks scientific support.
Comments on the Quality of English LanguageExtensive editing of English language required
Reviewer 2 Report
Comments and Suggestions for AuthorsThe paper is interesting but it is not written in an academic way well enough. The paper has major problems that consist in the insufficient description of the research area and the orographic factors, the lack of description of the research methods and the lack of discussions on the results obtained.
Consequently, I recommend the following major improvements:
- improve the abstract and do not use abbreviations in the abstract without explaining them.
- specify in the description of the research area the influence of orographic factors, soil types, climatic conditions, land use.
- specify the methods used and their references.
- specify in footnotes all abbreviations used in tables and figures.
- the discussions are simple statements, you have no references; they must be analyzed comparatively and supported with similar references.
- increase the accuracy of the paper and write according to the Water writing instructions, including the bibliography.
Comments on the Quality of English LanguageThe paper is interesting but it is not written in an academic way well enough. The paper has major problems that consist in the insufficient description of the research area and the orographic factors, the lack of description of the research methods and the lack of discussions on the results obtained.
Consequently, I recommend the following major improvements:
- improve the abstract and do not use abbreviations in the abstract without explaining them.
- specify in the description of the research area the influence of orographic factors, soil types, climatic conditions, land use.
- specify the methods used and their references.
- specify in footnotes all abbreviations used in tables and figures.
- the discussions are simple statements, you have no references; they must be analyzed comparatively and supported with similar references.
- increase the accuracy of the paper and write according to the Water writing instructions, including the bibliography.
