A PANN-Based Grid Downscaling Technology and Its Application in Landslide and Flood Modeling
Round 1
Reviewer 1 Report
This manuscript proposes a physical adaption neural network to optimize the coarse grid representation from fine grid. In general, the goal of this manuscript is significant, because it can greatly promote the computational efficiency of numerical simulations. I recommend a minor revision with the following comments revised.
1. First personal tone should be avoided in a scientific paper, and “we” is suggested to be changed to “the authors”.
2. Besides the distance along cross-section, more quantitative comparisons are suggested to be carried out to clearer reflect the precision and efficiency of the model.
3. The section Discussion and Conclusions is suggested to be reorganized. Discussion and Conclusions are suggested to be described separately.
4. Two papers about the numerical simulations of Baige landslide are related to the manuscript and are suggested to be introduced in the main text.
1. Dynamic process analysis of the Baige landslide by the combination of DEM and long-period seismic waves. (2021) Landslides 18, 1625–1639.
2. Dynamic process of a high-level landslide blocking river event in a deep valley area based on FDEM-SPH coupling approach. (2023) Engineering Geology 319, 107108.
Author Response
Please check the response files. Thanks.
Author Response File: Author Response.pdf
Reviewer 2 Report
1. Title mentions "Earth-surfaced flow", which is quite ambiguous. Maybe authors should change the title to something like "A PANN-based Grid Downscaling Technology and its Application in landslide and flood modelling".
2. English should be improved as some sentences aren't understandable. E.g. page 2, lines 64-65 "Tackling the problem from another perspective, most didn’t inspect how to optimize coarse grid in external environment..." isn't understandable at all.
3. Some sections contain info that belongs to different sections. E.g. On page 5, lines 169-172 (section related to Methodology) state the performance of PANN, which should appear in either the Introduction (and reference previous research) or in the Results section.
4. Methodology section should provide more detailed explanation of the PANN methodology (previous research of the authors, etc.). By contrast, the part of the shallow-water equations is a much more known topic and can be completely left out as it doesn't provide any valuable info (authors can reference this in the introduction).
5. In Discussion section, authors have mentioned a value of 2.339e-36 (this was also mentioned in the Methodology section) but do not provide the supporting evidence for this claim.
6. Discussion section is merged with the Conclusions which might be somewhat more difficult for the readers. Authors should split it from Conclusions, and give it a separate section or merge it with the Results section.
English should be improved as some sentences aren't understandable. E.g. page 2, lines 64-65 "Tackling the problem from another perspective, most didn’t inspect how to optimize coarse grid in external environment..." isn't understandable at all.
Author Response
Please check the response file. Thanks.
Author Response File: Author Response.pdf
Reviewer 3 Report
The topic of this MS is interesting and it has the potential to prove its applicability. However, there are still many issues to be clarified, and a major revision is recommended.
The introduction part needs to be rewritten, as the foundation and the basic concept of the proposed method have not been clarified. Too many previous studies and results have been listed without a strong bond to the proposed new method. Furthermore, the research gaps and the novelty of this study should be clearly stated.
Try to illustrate the framework or the flow chart of the PANN in one figure, so Figs. 1 and 2 are suggested to be merged and reorganized.
The assessing indicators should be introduced and how they were selected should be explained.
The discussion part is important and the comparison of the performances of new methods with the traditional methods should be conducted. Both the accuracy and the efficiency should be carefully compared in the discussion. The sensitivity analysis of different parameters or resolutions is supposed to be conducted.
Try to quantify the performance of the proposed model.
Although the MS is vastly understandable, there are still too many grammar errors. Please try to polish the writing.
Author Response
Please check the response file. Thanks
Author Response File: Author Response.pdf
Reviewer 4 Report
Dear authors,
congratulations for your interesting attempt. I really enjoyed reading the paper. Just a very few minor edits suggested within my comments in the pdf file.
Comments for author File: Comments.pdf
A few edits on English language are suggested within my comments in the pdf file.
Author Response
Please check the file. Thanks
Author Response File: Author Response.pdf
Round 2
Reviewer 3 Report
I think this manuscript can be accepted at this stage.
acceptable.