Impact of Levee-Breach Width on the Channel–Levee–Floodplain: A Case Study in the Huaihe River Basin, China

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The 300-year flood does not convince me; what is the relevance of doing this? In 300 years, many geomorphic changes can happen based on hydroclimatic conditions. I would say that authors may first use the discharge data for the last 50-60 years to get how severe flood (in terms of return period) magnitudes are experienced in that area. Then, based on that, scenarios can be generated. For example, if the area in the last 50 years has not visited a flood event of 100 years, there is no need to do a 300-year flood; instead, 25-year, 50-year, and 75-year scenarios can be generated, which could be much more relevant to administrators. 

Again, in generated scenarios, I would like to know what the breach width is in current days. Is there any record of changing breach width in the last few decades? Any survey document? Authors may also use historical Landsat imageries to estimate the possible changes in breach width. Based on the historical changes, authors may stipulate some scenarios of breach width. This is because the scenario of a breach width of 10m to 2000m is unrealistic at the exact location. Such changes would require several decades to centuries, which would not benefit the actual purpose. 

I observed that the water stages for 100-year floods over the Yinghe and Huaihe rivers differ. What could be the reason behind this?

In Fig 1, 1st level and other levees are not distinguishable from the map. Use some different colours and increase line thickness. Use the better resolution to export the image. Huaihe River is not demarcated. 

Line 92, "ten-thousandth slope", instead mentions slope value in degrees, which would be more convenient for readers. 

Lines 129-131 mention the resolution of the original DEM and interpolated terrain. 

Line 131-132, "micro landforms... are represented by linear structures". 

The authors missed the notations in Fig. 8. 

Line 254-257, Justify the reasoning. 

Line 260-264, how is it calculated? Mention the equations or methods with references. 

Author Response

Dear Reviewer,

We have studied your comments carefully and tried our best to revise our manuscript according to the comments.

Please see the attachment.

Thanks for your valuable comments and suggestions, again.

Best wishes to you.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Dear authors,

Please, read the attached file.

Comments for author File: Comments.pdf

Author Response

Dear Reviewer,

We have studied your comments carefully and tried our best to revise our manuscript according to the comments.

Please see the attachment.

Thanks for your valuable comments and suggestions, again.

Best wishes to you.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

I am suggesting a minor revision as the manuscript is not clearly written and there are some issues with the methodology. The authors should be able to address all my comments without re-doing the hydrodynamic simulations.

The biggest issue is the lack of validation. Without validation, the entire study looks like a computer game. For example, how can we confirm the through-levee discharge is the same as reported? How can we confirm the flooded areas based on simulations? There are two solutions. If validation is absent, then why not run simulations in a sandbox with variable breach widths? In this case, the authors can do more analysis with the river size and breach size. Another solution is to find satellite images during flooding and validate the flood maps. Many free satellite data are available such as Landsat, Sentinel-1, and Sentinel-2. Alternatively, use ground gauges to validate the results.

I have not seen the objective of this study, and am unsure which area has a higher priority to be protected from flooding. Levees were built to prevent flooding, but the analysis showed that a wider breach allows more water to enter the floodplain and suggests that flood peaks in rivers dropped. So, which is more important, rivers, floodplain, or downstream cities? This must be clarified.

A related issue is about the study site. I checked satellite images and saw the floodplain has been largely converted to agricultural fields and villages. The so-called levees are not visible in images, and the marked levee breaches seem to be roads. Since the authors have reported levee height, I guess field photos should be available, and should be added to the manuscript.

Line specific comments

L2: which is more appropriate, channel-breach-floodplain or levee-breach floodplain?

L13: why is it hard? Levee breaches are easily seen on DEMs or imagery in a natural, unforested floodplain.

L14: what system?

L23-27: make this concise

L34: solid rings? What are these?

The introduction was poorly written. Levees can be natural or artificial. The introduction should provide a more comprehensive summary of the current knowledge on this topic. There are many papers on natural levee breaches and crevasse splays, and their roles in floodplain water circulation, because levee breaches play similar roles regardless of the origin. It must be clearly stated that this paper focuses on artificial levees.

L59-60: if the water elevation difference is available, report it here and add sources.

L68-79: this is a summary of methods and does not fit here. What is the objective of this study? When talking about methods, add some references of similar studies on 2D models on levee breaches.

Fig 1: This figure is difficult to read. I can’t tell exactly where the levees are. Please fix it and add field photos showing levees and levee breaches.

Fig 2: I don’t know what this is, and this was not explained. What do the black arrows mean?

L117-124: I guess the 1D model only works in the channels and the 2D model works on the floodplain. But don’t let readers guess. The roughness values were only mentioned in rivers, but the floodplain has different land covers and should also be mentioned.

L129: what is original DEM?

L125-137: add necessary references to the model setup.

L131: If levees were eliminated, how could you study levee breaches?

L144: what is project flood stage, the flood stage in this research project?

L155: Isn’t the levee breach a fixed feature on the ground? What is the ground truth width? As a reader I am lost.

Fig 3: This hydrograph looks like real floods, but the manuscript did not mention it. I am assuming the simulations were not steady state since the discharge changes over time. Further, are these flood periods (1 month) typical in the study site? These should be mentioned in the text.

Fig 4: I have no idea what this figure is trying to say. Why is the inflow discharge a single curve instead of two? What are the color schemes of c-2 and c-3? I have a suggestion on how to make good figures. Ask a colleague who is unfamiliar with this study and see if the colleague can understand the figure solely based on the figure caption without referring to the text.

Fig 5b: is the total volume an artifact of flood duration? For example, if the flood lasted longer, should volume increase?

Section 4.1: This part tried to generalize levee breaches in floodplains. This is a common mistake in hydrology or floodplain studies. The study has only looked at one floodplain but tried to make the findings universal. For example, on L361-363, the authors claimed that the river stage in the floodplain rises faster than in the main channel. In this specific site it is true, but in floodplains where the receiving basin is large enough and dry enough, or the basin is connected to large downstream basins, this may not be true. This is related to a point the authors mentioned on L59. In a wide floodplain the lateral free surface gradient is likely to be larger, while a narrow floodplain tends to have zero lateral gradient (Adams et al. 2004). This is the reason I suggested doing more homework on natural levees and position this study better in the existing levee studies.

L352: it helps to plot a few curves using values of the study site. Add a reference to this equation.

I assume the driving force of flooding in this floodplain is river flood, not local precipitation. This should be mentioned somewhere but not necessarily in the discussion.

 

Adams, P.N., Slingerland, R.L. and Smith, N.D., 2004. Variations in natural levee morphology in anastomosed channel flood plain complexes. Geomorphology, 61(1-2), pp.127-142.

Author Response

Dear Reviewer,

We have studied your comments carefully and tried our best to revise our manuscript according to the comments.

Please see the attachment.

Thanks for your valuable comments and suggestions, again. Best wishes to you.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Section 4.1 mostly talks about results, should be shifted in result section with equation, not in discussion.

Section 4.3: for lines 429-431, references required.

The 3rd point in conclusion can be included in discussion, but not in conclusion. Instead, tell significance of the observation regarding relationship of the inundation area and the breach width and why it was different for different rivers.

Author Response

Dear Reviewer,

We have studied your comments carefully and tried our best to revise our manuscript according to your comments and suggestions.

The attachment contains our detailed reply to your your comments and suggestions. Please see them in the  attachment.

Your valuable comments and suggestions are helpful to improve the quality of the manuscript and for our authors. Thanks for your comments and suggestions.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Dear authors,

I appreciate your effort to improve the original version of the paper. However, I have two minor observations:

- Figure 4 c2 and c3: to replace [3.0, ∞) and [1.0, ∞) with >3.0 and > 1.0 respectively

- You agreed to change the volume units in Figure 5 b from hundreds of millions of m3 to billions of m3, but this change was not made. If you find it difficult, you can leave the figure as it is.

Greetings

  

Author Response

Dear review,

Thank you for your comments.

We use >3.0 and > 1.0 to replace [3.0, ∞) and [1.0, ∞) in Figure 4 c2 and c3.

The volume units in Figure 5 b  had been changed from hundreds of millions of m3 to billions of m3. We have down this work but forgot to replace it in the for previous revised draft.

Greetings.

Reviewer 3 Report

Comments and Suggestions for Authors

This is my second time reviewing this work. The general problem with this manuscript is lack of clarity. It was written in a way that only the research team can understand, and readers like me are quite confused and need to guess. This applies to the objective and some details. The authors have answered my comments but have made very few changes to improve the overall readability of the manuscript. I seriously suggest improving the writing so that readers can understand what has been done. It has been very difficult for me to read through the manuscript.

Now I doubt the two mentioned breaches even exist on the ground. The images are way too zoomed out and no breach can be confirmed. The study explored different widths of the two breaches but did not say the actual width on the ground. The manuscript mentioned levee crest elevation and ground elevation but did not mention the levee breach elevation. Are these two breaches imaginary? Are the levees intact at these two locations? Why pick these two locations? Will the administration create breaches before large floods?

I gave a suggestion on how to make good figures in the previous round. I said that show the figure to a colleague outside the research team and see if the colleague can understand the figure without referring to the text. I don’t see the authors take my suggestion seriously and some figures are still problematic in the revised manuscript. These include Figs 1, 2, and 4. For example, Fig 1 is a mess and I don’t know where the levees are – they seem to be everywhere. Fig 2, what is “floodway distinct”? It was never explained and seems unrelated to this research.

There are some minor examples of unclear writing. I am pointing out the problem of “solid rings” on line 35 again. Is solid ring a well-accepted term as a synonym of levee? I have not found any published papers using this term. Did the authors create this term? Do people think of levees when hearing the term solid rings? If authors believe this is a common term, provide references.

The title said two things: levee breach and channel breach. Are these different? Can a channel breach occur without breaching levees? But this work is about levee breaches.

L13: why is it important to predict breach geometry prior to a breach? Is this the focus of this study? This distractive sentence should be deleted.

Comments on the Quality of English Language

See comments above

Author Response

Dear Reviewer,

We have studied your comments carefully and tried our best to revise our manuscript according to your comments and suggestions.

The attachment contains our detailed reply to your valueable attachment. Please see them in the attachment.

Your valuable comments and suggestions are helpful to improve the quality of the manuscript and for our authors.

Thanks for your comments and suggestions.

 

Author Response File: Author Response.pdf