Alluvial cover on bedrock channels:  applicability of existing models

Mishra, Jagriti; Inoue, Takuya

doi:https://doi.org/10.5194/esurf-8-695-2020

Articles | Volume 8, issue 3

https://doi.org/10.5194/esurf-8-695-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/esurf-8-695-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 8, issue 3

Review article

|

13 Aug 2020

Review article |

| 13 Aug 2020

Alluvial cover on bedrock channels: applicability of existing models

Jagriti Mishra and Takuya Inoue

Download

Final revised paper (published on 13 Aug 2020)
Preprint (discussion started on 04 Feb 2020)

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC1: 'Review of Mishra and Inoue', Anonymous Referee #1, 05 Mar 2020
- AC4: 'The Influence of Bed Roughness and Sediment Supply on Alluvial Cover in Bedrock Channels: Author Response', Jagriti Mishra, 13 May 2020
- AC5: 'The Influence of Bed Roughness and Sediment Supply on Alluvial Cover in Bedrock Channels: Author Response', Jagriti Mishra, 13 May 2020
RC2: 'Review: State of the art study of the influence of bed roughness and alluvial cover on bedrock channels and comparisons of existing models with laboratory scale experiments', Rebecca Hodge, 16 Mar 2020
- AC3: 'The Influence of Bed Roughness and Sediment Supply on Alluvial Cover in Bedrock Channels: Author Response', Jagriti Mishra, 13 May 2020
EC1: 'Editorial comment', Jens Turowski, 16 Mar 2020
- EC2: 'New paper on the cover effect', Jens Turowski, 20 Apr 2020
  - AC2: 'The Influence of Bed Roughness and Sediment Supply on Alluvial Cover in Bedrock Channels: Author Response', Jagriti Mishra, 13 May 2020
- AC1: 'The Influence of Bed Roughness and Sediment Supply on Alluvial Cover in Bedrock Channels: Author Response', Jagriti Mishra, 13 May 2020

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

AR by Jagriti Mishra on behalf of the Authors (13 May 2020) Author's response Manuscript

ED: Referee Nomination & Report Request started (20 May 2020) by Jens Turowski

RR by Anonymous Referee #1 (03 Jun 2020)

Suggestions for revision or reasons for rejection

Review of Mishra and Inoue

Thanks very much to the authors for all of their hard work on this revised manuscript. My comments in the first review mainly focused on the presentation (expanding the review, better explaining the models being tested, and clearly stating the experimental procedure). The authors have done a good job of addressing these concerns. I have a few remaining small comments below, but nothing major.

General comments:

1) I find sections 1.1-1.6 to be a great improvement. The structure of the paper, and the nature of the models being compared, is now quite a lot easier for readers to understand. Also, definitions of concepts such as cover hysteresis and clast rough/smooth have been moved closer to the beginning of the paper, which is very helpful. I have one final request to make with regard to the introduction/presentation of the models: please consider keeping consistent the order in which the models are discussed. In line ~125, the models are listed in the text. Then in sections 1.2-1.6, they are presented in a different order. Then in lines 553-555, they are listed again in an order that matches sections 1.2-1.6 but does not match the first list. Understanding this paper requires that readers keep a lot of concepts in mind at once; please try to make it as easy for them as possible by keeping the model order consistent.
2) Abstract: the results reported in the abstract are vague: “the results suggest a fit of certain models…” Consider revising the abstract to better clarify your results (for example, that the two simplest models do well on clast-rough but not clast-smooth beds). As much as we might like to believe otherwise, the abstract is all that some people will ever read!
3) I appreciate the much-improved introduction, with its broader scope and much more complete inclusion of the recent literature. I thought hard about why I still find the revised introduction a little bit difficult to follow, and I think it’s because the purpose of a given paragraph is not always made clear in its first sentence. For example, a couple of paragraphs just start with sentences describing what a single previous study did, so it’s hard to understand what the paragraph is trying to say up front. This is up to the authors, but the introduction might be easier to understand if each paragraph started with a more general statement. For example, the second paragraph (lines 34-50) might be easier to read if it started with the sentence (from lines 49-50) “The ratio of sediment supply to capacity controls the alluvial cover ratio, bedrock incision rate, and morphodynamics in bedrock rivers.” This way, readers immediately see the point of all the descriptions of past work.
4) Terminology: the paper repeatedly states that the study evaluates the “efficiency” of models. To most people I think this would imply that you are testing how quickly simulations can be run. Really what you are testing is the “applicability” (or any similar word of your choice) of each model to a benchmark set of experiments.

Line comments:

15: Similar to general comment #2, the last sentence of the abstract could be a little bit more specific about which models did best, and what their basic limitations are.

95: The model of Shobe et al (2017) also includes both the cover effect and a statement of mass conservation (based on entrainment/deposition).

135: This sentence is confusing; consider re-wording or deleting the second clause for clarity.

138: typo: linearly.

149-150: Please add a quick (~1 sentence) explanation of what you mean by this. How/why is their assumption incorrect?

177: I’ve never seen a footnote in a paper before—it might be better to just incorporate it into the text.

221-224: As with general comment #1, please just make sure that whenever you list all the models they occur in the same order.

280: “unless” should be “until,” I think.

374: consider spelling out “with respect to” or using another phrase

375: Do you mean “for example?”

419: Missing an “and” in the list of models.

436: This sentence uses “Run#” terminology as in the initial paper draft, whereas the authors have helpfully changed their terminology to be “Gravel#/Net#.” As such, this sentence doesn’t match up with the figure and the run names should be changed in the text.

450: Needs a comma instead of a period.

490-494: This sentence is confusing; consider breaking it up into multiple sentences.

546-547: Do you mean the change in alluvial cover with the sediment supply rate?

553: Again, just make sure to keep the model order constant in whichever order you think is best.

567: “into a plane two dimensional” is confusing; consider rewording.

Hide

RR by Rebecca Hodge (10 Jun 2020)

Suggestions for revision or reasons for rejection

Overall I think that the authors have done a good job of editing this paper, but there are still a number of sections that could be clearer in places and some ideas could be further developed in the discussion. I have attached a copy of the paper with detailed comments, but present some more general ones below.

Section 1 covers a wide range of relevant papers, but it’s still not clear what the main points are. It might help to structure it around a series of ideas (e.g. the role of bedload, relationships between sediment supply and sediment cover, roughness), and then use the papers to explain how these ideas have been investigated. The current structure just presents one study after another, making it difficult to identify what the key ideas are, and when the argument is moving on from one idea to the next. The same applies to most of section 1.1, with the exception of the final paragraph.

Both hydraulic and bedrock topographic roughness lengths are calculated for all beds, and shown to be related to each other (Fig 2). It is far easier to calculate the bedrock roughness length from topographic data, in contrast to calculating the hydraulic roughness from flow data. Can you comment on whether it would be possible to use the bedrock topographic length to predict flow and sediment cover, or do you need the hydraulic data to make accurate predictions?

My understanding is that Johnson’s model predicts the roughness length from the topography, whereas Inoue’s model uses the measured value. Does this difference explain why Inoue’s model better predicts the data? If you substituted the measured hydraulic roughness lengths into Johnson’s model, would that improve its performance?

It’s not entirely clear to me how the values of qbca are calculated for the flume data in Fig. 10 onwards. Is it using one of the initial equations, or was it measured in the flume? Furthermore, was the initial sediment feed rate calculated to be equal to the transport rate for an alluvial bed? It’s notable that for Gravel5 and Net2 the qbs/qbca values are mainly greater than one, but this is not mentioned in the text, and I think that it should be pointed out.

Referee Report: PDF

Hide

ED: Publish subject to minor revisions (review by editor) (11 Jun 2020) by Jens Turowski

Dear authors,

thank you for the revised manuscript. We were able to solicit two reviews from the original reviewers, and both agree that the paper is much improved. However, both reviewers see room for improvement in presentation, writing and language and ask for a number of clarifications.

I agree with this assessment from my own reading. You have done a nice job including the extra literature and putting the matieral into an overall coherent structure. Summarizing the weaknesses:

1) The literature review in the beginning of the paper can still be improved. I agree with both reviewers that over most of the section, it reads like a collection of disconnected statements about various papers, without a coherent summary of the state of knowledge, a critical assessment of what we do and do not know, and a synthesis. Both reviewers make concrete suggestions about how you can improve this. Please rewrite to establish a clear line of argument, leading to a few points that go beyond a few statements about what previous authors have done.

2) The description of the selected models can still be improved. This is much better than in the original manuscript, but partially, the physical and rational basis of the models is still clear. With respect to the tests, the differences in model prediction could be better worked out. This aspect can also be picked in the discussion, by putting the model performance into the context of the assumptions that went into their construction (see also reviewer #2).

3) I would appreciate clear conclusions, if possible with recommendations of which models are suitable under which circumstances. At the moment, you provide a summary, but no conclusions. Also, a summary figure including all models and fits (e.g., the RMSE of all models plotted for the different experimental runs), and a table summarizing the goodness of fit statistics would be helpful.

4) Finally, the manuscript would benefit from thorough language editing.

I decide on minor revisions, since the changes will affect the writing, but do not entail further analyses. In adition to the revised manuscript, please submit a detailed rebuttal to the comments.

With best wishes, Jens Turowski

Comments by line
23 …in controlling landscape evolution and determining the morphology of rivers…
24-30 Nice to see that the list of references is comprehensive. Still, for readability, it may be good to limit it to some key papers at this point.
Some of the references are listed in the wrong groups. Shepherd 1972 is an experimental paper and Hobley et al. 2011 a field paper. Sklar and Dietrich 1998 do not describe experiments, but develop concepts and the basics of the model published later in their 2004 based on general field observations. Please check through. Some of the references do not suggest the existence of the tools and cover effects, but build on these notions, and are therefore ill-placed in the list.
32 The notion of impacts of particles driving the erosion should be mentioned here.
35 …at the Erlenbach…
38 unclear, rewrite for clarity. Maybe there is a word confusion and you mean ‘hardly’ instead of ‘merely’.
44 …showed the formation…
54 …roughness of the bedrock bed…
56 …excavating a channel into natural bedrock…
60 …is proportional to…
115 …introduced a reach-scale…
145 typo, Turowski
149 you need to be more specific here on which part exactly was incorrect. The derivation of the exponential model for mass is correct, the transformation from mass to fluxes is not correct.
149 It is unclear why this model is tested even though it is deemed to be incorrect.
181 Grammatical problem in the sentence about Meyer-Peter and Müller. Unclear.
193 What does this mean that Inoue et al. directly use hydraulic roughness? Please clarify.
210 They did not propose it, it’s a prediction of the model (for the specific assumption of equal probability of deposition on exposed and cover bed).
212 I do not understand this statement. Similar to which other models? And similar in what way?
212 Any probability can take values between 0 and 1.
213 It may help a reader unfamiliar with that paper to explain the rationale behind this equation.
214 …and potentially other variables…
215 unclear, which exponent?
217 I am not sure where you found this, but I am pretty sure that we did not say this.
219 None of the parameters have obscure physical meaning. The meaning of the parameters is clearly explained in the paper. We do not have any observational constraints on some of the parameters, but this is something your analysis could provide.
230 …sandbar conditions
233 …were formed…
234 width-depth ratios
238 This could do with some more detail. What kind of mortar? What was the sand fraction? Did you quantify the hardness in some way? Was the mortar erodible by the bedload?
238-240 The last sentence repeats information of the previous sentence. This could be consolidated.
273 The term ‘validation’ does not fit here.
350-353 This is an interpretation and belongs to the discussion.
367 This is an interpretation and belongs to the discussion.
378 The bed does not try anything. Please reformulate.
380 …when the bed…
387-392 This seems to be a description of methods to me.
390 calibrated
397-407 It is unclear to me if one of the models can be rejected on grounds of the data.
406 the statement here is self-contradicting. Are they consistent or not?
436 What is ‘rather scarce deposition’? Can you make a quantitative rather than a qualitative statement?
440…increases, the cover does no increase…
467 what does ‘accurately’ mean in this context?
469 Why is that particularly important?
495 An optimization procedure is not possible?
501 For alluvial beds, different measures of roughness have been investigated by Schneider et al., WRR 2015
503-509 You could discuss the fitted values of omega here, and their physical meaning and implications.
504 Again, an optimization, for example using Chi squared, should be possible.
504 Actually, in this case, physically, you would need to use different P functions for entrainment and deposition.
507 The mentioning of the Turowski 2020 sits a bit odd here, and the statement of the model content is misleading. Either provide some context or remove.
509 The sentence gives next to no information to the reader. Bramante et al. looked at tools and cover effects in oscillatory flows and successfully fitted the Turowski and Hodge value. For context, you could cross-compare the fit values of omega.
Fig. 12: For some of the plots, especially a, the cover function based on the exponential model may provide a better fit (eq. 39 in the Turowski and Hodge paper). This would be equivalent to the limit in which omega approaches 1.
This is a side note, but I still wonder why it is not possible to evaluate the P-function directly. In my mind that would lead to deeper insights into the physics.
551 When sediment supply exceeds transport capacity, the bed is abruptly covered by sediment and quickly reaches a completely alleviated bed.
557 The statement here is somewhat misleading, because a calibration of some of the model parameters to the experimental conditions is required.

Hide

AR by Jagriti Mishra on behalf of the Authors (22 Jun 2020) Author's response Manuscript

ED: Publish subject to technical corrections (23 Jun 2020) by Jens Turowski

Dear authors,

thanks for the revised paper. I am happy with the changes and think the manuscript is nearly ready for publication. There are a number of small issues, which I ask you to correct. These are largely technical, so I have decided on technical corrections, meaning that the paper is passed to production directly when you upload the new version.

Thanks for working constructively with the reviewers and me on the manuscript. I believe the quality has tremendously improved in the process. As a result, your paper will be a very valuable contribution to the literature on the cover effect.

Best wishes,
Jens Turowski

There seems to be a problem with the figure numbering. Figures 7 & 8 exist twice and 12 and 13 are missing. Please correct this.
7 maybe ‘predict’ instead of ‘mimic’
26 …known as the tools and cover effects
27 Sediment acts as a tool for erosion by…
42 Their field surveys of the bedrock gorge cut by the Daan River…
43 …that the channel was not eroded…
73 Predicting the tools and cover effects…
88 …using a straight channel…
93 also as ‘run-away alluviation’
96 Add the citation of Hodge and Hoey 2016a to this statement.
117 …happens…
118-122 This is still a collection of unrelated statements. It is unclear why the reader should be interested in this.
142 Only the linear and exponential models, not the Turowski and Hodge model.
161 …determines whether sediment deposition is more likely on…
232 Equations
233 …we employed…
235 …refer to…
438 …do not include free model parameters.
449 The rebuttal states that the fit is now made by minimizing RMSE. Apparently, the manuscript has not been updated regarding this change.
477 Figure 12 seems to be missing from the manuscript.
480 see comment on line 449.
529-531 This sentence would be better placed in the figure caption.
597 proportionally
618 …capture bar formation…

Hide

ED: Publish subject to technical corrections (29 Jun 2020) by Andreas Lang (Editor)

AR by Jagriti Mishra on behalf of the Authors (02 Jul 2020) Author's response Manuscript

Short summary

This study provides an extensive review of field- and laboratory-scale studies and mathematical models used for predicting the sediment processes in bedrock river channels. We tested the model capabilities by reproducing and comparing the results with laboratory-scale experiments. This study provides an insight into the strengths and shortcomings of various available models.