Back calculation of the 2017 Piz Cengalo–Bondo landslide cascade with r.avaflow: what we can do and what we can learn

Mergili, Martin; Jaboyedoff, Michel; Pullarello, José; Pudasaini, Shiva P.

doi:https://doi.org/10.5194/nhess-20-505-2020

Articles | Volume 20, issue 2

https://doi.org/10.5194/nhess-20-505-2020

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

https://doi.org/10.5194/nhess-20-505-2020

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

Articles | Volume 20, issue 2

Research article

|

21 Feb 2020

Research article |

| 21 Feb 2020

Back calculation of the 2017 Piz Cengalo–Bondo landslide cascade with r.avaflow: what we can do and what we can learn

Martin Mergili, Michel Jaboyedoff, José Pullarello, and Shiva P. Pudasaini

Download

Final revised paper (published on 21 Feb 2020)
Preprint (discussion started on 08 Jul 2019)

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC1: 'Review of the manuscript: "Back-calculation of the 2017 Piz Cengalo-Bondo landslide cascade with r.avaflow"', Anonymous Referee #1, 09 Aug 2019
- AC1: 'Reply to the comments of Referee #1', Martin Mergili, 04 Nov 2019
RC2: 'Review of Mergili et al., 2017 Piz Cengalo-Bondo landslide runout modelling', Brian McArdell, 25 Sep 2019
- AC2: 'Response to the comments of Referee #2', Martin Mergili, 04 Nov 2019

Peer-review completion

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

ED: Reconsider after major revisions (further review by editor and referees) (07 Nov 2019) by Margreth Keiler

AR by Martin Mergili on behalf of the Authors (08 Nov 2019)

ED: Referee Nomination & Report Request started (19 Nov 2019) by Margreth Keiler

RR by Anonymous Referee #1 (10 Dec 2019)

Suggestions for revision or reasons for rejection

Dear authors. I appreciate your answer that has addressed all of my comments. I think that the paper now is clearer and that in this form is somewhat more interesting for the readers (a case study is always just a case study, the narration of the struggling with the modelling of a case study, with all the problems well addressed, is more interesting). I still have some comments, but they are minor ones.
Broad comments:
1) I still have some issues with the title, consider to incorporate “key challenges” – a form you use yourself in line 73 – in the title
2) Even though it is good that you explicit the fact that δ scales linearly with the solid fraction, you would agree with me that it is somewhat strange that the solid particles of your flow increase their friction angle along their path. Probably you also increase the fluid fraction and the “real, combined, effective δ” (the equivalent of a single phase model) is lower but if so I still believe that this should be explicated in the table with another column (either solid fraction or “effective δ”). Moreover, if in zone E the flow is so channelized how comes that no entrainment occurred (lateral instabilities for examples) when such a mass of solid and fluid material flowed in its path? Are zone D and E all composed by rocky outcrops (it does not seem so from the pictures)?
3) Table 3. The threshold of 50% for empirical adequacy allows clearly a large space of wiggle room and it is maybe too generous. However the importance is that it is explicated in the text so the reader could weight better the results

Specific comments
1) Line 57-61: this sentence is not clear and well articulated
2) Line 149: approximately
3) Line 387: please explicit what these “numerical issues related to the narrow gorge” are

Hide

RR by Brian McArdell (13 Dec 2019)

Suggestions for revision or reasons for rejection

Re-review of NHESS manuscript submitted by Mergili et al., Back-calculation of the 2017 Piz Cengalo-Bondo landslide cascade with r.avaflow

The manuscript has been substantially improved from the last version. It describes the difficulties in modelling the Piz Cengalo-Bondo mass movement event cascade. I still have a few questions about the manuscript that I think should be considered before publication. I still think that the manuscript will be of interest to the readers of NHESS.

Lines 75-79: I’m not sure what the policy of NHESS is regarding repetition and word count, but lines 75—79 don’t really add much to what has already been written in the previous page and therefore could be omitted. Instead, I would prefer more test to explain some novel aspects of your model, e.g. the nature of the coupling between the phases (another comment on that below).

Line 134: What is MNS? I presume that it’s the French-language abbreviation of the English term DSM? Most readers won’t be familiar with the French abbreviation. Additionally, why did you use the surface model from Swisstopo instead of the terrain model? The surface model would also include vegetation, so using a surface model from 2011 would have included about 6 years of vegetation growth for areas not modified by the 2011 event, and vegetation removed by the 2017 event would lead to systematic errors in your mass balance. Many trees which survived the 2011 rock avalanche were destroyed by the 2017 event, so this could easily lead to errors. I presume that this has been accounted for by the authors, however it may not be clear to readers.

Section 3: Please include more details on how the fluid and solid phases are coupled. You gave carefully-worded explanations in the response to the reviews, and it would certainly be interesting for the reader to see the details of how the phases are coupled—even if only in plain text without equations. When looking at the video of the rock avalanche event and the videos of the various debris-flow surges in the village of Bondo, it is clear that phase separation is important, and this manuscript would be incomplete without at least some description of the coupling, as well as some description of the difference in travel speed of the two phases.

Line 238 ff: The model has been tuned by adjusting the coefficients in six different regions of the runout zone. This strongly suggests that the model cannot be used for prediction without calibration. To help the reader understand how important the parameter tuning exercise is for applying the model, perhaps you could compare the results with results calibrated based only on an inspection of the aerial photographs, maps, and other materials prior to the 2017 event.

Section 5.1: Are the statistical metrics CSI and D2PC from Formetta et al., 2015? I haven’t heard of these metrics before, or perhaps not with these names. This section is a bit confusing to read because the meaning of the statistical metrics and critical values of those parameters is only roughly explained.

Line 283: What is the justification of using a fluid with a density of 1400 kg m^-3? Is this an empirical value or based on some physical assumptions? This value seems to be quite large, and appears to have been arbitrarily defined. It is large enough that your solid concentrations will consequently appear to be quite small.

Line 328: Please explain what you mean by “leaks” or re-write the sentence to more clearly explain the effect.

Line 357 & 358. Please remove the sentence about this model being considered “best practice.” Given only one case study, with all of the tuning that went into reproducing the observations, and the other limitations described in the preceding paragraphs, this statement cannot be really taken seriously.

Line 362: Given what I’ve read of the Iverson & George model, it appears to also be useful for geophysical flows that would be comparable to the ones that you describe in the paper. I think that this statement in the paper is incorrect. I encourage you to focus on the advantages of your own model instead of highlighting possibly fictitious disadvantages of other models.

Line 400: The sediment leaving the area was of major interest to the communities downstream. This is part of the process cascade that would be relevant for hazard managers. If your model is capable of describing sediment transport (suspended and bedlam), then you may want to highlight that here but state that you did not focus on that aspect of the work.

Conclusions: This section is difficult to read: you may also want to include some more information about the case that you modelled in the paper and why: the conclusions should also contain a concise summary of the problem, the methods, results, and major discussions points.

Table 1, Zone c: The exponent on the CE parameter is typeset as “-8-0”. Is this a typo?

Hide

ED: Publish subject to minor revisions (review by editor) (15 Dec 2019) by Margreth Keiler

AR by Martin Mergili on behalf of the Authors (20 Dec 2019) Author's response Manuscript

ED: Publish as is (16 Jan 2020) by Margreth Keiler

AR by Martin Mergili on behalf of the Authors (17 Jan 2020) Manuscript

Short summary

Computer simulations of complex landslide processes in mountain areas are important for informing risk management but are at the same time challenging in terms of parameterization and physical and numerical model implementation. Using the tool r.avaflow, we highlight the progress and the challenges with regard to such simulations on the example of the Piz Cengalo–Bondo landslide cascade in Switzerland, which started as an initial rockslide–rockfall and finally evolved into a debris flow.