An efficient two-layer landslide-tsunami numerical model: effects of momentum transfer validated with physical experiments of waves generated by granular landslides

Franz, Martin; Jaboyedoff, Michel; Mulligan, Ryan P.; Podladchikov, Yury; Take, W. Andy

doi:https://doi.org/10.5194/nhess-21-1229-2021

Articles | Volume 21, issue 4

https://doi.org/10.5194/nhess-21-1229-2021

© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/nhess-21-1229-2021

© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 21, issue 4

Research article

|

20 Apr 2021

Research article |

| 20 Apr 2021

An efficient two-layer landslide-tsunami numerical model: effects of momentum transfer validated with physical experiments of waves generated by granular landslides

Martin Franz, Michel Jaboyedoff, Ryan P. Mulligan, Yury Podladchikov, and W. Andy Take

Download

Final revised paper (published on 20 Apr 2021)
Preprint (discussion started on 20 Jan 2020)

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC1: 'Review of the paper "An efficient two-layer landslide-tsunami numerical model: effects of momentum transfer validated with physical experiments of waves generated by granular landslides"', Anonymous Referee #1, 17 Feb 2020
- AC1: 'Answer to Anonymous referee #1', Martin Franz, 23 Aug 2020
RC2: 'Represent the ordinary shallow-water equations still the state-of-the-art in landslide-tsunami modelling?', Anonymous Referee #2, 20 Feb 2020
- AC2: 'Answer to Anonymous Referee #2', Martin Franz, 23 Aug 2020

Peer-review completion

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

ED: Reconsider after major revisions (further review by editor and referees) (09 Sep 2020) by Mauricio Gonzalez

AR by Katja Gänger on behalf of the Authors (05 Nov 2020) Author's response Manuscript

ED: Referee Nomination & Report Request started (05 Nov 2020) by Mauricio Gonzalez

RR by Anonymous Referee #2 (14 Nov 2020)

Suggestions for revision or reasons for rejection

Summary:

The Authors improved many aspects in the manuscript and the modelling of the slide phase includes some advancements with the limitations (e.g. the shape factor as a fitting parameter) clearly communicated. However, the modelling of the landslide-tsunamis is a step backward (shallow-water equations (SWEs), excluding frequency dispersion) in my view, reflecting the state of the art 10-15 years ago. The Authors do not fully acknowledge these limitations in the article, despite the concerns of both Reviewers. Yes, there is a new statement on L368 about these limitations, but these limitations also need to be made clear in other key sections such as the Introduction, Conclusions, during the discussions of Figs. 4 and 9 (where the limitations of the applied model are obvious), etc. Further, there are still some linguistic issues, typos and smaller inconsistencies. Finally, as already highlighted in the 1st review, I do not think that some key conclusions are supported by the findings in the article. More details are given below.

Specific comments:

L39: Good that the RANS equations are now introduced as well as an alternative to the model applied by the Authors, but the full term needs to be introduced, not only the abbreviation. More background into other options (DNS, LES) would also be desirable.

L46/L185: It is not clear what the Authors mean by “free-surface nature”. Do not all mentioned models have a free surface? Or do they mean the way the free surface is treated (e.g. Volume of Fluids)?

L315 and Fig. 9: “…the numerical simulation reproduces the wave observed in the physical experiment very well in terms of amplitude and timing at each probe.” The “very well” is subjective and needs to be backed up with a more objective criterion (goodness of fit parameter). It is fine if the Authors do not want to use the nRMSE, but they need to use an alternative way to scientifically quantify the goodness of fit (e.g. a % deviations between the amplitudes), as already pointed out by both Reviewers in the 1st review round in other contexts.

Fig. 9 (first panel): The numerical wave appears to be cut? This makes it impossible to appreciate the quality of the agreement.

L359: This new paragraph is helpful showing that a simulation is performed efficiently. However, I believe this section is at the incorrect place in the manuscript (it should be part of the Methodology). Further, have convergence tests been performed?

L396: “…overall, the model effectively handles the complex phenomenon occurring during the interaction between the landslide and the water.” I do not fully agree with this conclusion, given that the model cannot model impact craters and wave breaking, as stated in the manuscript on L365: “This lack also implies that the model cannot consider the impact crater as long as the steepness of the water surface is not steeper than sub-vertical”. I do not think many scientists and engineers would find the shallow-water equations (SWEs) a wise choice to model subaerial landslide-tsunami generation.

L402 and Fig. 10: “…by the fact that the results of our model also fit well with those experiments.” I understand that this mainly based on the comparison in Fig. 10. Yes, the data scatter between the +/-30% for P < 9 (investigated in the original study). However, it is obvious that the data trend systematic deviates from the prediction, for both P < 9 and more significantly over the full presented range of P. It is further obvious that the trend of the data conducted for water depths of up to 0.17 m is very different for the experiments conducted at water depths of 0.10, 0.08 and 0.05 m. There are clear recommendations in the technical literature that laboratory experiments should not be performed certain Weber and Reynolds number limitations (roughly corresponding to water depths smaller than 0.20 m) to avoid significant scale effects. There appears currently to be no attempt in the manuscript to explain this systematic deviation in Fig. 10 with scale effects, or another potential explanation offered by the Authors.

L403: “Finally, our model is validated by a benchmark test performed herein, as this approach is very simple to implement and is very efficient in terms of computational resources. Therefore, we consider our model as a tool of choice for the assessment of landslide-generated tsunami hazards.” I agree that the model is very efficient, but I do not feel that “Therefore” is justified as the efficiency of a tool is not the only criterion for hazard assessment (it also needs to represent the underlying physics to an appropriate level, as appreciated by the Authors in the article). I do not think it would be a wise choice to select a SWEs model for landslide-tsunami hazard assessment due to points L396 and L402 above and the fact that the model is not able to consider frequency dispersion (partially responsible for the dangerous underestimation in Fig. 9 in some of the panels). I do not think the answer from the Authors in the response “The comments of both reviewers show that they understand the statement of the balance “correctness-efficiency” and that we had to make a choice of the level of approximation. We chose the non-linear shallow water equations (without multiple add-ons), hence, the inherent approximations and incomplete physics (such as frequency dispersion). We won’t discuss in the paper the well-known limitations of this approach.” is satisfactory. There are efficient, more appropriate and widely applied alternatives to the SWEs (non-hydrostatic non-linear SWEs, RANS equations, coupled approaches, GPU acceleration, etc.) to model landslide-tsunamis available. The Reviewers review the article not for themselves, but on behave of the readership of the Journal. My feeling as a Reviewer is that 95% of the readership of the Journal are not fully aware of the “correctness-efficiency” aspects and the fundamental limitations of the SWEs to model subaerial landslide-tsunami generation and propagation and will be misled by such statements if the limitations are not better communicated in the article.

Suggested grammatical corrections and minor points:

Abstract: There are still formatting aspects which should be improved, e.g. the abstract should be presented as 1 paragraph (not several small ones) and the text/figures need to be arranged such that there are no large free spaces on the pages (see e.g. at the bottom of page 10 or 13).

L57: Please add the missing free space in “whichis”.

L64: Please drop one of the “that”.

L84: Please revise the unclear expression “static critical state friction”.

L90: Please replace “paper” (rather informal) with “article”.

L96: I do not think “eq. 1” is the common writing style in this journal, normally it would be “(1)”. Please check and apply it correctly for all equations.

L100: Again, it is common practice to write parameters in italic in research articles. This is still a general issue in the article, there are many parameters with an inconsistent writing style (L100, L106, L147, L218, caption Fig. 2, L252, L254, L255, L256, L257, L279, captions of Figs. 5, 7, 8, 9 and 10 and many times in the Notation).

L121 (Eq. (8)): On the other hand, numbers, “sin”, “cos” and “tan”, Fr, P, Re and brackets should not be written in italic (Eqs. (8), (9), (10), (11) and (12), L147, L153, L155, Eqs. (20), (21) and (22), Table 1, L248, L329, L331, L335, L336, L338, L341, Eq. (37), L345, Eq. (39), L350, L355, L356, L400 and in the Notations).

L181: Please write “…when applied in our, never fit the experimental data.”

Legend of Fig. 2: Please improve the presentation and remove the typo in “Differnce”

Figs. 4, 6, 7 and 8: Should it read “z” and “x” rather than “Z” and “X” on the axes? Further, some of the numbers on the axes are too small.

L212-L218 and legend of Fig. 2: The writing style of the shape factor is inconsistent (SF, FS, S_F (F as subscript)).

L231: Please write “…in two-dimensions.”

L235: Please write “…the momentum transfer is… the wave propagates…”

L257 and L258: A full stop after “al” is required.

L266: The meaning of the word “discrimination” in this context is unclear.

L272: Please write “…identify the best…”

L284: Please write “…is the finally chosen rheological model.”

L339: The parameters have just been introduced, so just write “The relationships between P and Fr and S…”

L370: Please write “…lack of modelling frequency dispersion…”.

Notation: P is used for two different parameters. Also S is used twice, this should not be the case.

References: Please remove inconsistencies such as inconsistent use of upper and lower case letters in the article titles, abbreviation and no abbreviation of Journal titles, typos (e.g. L437 “genrated”), etc. This was already pointed out in the 1st review round.

Hide

RR by Anonymous Referee #1 (17 Nov 2020)

ED: Reconsider after major revisions (further review by editor and referees) (21 Nov 2020) by Mauricio Gonzalez

AR by Martin Franz on behalf of the Authors (02 Jan 2021) Author's response Author's tracked changes Manuscript

ED: Publish as is (27 Jan 2021) by Mauricio Gonzalez

Short summary

A landslide-generated tsunami is a complex phenomenon that involves landslide dynamics, wave dynamics and their interaction. This phenomenon threatens numerous lives and infrastructures around the world. To assess this natural hazard, we developed an efficient numerical model able to simulate the landslide, the momentum transfer and the wave all at once. The good agreement between the numerical simulations and physical experiments validates our model and its novel momentum transfer approach.