Reproducing an extreme flood with uncertain post-event information

Fuentes-Andino, Diana; Beven, Keith; Halldin, Sven; Xu, Chong-Yu; Reynolds, José Eduardo; Di Baldassarre, Giuliano

doi:https://doi.org/10.5194/hess-21-3597-2017

Articles | Volume 21, issue 7

https://doi.org/10.5194/hess-21-3597-2017

© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 3.0 License.

https://doi.org/10.5194/hess-21-3597-2017

© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 3.0 License.

Articles | Volume 21, issue 7

Research article

|

17 Jul 2017

Research article |

| 17 Jul 2017

Reproducing an extreme flood with uncertain post-event information

Diana Fuentes-Andino, Keith Beven, Sven Halldin, Chong-Yu Xu, José Eduardo Reynolds, and Giuliano Di Baldassarre

Download

Final revised paper (published on 17 Jul 2017)
Supplement to the final revised paper
Preprint (discussion started on 28 Sep 2016)

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

SC1: 'Some comments', Aristotelis Tegos, 03 Oct 2016
- AC1: 'Reply to A. Tegos (SC1)', Diana Fuentes, 12 Dec 2016
- AC2: 'Reply to Referee #1', Diana Fuentes, 12 Dec 2016
RC1: 'Less emphasis on the methods and more on the analyses', Anonymous Referee #1, 16 Oct 2016
- AC3: 'Reply to Referee #1', Diana Fuentes, 12 Dec 2016
RC2: 'Flood modeling under large data uncertainty within the GLUE framework', Anonymous Referee #2, 30 Oct 2016
- AC5: 'Reply to Referee #2', Diana Fuentes, 12 Dec 2016
SC2: 'Review on "Reproducing an extreme flood with uncertain post-event information" by D. Fuentes-Andino et al.', Anouk Sprong, 31 Oct 2016
- AC6: 'Reply to Anouk Sprong (SC2)', Diana Fuentes, 12 Dec 2016
SC3: 'Comment on “Reproducing an extreme flood with uncertain post-event information” by Diana Fuentes-Andino et al. Ilja America', Ilja America, 02 Nov 2016
- AC7: 'Reply to Ilja America (SC3)', Diana Fuentes, 12 Dec 2016
SC4: 'Interactive comment on “Reproducing an extreme flood with uncertain post-event information” by Diana Fuentes-Andino et al.', Kenechukwu Okoli, 08 Nov 2016
- AC8: 'Reply to Kenechukwu Okoli (SC4)', Diana Fuentes, 12 Dec 2016
SC5: 'Interactive comment on “Reproducing an extreme flood with uncertain post-event information” by Diana Fuentes-Andino et al.', Timo Kelder, 11 Nov 2016
- AC9: 'Reply to Timo Kelder (SC5)', Diana Fuentes, 12 Dec 2016
RC3: 'Review of "Reproducing an extreme flood with uncertain post-event information"', Anonymous Referee #3, 05 Dec 2016
- AC10: 'Reply to Referee #3', Diana Fuentes, 12 Dec 2016
AC4: 'Major_revision', Diana Fuentes, 12 Dec 2016

Peer-review completion

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

ED: Reconsider after major revisions (further review by Editor and Referees) (27 Dec 2016) by Roger Moussa

Dear Authors,

The manuscript presents an interesting work using a coupled hydrological-hydraulic model calibrated against post-event surveys in order to simulate the 1998 extreme flood in Tegucigalpa induced by the Hurricane Mitch. One of the main originality of this paper is the combination of field observations and hydrological and hydraulic modelling as a support for data interpretation. All three reviewers say that the paper is of international interest, but the three reviewers raised fundamental questions concerning the uncertainties in data which will affect modelling results, the comparability between the different post-event data, the lack of validation of the model, the need that the manuscript be more self-contained, the consolidation of the state-of-the-art, a discussion on strategies for improvements on the post-event data collection campaigns, the added value and the domain of application of the method proposed.

I would like to thank Aristotelis Tegos, Ryan Teuling and the students Anouk Sprong, Ilja America and Timo Kilder of the Master programme Earth & Environment at Wageningen University, Giuliano Di Baldassarre and the student Kenechukwu Okoli from the Department of Earth Science at Uppsala University, for their valuable comments. I would like also to especially thank the authors for responding point by point to all comments.

The authors' responses address in detail the majority of the points addressed by the three reviewers and all the other comments. The paper will be reconsidered after “Major revisions” for a complete review process. In the revised version, authors are invited to respond point by point to each comment of each reviewer, conducting additional analysis, and giving convincing proofs to all problems raised by reviewers. Please highlight clearly what you changed in the revised manuscript, so the reviewers are able to assess your changes.

Kind regards,

Hide

AR by Diana Fuentes on behalf of the Authors (25 Jan 2017) Author's response Manuscript

ED: Referee Nomination & Report Request started (30 Jan 2017) by Roger Moussa

RR by Anonymous Referee #2 (06 Feb 2017)

RR by Anonymous Referee #1 (05 Mar 2017)

Suggestions for revision or reasons for rejection

Several modifications and additions are included in the second version of the manuscript, and described in an accompanying note. But, these modifications are mostly minor, many comments of the reviewers where acknowledged and even included in the manuscript, but not really answered. Basically the results and their analysis remain unchanged, where my feeling is that a more in depth analysis was necessary to enrich the proposed paper.
I am finally hardly overall convinced by the content of the manuscript. It was the sense of my numerous question and comments in my initial review. The authors focus on the technique and its implementation and seem to forget, doing this, what is the real objective of such a post event analysis: learn from the observations about the underlying processes. If we consider that the calibrated model parameter values summarize what has been learnt from the confrontation between observations and models, the conclusion of the study should be that very little has been learnt. The initial uncertainties about the parameter values could hardly be reduced (fig. 6 and 7). No surprise considering the scarcity of the available set of observation (even the rainfall amounts are relatively unknown). Presented in this way, this result is really disappointing. It is of course due to the scarcity of the available data set but also to the method used. Calibrating all the models and all their parameters using all available data, everything put in a single “pot” without further strategy, without separating and sorting out the various questions (runoff production, flood wave propagation, stage-discharge relations), without a real critical analysis of the available data, could hardly lead to a different result. In do not share the statement of the authors that the method proposed could be taken as a reference example.
The paper is well written, but I can finally hardly imagine what its real contribution to the hydrological or hydraulic science is. Roughly said, it shows that a large variety of combinations of parameters may produce “behavioural” results in a largely over-parameterized case study. No surprise at all. I do not really see what other sort of conclusion can be drawn from the presented approach. Since the models could not really be calibrated, I do not share the final the conclusion that the presented work could be of any help for useful planning and flood forecasting. The application of the Glue method does not hide that the manuscript is relying on much too scarce, inaccurate data set, furthermore analysed using a much too coarse approach.

Hide

RR by Anonymous Referee #3 (10 Mar 2017)

Suggestions for revision or reasons for rejection

The authors describe a way to reproduce an extreme flood event due to hurricane Mitch in Tegucigalpa in Honduras thanks to TOPMODEL and LISFLOOD-FP modelling tools. The problematic and the methodology are well explained. The points of the conclusion are clear. The article has been fully improved by the authors. The topic of the article is fully in the scope of the journal.

Some minor points have to be fixed before publishing this article, that is why I recommend minor revisions.

p. 2 l.4 "to reasonable reproduce" I think it should be "to reproduce reasonably".
p. 3 l.2 "planning and prevention?" I think that the interrogation mark should be removed.
p.3 Concerning what is described from line 14 to line 27 which is very interesting, it might be nice to have a figure which summarises the methodology which is given here.
p.4 l.20 "is The Gen...": "The" should be changed into "the".
p.5 l.15 "... in areal means are likely, ..." I do not understand this sentence, I think that something is missing. It needs to be clarified.
p.7 l.4 "1D Mike 11 ..." a reference should be given concerning this hydraulic modelling tool.
p.7 l.28 "a uniform ..." should be changed into "an uniform ...".
p.8 l.17 "reason to selecting ..." I think it should be "reason to select ...".
p.9 l.26-27 "in Fig. 1 and 2 and Table1 ..." should be changed into "in Figs. 1 and 2 and 2 and Table 1 ...".
p.11 l.20 There is a mix between the physically based/mathematical model and the numerical method which is not very easy to understand. Moreover there is an error in the mass equation. So I recommend to write the mathematical model first and then to say few words on the way they are solved. Which should give :
\partial h/ \partial t +\partial Qx / \partial x +\partial Qy/ \partial y=0
\partial Qx/ \partial t = gA(Sx-n^2Qx|Q|/(R^(4/3)A^2) )
\partial Qy/ \partial t = gA(Sy-n^2Qy|Q|/(R^(4/3)A^2) )
where Q=(Qx,Qy), Sx (Sy) is the opposite of the slope in x (y) direction. It seems to be the kinematic wave model with an extra term.
p.11 l.23 "where Qis ..." into "where Q is".
p.11 l.26 "This model formulation is computationally more efficient than a full 2D dynamic model ..." it should be told why and by which factor.
p.13 l.33 "... given to predicting ..." I think it should be "... given to predict ...".
p.17 l.3 "The mean rainfall multiplier ... were ..." it should be plural on multiplier or singular on were.
p.17 l.8 "... for the these ..." the should be removed.
p.19 l.1 "the one-dimensional HEC-RAS in ..." should be changed into "the one-dimensional HEC-RAS model on ...". And a reference should be given concerning this modelling tool.
p.23 formula (A3) a point is missing at the end of the formula.
p.23 formula (A4) the same.
p.23 ( NWF) should be changed into (NWF).
p.23 l.10 and l.12 these are the only references for which "et al." is in italic. I recommend to put all Latin in italic (el al., i.e., ...).
p.24 "An initial ..." should be changed into "an initial ...".
p.24 l.2 "B1." should be changed into "B1:".
p.24 At the end of formula (B1) a point should be added.
p.24 l.4 "Eq. B2." into "Eq. B2:".
p.25 l.9 formula (B12) "...Ot," into "... Ot.".
p.25 l.16 formula (C1) a point should be added.
I recommend to center all the formulas.
All figures are fine accept figure 6 which is too small.

Hide

ED: Publish subject to revisions (further review by Editor and Referees) (20 Mar 2017) by Roger Moussa

AR by Diana Fuentes on behalf of the Authors (29 May 2017) Manuscript

ED: Publish as is (08 Jun 2017) by Roger Moussa

AR by Diana Fuentes on behalf of the Authors (13 Jun 2017)

Short summary

Reproduction of past floods requires information on discharge and flood extent, commonly unavailable or uncertain during extreme events. We explored the possibility of reproducing an extreme flood disaster using rainfall and post-event hydrometric information by combining a rainfall-runoff and hydraulic modelling tool within an uncertainty analysis framework. Considering the uncertainty in post–event data, it was possible to reasonably reproduce the extreme event.