Uncertainties in coastal flood risk assessments in small island developing states

Parodi, Matteo U.; Giardino, Alessio; van Dongeren, Ap; Pearson, Stuart G.; Bricker, Jeremy D.; Reniers, Ad J. H. M.

doi:https://doi.org/10.5194/nhess-20-2397-2020

Articles | Volume 20, issue 9

https://doi.org/10.5194/nhess-20-2397-2020

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

https://doi.org/10.5194/nhess-20-2397-2020

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

Articles | Volume 20, issue 9

Research article

|

11 Sep 2020

Research article |

| 11 Sep 2020

Uncertainties in coastal flood risk assessments in small island developing states

Matteo U. Parodi, Alessio Giardino, Ap van Dongeren, Stuart G. Pearson, Jeremy D. Bricker, and Ad J. H. M. Reniers

Download

Final revised paper (published on 11 Sep 2020)
Preprint (discussion started on 16 Dec 2019)

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC1: 'Review for Parodi et al. NHESS article', Anonymous Referee #1, 19 Jan 2020
- AC1: 'Authors reply to Reviewer 1', Matteo Ulisse Parodi, 01 Apr 2020
RC2: 'Reviewer's comments', Anonymous Referee #2, 05 Feb 2020
- AC2: 'Authors reply to Reviewer 2', Matteo Ulisse Parodi, 01 Apr 2020
AC3: 'Authors Reply Letter to the reviewers', Matteo Ulisse Parodi, 12 Apr 2020
- AC4: 'Marked Up version', Matteo Ulisse Parodi, 12 Apr 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) (03 Apr 2020) by Animesh Gain

AR by Matteo Ulisse Parodi on behalf of the Authors (12 Apr 2020) Author's response Manuscript

ED: Referee Nomination & Report Request started (16 Apr 2020) by Animesh Gain

RR by Anonymous Referee #1 (18 Apr 2020)

Suggestions for revision or reasons for rejection

The paper incorporates most of my earlier comments. However, still, a few things need to be clarified. The paper analyzes uncertainty among different ESL components and found DEM was the main uncertain component. However, no significant statistical tests were performed to partition sources of uncertainty.

[1] In Abstract, line 14, “input data with uncertain accuracy” replace with “input uncertainty”. Further, in Abstract, line 23, “predictive error”; this is not an error, it is “predictive uncertainty”

[2] In Page 2, Line 26, double period.

[3] In Page 2, Line 31, “…often damage curves are taken from literature and applied in different areas making few if any, adjustments ..” In your case as well, you have taken return level estimates from Muis et al. which is again from disparate data sources and directly implemented for your analyses. So what are the difference b/n 2nd order (where damage curve being deployed from different sources) and first-order (in your case, you are taking the specific quantile corresponds to 100-yr RL) sources of uncertainty estimates?

[4] In Page 3, Line 20, any literature that points to the vulnerability of coastal flooding in the area considered? Please cite.

[5] Page 4, Line 27, the small differences in intensity at different return levels could be shown in the Appendix.

[6] In Page 5, Line 7, for significant wave-height, please provide the information regarding variable used in parentheses; authors possibly used “Significant height of combined wind waves and swell" at 6-hourly timescale archived at ERA-Interim. But it would be wise to mention that for further reproducibility of the result.

[7] In Page 7, line 21, the statement, “The choice of RCP 8.5 relies on the fact that the 90% confidence interval of this projection also captures most of the 90% confidence interval values under the RCP 4.5 scenario.” This statement is not so clear, RCP8.5 is altogether a different scenario when GHG emission reaches 8.5 W/m2 and considered to be most extreme; an overlap b/n two scenarios may misrepresent risk.

[8] In Page 11, line 18, the goal of the work is to model multiple sources of uncertainty among different components; then why not authors have performed a simple ANOVA analysis, where you determine if there is any statistically significant difference b/n multiple independent groups and which component contribute more to projection uncertainty? Right now, a simple sensitivity analysis was performed to show DEM is the main contributor of uncertainty, however, whether the identified factor is statistically significant is not yet shown. Authors may refer (Gangrade et al., 2020) for this.

[9] In Page 12, line 3-4, instead of cumbersome Monte Carlo based analyses a simple ANOVA may be performed to know which factor has significant contribution in overall uncertainty assessment.

[10] In Page 12, line 10-11, while the cited paper analyzed compound flooding between storm surge and river discharge, (Ganguli and Merz, 2019) shows how compound flooding resulting from total coastal water level significantly impact the risk of inland riverine flooding over continental Europe.

References

Gangrade, S., Kao, S.-C. and McManamay, R. A.: Multi-model Hydroclimate Projections for the Alabama-Coosa-Tallapoosa River Basin in the Southeastern United States, Scientific Reports, 10(1), 1–12, doi:10.1038/s41598-020-59806-6, 2020.

Ganguli, P. and Merz, B.: Extreme Coastal Water Levels Exacerbate Fluvial Flood Hazards in Northwestern Europe, Sci Rep, 9(1), 1–14, doi:10.1038/s41598-019-49822-6, 2019.

Hide

RR by Anonymous Referee #3 (27 May 2020)

ED: Publish subject to minor revisions (review by editor) (02 Jun 2020) by Animesh Gain

AR by Matteo Ulisse Parodi on behalf of the Authors (16 Jun 2020) Author's response Manuscript

ED: Publish subject to minor revisions (review by editor) (27 Jun 2020) by Animesh Gain

AR by Matteo Ulisse Parodi on behalf of the Authors (30 Jun 2020) Author's response Manuscript

ED: Publish as is (17 Jul 2020) by Animesh Gain

AR by Matteo Ulisse Parodi on behalf of the Authors (29 Jul 2020)

Short summary

We investigate sources of uncertainty in coastal flood risk assessment in São Tomé and Príncipe, a small island developing state. We find that, for the present-day scenario, uncertainty from depth damage functions and digital elevation models can be more significant than that related to the estimation of significant wave height or storm surge level. For future scenarios (year 2100), sea level rise prediction becomes the input with the strongest impact on coastal flood damage estimate.