the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 10 Jul 2024
Flood risk assessment through large-scale modeling under uncertainty
Abstract. The complexity of flood risk models is intrinsically linked to a variety of sources of uncertainty (hydrology, hydraulics, exposed assets, vulnerability, coping capacity, etc.) that affect the accuracy and reliability of the analyses. Estimating the uncertainties associated with the different components allows us to be more confident in the risk values on the ground, thus providing a more reliable assessment for investors, insurance and flood risk management purposes. In this study, we investigate the flood risk of the entire Central Apennines District (CAD) in Central Italy using the laRgE SCale inUndation modEl – Flood Risk, RESCUE-FR, focusing on the interaction between the uncertainty of the hydraulic Manning parameter and the risk variability. We assess the coherence between the quantile flood risk maps generated by our model and the official risk maps provided by the CAD authority and focusing on three specific zones within the CAD region. Thus, RESCUE-FR is used to estimate the Expected Annual Damage (EAD) and the Expected Annual Population Affected (EAPA) across the CAD region and to conduct a comprehensive uncertainty analysis. The latter provides a range of confidence of risk estimation that is essential for identifying vulnerable areas and guiding effective mitigation strategies.
- Preprint
(7057 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on nhess-2024-114', Anonymous Referee #1, 29 Aug 2024
The study presents a new approach to assessing flood risk, based on monetary assessments of damage and exposed population, considering a more simplified modeling, which allows its application on a large scale. The main novelty of the study is accounting the uncertainties associated with the Manning roughness parameter (n). The approach is innovative and useful to assist in the decision-making process to increase the resilience of localities to floods. The results presented interesting inferences and discussions. I really enjoyed reading the text and learning about the methodology applied. Below, I present some considerations that can be incorporated into the text to deepen the discussion and some doubts that I hope will be better clarified.
The authors mention in the introduction some statements that can be better detailed to bring greater value. For example: “Traditional methods for large-scale analysis often rely on empirical or simplified models that may overlook important spatial variability and uncertainty in flood hazard and impact estimates” – such as?; “…and multivariate, assuming that flood damage is influenced by several variables” - which ones?
In the methodology, the authors mention that “We adopt here a uniform level of protection by assuming the threshold value Q∗ in Eq. (1) equal to design return period of the flood protection works.” I found the approach interesting, but this is also a level of uncertainty, and I believe it is even more significant than the variation in manning coefficient. From an action planning and efficient decision-making point of view, the uncertainties associated with protection actions are more interesting and useful for evaluation than manning coefficient, right? Since it is important for decision-makers to know whether the protection actions adopted are effective or not, including from a spatial point of view. Manning is not a parameter that can be varied by them. Do you have any suggestions on how this could be incorporated into future work?
Line 137 - Why use the rational formula? It usually has good applicability for small basins, but in your case the basins are large. How were the C values determined for the different areas? Is the study area well delimited and coincident with the entire area of the basin? Is there no “external” flow entering the basin?
Figure 1 – Is the hydrography in blue?
I suggest including a brief description of the study area (with size) in the introduction, to familiarize the reader with the type of floods and scale that will be addressed. I later resolved many of my doubts about the methodology when I read the section on the study area and had a better understanding of the spatial scale.
Line 308 – C3 above 10^4 euro/year, right?
Line 303 – The authors states: “Note that agricultural land use has significantly different maximum damage values compared to the other land use types, varying by about two orders of magnitude.” But the reader cannot note this, since the authors don’t show this result, they present the risk value already in classes in Figure 2. Or does this variation in two orders of magnitude refer to the curves coming from Huzinga?
Either way, the authors provide many discussions in the next paragraphs in terms of LULC and the differences in the calculated maps and the ones obtained from PGRAAC. Therefore, I recommend including a figure of LULC in the study case section, to help the reader understand the discussions and draw their own conclusions.
What is the advantage of using this proposed methodology in an area that already has risk maps constructed with resolutions and greater data accuracy? What new information is provided by the authors' approach? Or were the area chosen precisely to allow a comparison with reliable results and then expand the use of the methodology to other areas that do not have this detail with greater reliability? I missed a bit of this discussion, and I think the innovation points presented in the introduction do not address these issues.
Line 362 – There is a repeated “the”.
Citation: https://doi.org/10.5194/nhess-2024-114-RC1 -
AC2: 'Reply on RC1', Luciano Pavesi, 24 Sep 2024
The comment was uploaded in the form of a supplement: https://nhess.copernicus.org/preprints/nhess-2024-114/nhess-2024-114-AC2-supplement.pdf
-
AC2: 'Reply on RC1', Luciano Pavesi, 24 Sep 2024
-
RC2: 'Comment on nhess-2024-114', Anonymous Referee #2, 30 Aug 2024
Pavesi et al have presented an estimation of expected regional flood risk taking into account uncertainty within the Manning coefficient. While the uncertainty is considered for only one parameter, the framework resented by the authors can be used to account for other uncertainties. The analysis provided is rigorous and comprehensive, and I only have a few minor comments to further improve the manuscript.
1. Is the same Manning coefficient used across the entire spatial domain in a single simulation?
2. If the same Manning coefficient is used across the entire spatial region, it will be interesting to see a plot of regional risk R_n with respect to Manning coefficient.
3. Line 120 - DTM is defined after its first usage.
4. Line 155 - It would be helpful for the reader if the authors elaborated on justification for their assumption - Uncertainty in Manning coefficient is considered as representative of most of the uncertainty characterizing the hydraulic modeling of flood risk.
5. Line 178 - The choice of normal distribution for Manning coefficient appears arbitrary. The authors have only stated that uniform distribution used by Pavesi et al. (2022) resulted in overestimation of water depth, hence they selected the normal distribution. The appropriate distribution should be chosen based on the inherent uncertainty of the parameter, instead of the distribution’s impact on the result. Is there any literature that can be referenced for the appropriate probability distribution of Manning coefficient?
6. Line 182 - Parenthesis around DTM are unnecessary.
7. Line 205 - between all the country in which -> among all the countries for which
8. Line 207 - Is there supposed to be `and` between GDP and PPS?
9. Line 308 - classed -> classed
10. Line 285 - What is the purpose of this statement? - Where Q100 is the 1 − 1/100 = 0.99 quantile of the probability distribution of the peak discharge.Citation: https://doi.org/10.5194/nhess-2024-114-RC2 -
AC1: 'Reply on RC2', Luciano Pavesi, 24 Sep 2024
The comment was uploaded in the form of a supplement: https://nhess.copernicus.org/preprints/nhess-2024-114/nhess-2024-114-AC1-supplement.pdf
-
AC1: 'Reply on RC2', Luciano Pavesi, 24 Sep 2024
Status: closed
-
RC1: 'Comment on nhess-2024-114', Anonymous Referee #1, 29 Aug 2024
The study presents a new approach to assessing flood risk, based on monetary assessments of damage and exposed population, considering a more simplified modeling, which allows its application on a large scale. The main novelty of the study is accounting the uncertainties associated with the Manning roughness parameter (n). The approach is innovative and useful to assist in the decision-making process to increase the resilience of localities to floods. The results presented interesting inferences and discussions. I really enjoyed reading the text and learning about the methodology applied. Below, I present some considerations that can be incorporated into the text to deepen the discussion and some doubts that I hope will be better clarified.
The authors mention in the introduction some statements that can be better detailed to bring greater value. For example: “Traditional methods for large-scale analysis often rely on empirical or simplified models that may overlook important spatial variability and uncertainty in flood hazard and impact estimates” – such as?; “…and multivariate, assuming that flood damage is influenced by several variables” - which ones?
In the methodology, the authors mention that “We adopt here a uniform level of protection by assuming the threshold value Q∗ in Eq. (1) equal to design return period of the flood protection works.” I found the approach interesting, but this is also a level of uncertainty, and I believe it is even more significant than the variation in manning coefficient. From an action planning and efficient decision-making point of view, the uncertainties associated with protection actions are more interesting and useful for evaluation than manning coefficient, right? Since it is important for decision-makers to know whether the protection actions adopted are effective or not, including from a spatial point of view. Manning is not a parameter that can be varied by them. Do you have any suggestions on how this could be incorporated into future work?
Line 137 - Why use the rational formula? It usually has good applicability for small basins, but in your case the basins are large. How were the C values determined for the different areas? Is the study area well delimited and coincident with the entire area of the basin? Is there no “external” flow entering the basin?
Figure 1 – Is the hydrography in blue?
I suggest including a brief description of the study area (with size) in the introduction, to familiarize the reader with the type of floods and scale that will be addressed. I later resolved many of my doubts about the methodology when I read the section on the study area and had a better understanding of the spatial scale.
Line 308 – C3 above 10^4 euro/year, right?
Line 303 – The authors states: “Note that agricultural land use has significantly different maximum damage values compared to the other land use types, varying by about two orders of magnitude.” But the reader cannot note this, since the authors don’t show this result, they present the risk value already in classes in Figure 2. Or does this variation in two orders of magnitude refer to the curves coming from Huzinga?
Either way, the authors provide many discussions in the next paragraphs in terms of LULC and the differences in the calculated maps and the ones obtained from PGRAAC. Therefore, I recommend including a figure of LULC in the study case section, to help the reader understand the discussions and draw their own conclusions.
What is the advantage of using this proposed methodology in an area that already has risk maps constructed with resolutions and greater data accuracy? What new information is provided by the authors' approach? Or were the area chosen precisely to allow a comparison with reliable results and then expand the use of the methodology to other areas that do not have this detail with greater reliability? I missed a bit of this discussion, and I think the innovation points presented in the introduction do not address these issues.
Line 362 – There is a repeated “the”.
Citation: https://doi.org/10.5194/nhess-2024-114-RC1 -
AC2: 'Reply on RC1', Luciano Pavesi, 24 Sep 2024
The comment was uploaded in the form of a supplement: https://nhess.copernicus.org/preprints/nhess-2024-114/nhess-2024-114-AC2-supplement.pdf
-
AC2: 'Reply on RC1', Luciano Pavesi, 24 Sep 2024
-
RC2: 'Comment on nhess-2024-114', Anonymous Referee #2, 30 Aug 2024
Pavesi et al have presented an estimation of expected regional flood risk taking into account uncertainty within the Manning coefficient. While the uncertainty is considered for only one parameter, the framework resented by the authors can be used to account for other uncertainties. The analysis provided is rigorous and comprehensive, and I only have a few minor comments to further improve the manuscript.
1. Is the same Manning coefficient used across the entire spatial domain in a single simulation?
2. If the same Manning coefficient is used across the entire spatial region, it will be interesting to see a plot of regional risk R_n with respect to Manning coefficient.
3. Line 120 - DTM is defined after its first usage.
4. Line 155 - It would be helpful for the reader if the authors elaborated on justification for their assumption - Uncertainty in Manning coefficient is considered as representative of most of the uncertainty characterizing the hydraulic modeling of flood risk.
5. Line 178 - The choice of normal distribution for Manning coefficient appears arbitrary. The authors have only stated that uniform distribution used by Pavesi et al. (2022) resulted in overestimation of water depth, hence they selected the normal distribution. The appropriate distribution should be chosen based on the inherent uncertainty of the parameter, instead of the distribution’s impact on the result. Is there any literature that can be referenced for the appropriate probability distribution of Manning coefficient?
6. Line 182 - Parenthesis around DTM are unnecessary.
7. Line 205 - between all the country in which -> among all the countries for which
8. Line 207 - Is there supposed to be `and` between GDP and PPS?
9. Line 308 - classed -> classed
10. Line 285 - What is the purpose of this statement? - Where Q100 is the 1 − 1/100 = 0.99 quantile of the probability distribution of the peak discharge.Citation: https://doi.org/10.5194/nhess-2024-114-RC2 -
AC1: 'Reply on RC2', Luciano Pavesi, 24 Sep 2024
The comment was uploaded in the form of a supplement: https://nhess.copernicus.org/preprints/nhess-2024-114/nhess-2024-114-AC1-supplement.pdf
-
AC1: 'Reply on RC2', Luciano Pavesi, 24 Sep 2024
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 304 | 95 | 236 | 635 | 9 | 10 |
- HTML: 304
- PDF: 95
- XML: 236
- Total: 635
- BibTeX: 9
- EndNote: 10
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
