the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 24 Jan 2023
The potential of global coastal flood risk reduction using various DRR measures
Abstract. Coastal flood risk is a serious global challenge facing current and future generations. Several disaster risk reduction (DRR) measures have been posited as ways to reduce the deleterious impacts of coastal flooding. On the global scale, however, efforts to model the effects of DRR measures (beyond structural) in the future are limited. In this paper, we use a global-scale flood risk model to estimate the risk of coastal flooding, and to assess and compare the effectiveness and economic performance of various DRR measures, namely: dykes and coastal levees, dry-proofing of urban assets, zoning restrictions in flood-prone areas, and management of foreshore vegetation. To assess the effectiveness of each DRR measure, we determine the extent to which they can limit future flood risk as a percentage of regional GDP to the same value as today (the so-called relative-risk constant objective). To assess their economic performance, we estimate the economic benefits and costs. If no DRR measures are taken in the future, we estimate expected annual damages to exceed $2 trillion USD by 2080, directly affecting an estimated 15 million people. Over 90 % of sub-national regions in the world can achieve their relative-risk constant targets if at least one of the investigated DRR measures is employed. At the global scale, we find the effectiveness of dykes and coastal levees in achieving the relative-risk constant objective to be 98 %, dry-proofing to be 49 %, zoning restrictions to be 11 %, and foreshore vegetation to be 6 %. In terms of direct costs, the overall figure is largest for dry-proofing ($151 billion) and dykes and coastal levees ($86 billion), much more than those of zoning restrictions ($27 million) and foreshore vegetation ($366 million). While zoning restrictions and foreshore vegetation achieve the highest global benefit-cost ratios, they also provide the least benefits overall. We show that there are large regional patterns in both the effectiveness and economic performance of modelled DRR measures. Future research could assess the indirect costs and benefits of these four and other DRR measures as well as their subsequent hybridisation.
- Preprint
(1727 KB) - Metadata XML
-
Supplement
(526 KB) - BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on nhess-2022-284', Anonymous Referee #1, 03 Mar 2023
The manuscript by Mortensen et al. employs a global flood model to estimate the effectiveness of a range of disaster risk reduction (DRR) measures in limiting future flood risk, focusing on potential direct impacts. In their assessment, the authors use the constant relative-risk objective, with respect to the regional Gross Domestic Product (GDP). The paper is well structured, clearly written, the results are presented in detail and the limitations are acknowledged and outlined in a clear manner. The study produces interesting results regarding the effectiveness of the different measures in reducing risk and regarding the cost-benefit ration of these measures and are therefore interesting for developing regional strategies to manage coastal flood risk.
I would recommend the study for publication – nevertheless, I would like to post some comments that I believe need to be addressed or may be useful for the authors to improve the paper:
- I am unsure why the authors refer to the measures that they are exploring as DRR measures. I do understand that floods can be disastrous but, if I am not mistaken, the authors are not assessing risk based only on high-impact low-probability events that could lead to disasters; rather, they estimate flood probabilities integrating over a range of return periods (ranging from events with 2-year return period, which hardly constitute disasters, to events with 1000-year return period). To my knowledge, these measures are usually referred to in the literature as coastal adaptation measures or grouped under the IPCC coastal adaptation typology categories. I find this potentially confusing and would suggest the authors to either explain clearly why the term DRR is used or refer to the IPCC terminology for coastal adaptation.
- The authors use a GIS-based inundation model, considering attenuation of water levels and, if I am not mistaken, waves. It is however not clear to me how waves have been accounted for in the total water level. Could the authors please clarify? Also, I would assume that the wave values that the authors are using refer to offshore waves; or does the model propagate waves to the near shore? (and how near is the “near shore”, since wave height will change considerably as waves approach the coast)
- Although the authors outline very clearly the limitations of the study, there is hardly any discussion on uncertainty and how this is addressed – where do the main uncertainties in the results stem from? I guess it would be too much to ask the authors to conduct an uncertainty analysis but there is substantial literature regarding flood risk assessments and the authors should at least discuss this issue.
- Following my previous point, uncertainty (in e.g. socio-economic development) is often addressed with the use of scenarios. The authors use only one scenario combination (SSP2-RCP6.0), which is a middle of the road scenario; I am unsure what the value of this is since it gives us practically no information about the potential range of uncertainty. In this case, either a second scenario should be used or the authors should rather opt for a high- or low-end scenario which would indicate the upper or lower boundaries. Of course, there is value in comparing the different measures, however, in a different scenario combination results could look very different.
- I understand the need for a no-measures assessment. However, I believe that it should be clearly pointed out that this is just a theoretical exercise since, in reality, there will be a response to flooding and adaptation will take place in one form or another, at some point in the century.
- Foreshore vegetation can be partly effective in reducing flood risk – however, a high-end event would destroy foreshore vegetation, thus limiting its protective effects for the years to come. I assume that this has not been considered, I however believe that it would be useful to discuss.
- My last point is a suggestion: based on my experience, many of the differences in global flood impact assessments stem from the calculation of the floodplains. I would personally find it useful if the authors would make their floodplains freely available (not only upon request as this usually does not work) so that others can use them to produce estimates that are comparable. I believe there could be a lot of added value for the research community if everyone conducting global or continental impact assessments made their floodplains openly available.
I hope my comments help the authors to strengthen this very good manuscript.
Citation: https://doi.org/10.5194/nhess-2022-284-RC1 - AC1: 'Reply on RC1', Eric Mortensen, 02 Jun 2023
-
RC2: 'Comment on nhess-2022-284', Anonymous Referee #2, 21 Apr 2023
In this study, the authors estimate the effectiveness of DRR measures for coastal flooding and provide sub-national risk estimates. This is a complex topic given the dynamics in hazard, exposure and vulnerability components. DRR measures are very important for reducing flood risk. Firstly, thank you for addressing this important aspect in flood risk management.
The authors mention that one of the novel aspects of the study is the global scale of analysis. Unfortunately, I have major concerns regarding the assumptions behind the risk computation and hence, the overall take away from this study.
I like the concept of risk constant. However, many other assumptions are quite vague to generalize. The possibility to implement DRR measures and their effectiveness to reduce risk are very diverse across regions and countries. For example, the assumption such as a constant % of dry-proofed area and urban cell composition are too simplified for a cost analysis and could be wrong for many regions. The same with the generalized costs of zoning. The authors do mention that as a limitation, however, it is a significant limitation that questions the credibility and usability of the results presented.
I see that two out of the four DRR measures – dikes and foreshore vegetation are part of the previous work done by the co-authors (as cited in this manuscript). The new findings are the effects of dry-proofing and zoning (please clarify if I am missing something here).
I strongly believe that there is a definite need to motivate the implementation of DRR measures. However, the generalized assumptions made in the study without considering local processes make the risk numbers at the Global level questionable. Also, the authors have not provided uncertainty ranges or any sort of validation for any of the reported values (e.g. EAD and EAAP; risk-reduction due to measures).
I sincerely appreciate the intention to provide a Global quantification of effectiveness of DRR measures. However, I recommend that the authors analyze the effectiveness of DRRs (especially building- and community-level measures) considering local and regional dynamics with region-specific datasets and knowledge and then integrate them in such a global study.
Citation: https://doi.org/10.5194/nhess-2022-284-RC2 - AC2: 'Reply on RC2', Eric Mortensen, 02 Jun 2023
Status: closed
-
RC1: 'Comment on nhess-2022-284', Anonymous Referee #1, 03 Mar 2023
The manuscript by Mortensen et al. employs a global flood model to estimate the effectiveness of a range of disaster risk reduction (DRR) measures in limiting future flood risk, focusing on potential direct impacts. In their assessment, the authors use the constant relative-risk objective, with respect to the regional Gross Domestic Product (GDP). The paper is well structured, clearly written, the results are presented in detail and the limitations are acknowledged and outlined in a clear manner. The study produces interesting results regarding the effectiveness of the different measures in reducing risk and regarding the cost-benefit ration of these measures and are therefore interesting for developing regional strategies to manage coastal flood risk.
I would recommend the study for publication – nevertheless, I would like to post some comments that I believe need to be addressed or may be useful for the authors to improve the paper:
- I am unsure why the authors refer to the measures that they are exploring as DRR measures. I do understand that floods can be disastrous but, if I am not mistaken, the authors are not assessing risk based only on high-impact low-probability events that could lead to disasters; rather, they estimate flood probabilities integrating over a range of return periods (ranging from events with 2-year return period, which hardly constitute disasters, to events with 1000-year return period). To my knowledge, these measures are usually referred to in the literature as coastal adaptation measures or grouped under the IPCC coastal adaptation typology categories. I find this potentially confusing and would suggest the authors to either explain clearly why the term DRR is used or refer to the IPCC terminology for coastal adaptation.
- The authors use a GIS-based inundation model, considering attenuation of water levels and, if I am not mistaken, waves. It is however not clear to me how waves have been accounted for in the total water level. Could the authors please clarify? Also, I would assume that the wave values that the authors are using refer to offshore waves; or does the model propagate waves to the near shore? (and how near is the “near shore”, since wave height will change considerably as waves approach the coast)
- Although the authors outline very clearly the limitations of the study, there is hardly any discussion on uncertainty and how this is addressed – where do the main uncertainties in the results stem from? I guess it would be too much to ask the authors to conduct an uncertainty analysis but there is substantial literature regarding flood risk assessments and the authors should at least discuss this issue.
- Following my previous point, uncertainty (in e.g. socio-economic development) is often addressed with the use of scenarios. The authors use only one scenario combination (SSP2-RCP6.0), which is a middle of the road scenario; I am unsure what the value of this is since it gives us practically no information about the potential range of uncertainty. In this case, either a second scenario should be used or the authors should rather opt for a high- or low-end scenario which would indicate the upper or lower boundaries. Of course, there is value in comparing the different measures, however, in a different scenario combination results could look very different.
- I understand the need for a no-measures assessment. However, I believe that it should be clearly pointed out that this is just a theoretical exercise since, in reality, there will be a response to flooding and adaptation will take place in one form or another, at some point in the century.
- Foreshore vegetation can be partly effective in reducing flood risk – however, a high-end event would destroy foreshore vegetation, thus limiting its protective effects for the years to come. I assume that this has not been considered, I however believe that it would be useful to discuss.
- My last point is a suggestion: based on my experience, many of the differences in global flood impact assessments stem from the calculation of the floodplains. I would personally find it useful if the authors would make their floodplains freely available (not only upon request as this usually does not work) so that others can use them to produce estimates that are comparable. I believe there could be a lot of added value for the research community if everyone conducting global or continental impact assessments made their floodplains openly available.
I hope my comments help the authors to strengthen this very good manuscript.
Citation: https://doi.org/10.5194/nhess-2022-284-RC1 - AC1: 'Reply on RC1', Eric Mortensen, 02 Jun 2023
-
RC2: 'Comment on nhess-2022-284', Anonymous Referee #2, 21 Apr 2023
In this study, the authors estimate the effectiveness of DRR measures for coastal flooding and provide sub-national risk estimates. This is a complex topic given the dynamics in hazard, exposure and vulnerability components. DRR measures are very important for reducing flood risk. Firstly, thank you for addressing this important aspect in flood risk management.
The authors mention that one of the novel aspects of the study is the global scale of analysis. Unfortunately, I have major concerns regarding the assumptions behind the risk computation and hence, the overall take away from this study.
I like the concept of risk constant. However, many other assumptions are quite vague to generalize. The possibility to implement DRR measures and their effectiveness to reduce risk are very diverse across regions and countries. For example, the assumption such as a constant % of dry-proofed area and urban cell composition are too simplified for a cost analysis and could be wrong for many regions. The same with the generalized costs of zoning. The authors do mention that as a limitation, however, it is a significant limitation that questions the credibility and usability of the results presented.
I see that two out of the four DRR measures – dikes and foreshore vegetation are part of the previous work done by the co-authors (as cited in this manuscript). The new findings are the effects of dry-proofing and zoning (please clarify if I am missing something here).
I strongly believe that there is a definite need to motivate the implementation of DRR measures. However, the generalized assumptions made in the study without considering local processes make the risk numbers at the Global level questionable. Also, the authors have not provided uncertainty ranges or any sort of validation for any of the reported values (e.g. EAD and EAAP; risk-reduction due to measures).
I sincerely appreciate the intention to provide a Global quantification of effectiveness of DRR measures. However, I recommend that the authors analyze the effectiveness of DRRs (especially building- and community-level measures) considering local and regional dynamics with region-specific datasets and knowledge and then integrate them in such a global study.
Citation: https://doi.org/10.5194/nhess-2022-284-RC2 - AC2: 'Reply on RC2', Eric Mortensen, 02 Jun 2023
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 737 | 294 | 26 | 1,057 | 37 | 17 | 16 |
- HTML: 737
- PDF: 294
- XML: 26
- Total: 1,057
- Supplement: 37
- BibTeX: 17
- EndNote: 16
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1