the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 27 Jul 2022
Holistic planning of human, water, and environmental impacts for regional flood management: A case study of aging dam infrastructure
Abstract. Urbanization and climate change have challenged the structural integrity of flood-control dams through increased storage requirements and internal water pressures. Many existing dams are aging and have been classified as deficient or having potential for life-threatening floods in the event of failure, thereby necessitating rapid and innovative mitigation strategies (e.g., optimized timing of releases, emergency warning systems, property buyouts, additional storage, diversion levees, underground tunnels). Such alternatives are often screened primarily through a cost-benefit analysis (CBA), where measures of flood-risk reduction are quantified according to inundation bounds and implementation costs. Secondary impacts associated with dam-induced flooding, such as environmental triggers (e.g., toxic pollutant releases, wastewater dispersion, soil erosion, habitat disruption) or social vulnerabilities (e.g., medical needs, language barriers, reinforced poverty, housing challenges), are often included at the screening stage as a series of narratives and are therefore largely indeterminant when ranking alternative strategies. This tendency to screen mitigation strategies through the lens of flood inundation may prioritize solutions with strong hydrological benefits while minimizing additional impacts associated with widespread flooding. To address this gap, we compare a reservoir mitigation strategy using traditional CBA metrics with composite socio-environmental risks through geospatial multi-criteria decision analysis (MCDA) and scenario-based hydrologic/hydraulic modelling. We demonstrate a case study of alternative mitigation options associated with the Addicks and Barker Reservoirs in Houston, Texas, USA under Hurricane Harvey rainfall conditions and compare performance outcomes between the traditional CBA approach and the spatial MCDA approach. This study illustrates how preferred flood management strategies may shift when hydrologic outputs are integrated explicitly with socio-environmental factors at the preliminary screening stage. By leveraging the strengths of composite risk indicators and simplified spatial overlay methods, the MCDA framework aids decision-makers in visualizing multi-functional benefits from disparate mitigation options and provides an additional layer of information for optimizing the system.
This preprint has been withdrawn.
-
Withdrawal notice
This preprint has been withdrawn.
-
Preprint
(4351 KB)
-
Supplement
(7882 KB)
-
This preprint has been withdrawn.
- Preprint
(4351 KB) - Metadata XML
-
Supplement
(7882 KB) - BibTeX
- EndNote
Interactive discussion
Status: closed
-
CC1: 'Referee comment on nhess-2022-193', Francesco Ballio, 30 Aug 2022
I suggest rejecting this manuscript. I have some conceptual objections which, of course, could be discussed. The main problem here is that the manuscript is too confused, key points are unclear. Due to this, at the end of two readings, I was not able to form an opinion on the main typical points of such a research project, that is:
- How robustly do the chosen variables represent the economic/social/… impacts of a flood they are intended to represent?
- How robust are results (preferences among mitigation solutions) with respect to all uncertainties in the intermediate steps of the procedure?
- The fact that a MCA provides a wider view with respect to a CBA within evaluation of flood mitigation solutions is well established. Which is the (methodological?) innovation with respect to other MCA in flood mitigation problems?
Details are given below.
- Definition of R is not clear to me. Symbol of intersection in eq. (7) means a product? If yes, please notice that Rk is either equal to 0 (if Ak=A1) or equal to Ik (if A1=1, Ak=0). In a few cases it could be equal to -Ik, if the mitigation strategy generates a larger flooded area than the baseline.
- Is R a cell-defined variable or it is a sum over the whole domain? At line 480 I read about “impacts per spatial unit”. I R an extensive or intensive variable?
- What is exactly the variable on the x-axis of fig. 6? R? some percentage of R?
- Based of my understanding of eq. 7 a value of R larger than 0 means that the mitigation strategy is less impacting than the baseline. But this is exactly the opposite of what I read. This means that I could not understand the definition of R.
- I cannot understand how most of mitigation strategies produce larger impacts than the to-do-nothing option. Authors write that “this is likely due to the areal approach used to quantify risk change from the baseline scenario,” But this does not explain much to me.
- At the end, having all these doubts, fig. 7 is simply not understandable to me.
- AHP weights depend on the scenario. This appears inconsistent to me. If I want to compare different scenarios all variables should be weighted, even if not relevant for that scenario, so that weights remain constant. The fact that medical facility are more or less important with respect to amenity disruption is an absolute evaluation, which cannot depend on the presence of either one or the other. Otherwise, I should have different weights for each cell, not only for each scenario.
- What are the Social Vulnerability circles in fig. 5b?
- Presentation is confused. The case is complex, I understand this. Authors try to present it from different perspectives and through a variety of alternatives. This is nice, but, at the end, the picture is not clear.
- Use of terms hazard, vulnerability, exposure is not standard. I do not like to be rigid on terminology, but authors should discuss and justify their unconventional choice.
- Many minor details should be polished (examples: ABRS is first used and then defined; fig. 1b contains quantities that are discussed much later – no clear how “language” and “disability” can be a flood impact; fig 1a is totally useless; SAW is not defined; some information is repeated; fig. 5 and 6 are inverted in the discussion; …)
Citation: https://doi.org/10.5194/nhess-2022-193-CC1 - AC1: 'Reply on CC1', Hanadi Rifai, 20 Oct 2022
-
RC1: 'Comment on nhess-2022-193', Anonymous Referee #1, 03 Sep 2022
The authors brought up an important and timely topic: aging dams, their potential failures in flood events, and the need for life-protection strategies. However, their current version of the manuscript requires a fundamental review of the concepts, followed by adjustments and improvements in the methodology. I recommend major revision and reconsideration in the future.
Specific comments:
- Line 105: What means the expression: “tripartite coupling of human-water-environmental systems”? What is the “environment system” in the manuscript context? Perhaps, this new expression is not suitable and results in misunderstanding. I suggest remove it.
- Line 111-113: The definition of risk in these lines is different from that depicted in Equation 1.
- Line 116: “Hazard represents a shock that may be triggered by flooding, and which poses a negative consequence to regional health” and Figure 1 uses “toxic hazard”. Are hazards associated with floods, dam failures, or general pollutants? How do measure shock? Flood hazards should be related to the flood characteristics, for instance, return period, water level, velocity, …
- Line 119 and Line 384: The definition of risk is very confusing. Risk is expressed differently in Equation 1 and Equation 7. In the current version, the term was used in different senses causing misunderstandings for the reader.
- Table 4: Is social vulnerability one of the eight social impact factors? The concept of vulnerability also is confusing. What variables are really included in social vulnerability? Is Risk analysis about vulnerability or environmental and social impacts?
Some points:
- According to the NHESS guidelines, abbreviations need to be defined at the first instance in the text, for instance, SAW and ABRS.
- Some citations should be re-formatting.
- Line 386: “food inundation boundary”, it would be “flood inundation boundary”?
Citation: https://doi.org/10.5194/nhess-2022-193-RC1 - AC2: 'Reply on RC1', Hanadi Rifai, 20 Oct 2022
-
RC2: 'Comment on nhess-2022-193', Francesco Ballio, 05 Sep 2022
I suggest rejecting this manuscript. I have some conceptual objections which, of course, could be discussed. The main problem here is that the manuscript is too confused, key points are unclear. Due to this, at the end of two readings, I was not able to form an opinion on the main typical points of such a research project, that is:
- How robustly do the chosen variables represent the economic/social/… impacts of a flood they are intended to represent?
- How robust are results (preferences among mitigation solutions) with respect to all uncertainties in the intermediate steps of the procedure?
- The fact that a MCA provides a wider view with respect to a CBA within evaluation of flood mitigation solutions is well established. Which is the (methodological?) innovation with respect to other MCA in flood mitigation problems?
Details are given below.
- Definition of R is not clear to me. Symbol of intersection in eq. (7) means a product? If yes, please notice that Rk is either equal to 0 (if Ak=A1) or equal to Ik (if A1=1, Ak=0). In a few cases it could be equal to -Ik, if the mitigation strategy generates a larger flooded area than the baseline.
- Is R a cell-defined variable or it is a sum over the whole domain? At line 480 I read about “impacts per spatial unit”. I R an extensive or intensive variable?
- What is exactly the variable on the x-axis of fig. 6? R? some percentage of R?
- Based of my understanding of eq. 7 a value of R larger than 0 means that the mitigation strategy is less impacting than the baseline. But this is exactly the opposite of what I read. This means that I could not understand the definition of R.
- I cannot understand how most of mitigation strategies produce larger impacts than the to-do-nothing option. Authors write that “this is likely due to the areal approach used to quantify risk change from the baseline scenario,” But this does not explain much to me.
- At the end, having all these doubts, fig. 7 is simply not understandable to me.
- AHP weights depend on the scenario. This appears inconsistent to me. If I want to compare different scenarios all variables should be weighted, even if not relevant for that scenario, so that weights remain constant. The fact that medical facility are more or less important with respect to amenity disruption is an absolute evaluation, which cannot depend on the presence of either one or the other. Otherwise, I should have different weights for each cell, not only for each scenario.
- What are the Social Vulnerability circles in fig. 5b?
- Presentation is confused. The case is complex, I understand this. Authors try to present it from different perspectives and through a variety of alternatives. This is nice, but, at the end, the picture is not clear.
- Use of terms hazard, vulnerability, exposure is not standard. I do not like to be rigid on terminology, but authors should discuss and justify their unconventional choice.
- Many minor details should be polished (examples: ABRS is first used and then defined; fig. 1b contains quantities that are discussed much later – no clear how “language” and “disability” can be a flood impact; fig 1a is totally useless; SAW is not defined; some information is repeated; fig. 5 and 6 are inverted in the discussion; …)
Citation: https://doi.org/10.5194/nhess-2022-193-RC2 - AC3: 'Reply on RC2', Hanadi Rifai, 20 Oct 2022
Interactive discussion
Status: closed
-
CC1: 'Referee comment on nhess-2022-193', Francesco Ballio, 30 Aug 2022
I suggest rejecting this manuscript. I have some conceptual objections which, of course, could be discussed. The main problem here is that the manuscript is too confused, key points are unclear. Due to this, at the end of two readings, I was not able to form an opinion on the main typical points of such a research project, that is:
- How robustly do the chosen variables represent the economic/social/… impacts of a flood they are intended to represent?
- How robust are results (preferences among mitigation solutions) with respect to all uncertainties in the intermediate steps of the procedure?
- The fact that a MCA provides a wider view with respect to a CBA within evaluation of flood mitigation solutions is well established. Which is the (methodological?) innovation with respect to other MCA in flood mitigation problems?
Details are given below.
- Definition of R is not clear to me. Symbol of intersection in eq. (7) means a product? If yes, please notice that Rk is either equal to 0 (if Ak=A1) or equal to Ik (if A1=1, Ak=0). In a few cases it could be equal to -Ik, if the mitigation strategy generates a larger flooded area than the baseline.
- Is R a cell-defined variable or it is a sum over the whole domain? At line 480 I read about “impacts per spatial unit”. I R an extensive or intensive variable?
- What is exactly the variable on the x-axis of fig. 6? R? some percentage of R?
- Based of my understanding of eq. 7 a value of R larger than 0 means that the mitigation strategy is less impacting than the baseline. But this is exactly the opposite of what I read. This means that I could not understand the definition of R.
- I cannot understand how most of mitigation strategies produce larger impacts than the to-do-nothing option. Authors write that “this is likely due to the areal approach used to quantify risk change from the baseline scenario,” But this does not explain much to me.
- At the end, having all these doubts, fig. 7 is simply not understandable to me.
- AHP weights depend on the scenario. This appears inconsistent to me. If I want to compare different scenarios all variables should be weighted, even if not relevant for that scenario, so that weights remain constant. The fact that medical facility are more or less important with respect to amenity disruption is an absolute evaluation, which cannot depend on the presence of either one or the other. Otherwise, I should have different weights for each cell, not only for each scenario.
- What are the Social Vulnerability circles in fig. 5b?
- Presentation is confused. The case is complex, I understand this. Authors try to present it from different perspectives and through a variety of alternatives. This is nice, but, at the end, the picture is not clear.
- Use of terms hazard, vulnerability, exposure is not standard. I do not like to be rigid on terminology, but authors should discuss and justify their unconventional choice.
- Many minor details should be polished (examples: ABRS is first used and then defined; fig. 1b contains quantities that are discussed much later – no clear how “language” and “disability” can be a flood impact; fig 1a is totally useless; SAW is not defined; some information is repeated; fig. 5 and 6 are inverted in the discussion; …)
Citation: https://doi.org/10.5194/nhess-2022-193-CC1 - AC1: 'Reply on CC1', Hanadi Rifai, 20 Oct 2022
-
RC1: 'Comment on nhess-2022-193', Anonymous Referee #1, 03 Sep 2022
The authors brought up an important and timely topic: aging dams, their potential failures in flood events, and the need for life-protection strategies. However, their current version of the manuscript requires a fundamental review of the concepts, followed by adjustments and improvements in the methodology. I recommend major revision and reconsideration in the future.
Specific comments:
- Line 105: What means the expression: “tripartite coupling of human-water-environmental systems”? What is the “environment system” in the manuscript context? Perhaps, this new expression is not suitable and results in misunderstanding. I suggest remove it.
- Line 111-113: The definition of risk in these lines is different from that depicted in Equation 1.
- Line 116: “Hazard represents a shock that may be triggered by flooding, and which poses a negative consequence to regional health” and Figure 1 uses “toxic hazard”. Are hazards associated with floods, dam failures, or general pollutants? How do measure shock? Flood hazards should be related to the flood characteristics, for instance, return period, water level, velocity, …
- Line 119 and Line 384: The definition of risk is very confusing. Risk is expressed differently in Equation 1 and Equation 7. In the current version, the term was used in different senses causing misunderstandings for the reader.
- Table 4: Is social vulnerability one of the eight social impact factors? The concept of vulnerability also is confusing. What variables are really included in social vulnerability? Is Risk analysis about vulnerability or environmental and social impacts?
Some points:
- According to the NHESS guidelines, abbreviations need to be defined at the first instance in the text, for instance, SAW and ABRS.
- Some citations should be re-formatting.
- Line 386: “food inundation boundary”, it would be “flood inundation boundary”?
Citation: https://doi.org/10.5194/nhess-2022-193-RC1 - AC2: 'Reply on RC1', Hanadi Rifai, 20 Oct 2022
-
RC2: 'Comment on nhess-2022-193', Francesco Ballio, 05 Sep 2022
I suggest rejecting this manuscript. I have some conceptual objections which, of course, could be discussed. The main problem here is that the manuscript is too confused, key points are unclear. Due to this, at the end of two readings, I was not able to form an opinion on the main typical points of such a research project, that is:
- How robustly do the chosen variables represent the economic/social/… impacts of a flood they are intended to represent?
- How robust are results (preferences among mitigation solutions) with respect to all uncertainties in the intermediate steps of the procedure?
- The fact that a MCA provides a wider view with respect to a CBA within evaluation of flood mitigation solutions is well established. Which is the (methodological?) innovation with respect to other MCA in flood mitigation problems?
Details are given below.
- Definition of R is not clear to me. Symbol of intersection in eq. (7) means a product? If yes, please notice that Rk is either equal to 0 (if Ak=A1) or equal to Ik (if A1=1, Ak=0). In a few cases it could be equal to -Ik, if the mitigation strategy generates a larger flooded area than the baseline.
- Is R a cell-defined variable or it is a sum over the whole domain? At line 480 I read about “impacts per spatial unit”. I R an extensive or intensive variable?
- What is exactly the variable on the x-axis of fig. 6? R? some percentage of R?
- Based of my understanding of eq. 7 a value of R larger than 0 means that the mitigation strategy is less impacting than the baseline. But this is exactly the opposite of what I read. This means that I could not understand the definition of R.
- I cannot understand how most of mitigation strategies produce larger impacts than the to-do-nothing option. Authors write that “this is likely due to the areal approach used to quantify risk change from the baseline scenario,” But this does not explain much to me.
- At the end, having all these doubts, fig. 7 is simply not understandable to me.
- AHP weights depend on the scenario. This appears inconsistent to me. If I want to compare different scenarios all variables should be weighted, even if not relevant for that scenario, so that weights remain constant. The fact that medical facility are more or less important with respect to amenity disruption is an absolute evaluation, which cannot depend on the presence of either one or the other. Otherwise, I should have different weights for each cell, not only for each scenario.
- What are the Social Vulnerability circles in fig. 5b?
- Presentation is confused. The case is complex, I understand this. Authors try to present it from different perspectives and through a variety of alternatives. This is nice, but, at the end, the picture is not clear.
- Use of terms hazard, vulnerability, exposure is not standard. I do not like to be rigid on terminology, but authors should discuss and justify their unconventional choice.
- Many minor details should be polished (examples: ABRS is first used and then defined; fig. 1b contains quantities that are discussed much later – no clear how “language” and “disability” can be a flood impact; fig 1a is totally useless; SAW is not defined; some information is repeated; fig. 5 and 6 are inverted in the discussion; …)
Citation: https://doi.org/10.5194/nhess-2022-193-RC2 - AC3: 'Reply on RC2', Hanadi Rifai, 20 Oct 2022
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 495 | 230 | 27 | 752 | 78 | 11 | 16 |
- HTML: 495
- PDF: 230
- XML: 27
- Total: 752
- Supplement: 78
- BibTeX: 11
- EndNote: 16
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Cited
1 citations as recorded by crossref.
Cyndi Vail Castro
This preprint has been withdrawn.
- Preprint
(4351 KB) - Metadata XML
-
Supplement
(7882 KB) - BibTeX
- EndNote