A Cost-Effectiveness Protocol for Flood-Mitigation Plans Based on Leeds’ Boxing Day 2015 Floods
Round 1
Reviewer 1 Report
Recommendation: minor revision
It is a clear, well-written and logically built-up paper. It offers innovative, fresh, look into cost-effectiveness protocol for local flood-mitigation plans. The paper does not develop a general theory, but is practically useful, as a grass-root idea of a response to a practical need. It could be of value for local stakeholders: administration, municipal authorities, decision makers and inhabitants, in general.
The assumptions of 2m excess depth (line 537) and the critical river-level threshold, hT, are of primary importance in the paper, being, in fact arbitrary. The authors may wish to offer advice as to the objective algorithm of determination of hT. In fact, any flood depth causes damage (e.g. water in cellars of buildings is also a problem, generating losses).
Numbering of references should follow the order of appearance of references in the text. Hence, for example, it should start from [1] in line 25 and [2] in line 28, rather than the present numbers [32] and [21], respectively.
The authors should support the statement: “though the amount of summer rainfall has decreased overall, it has contemporaneously intensified“ (l. 26-27) by an appropriate British reference.
Damage estimates given in line 30 were exceeded by far in earlier floods in the UK (e.g. 2000, 2007), see Kron et al. (2019). Statement on flood generating mechanism in lines 30-32 is restricted to the UK conditions being examined (elsewhere, snowmelt runoff also occur, see Kron et al., 2019).
When discussing flood risk reduction, in lines 32-36, see Driessen et al. (2018) and Kundzewicz et al. (2018).
Line 78 Replace: “river levels returning on average 1 : 10, 1 : 25, 1 : 100 or 1 : 200 years” by: “river levels with probability of exceedance in any one year being 1 : 10, 1 : 25, 1 : 100 or 1 : 200”
Lines 93-94: „the flood volume is a key parameter in the design of mitigation measures”. Yes, it is one of key parameters. Flood peaks stage is important for design of flood walls and levees. Flood duration also matters.
How about flood hazards projections for the future? Climate track is mentioned in this paper via references to Hodgkins et al. (2017), IPCC (2013), Pacala and Socolow (2004). However, even if projections are uncertain (cf. Kundzewicz et al. 2017), flood hazard is likely to increase in much of Europe, including the UK. Hence, by referring to projections, the usefulness of the proposed approach can be strengthened.
Lines 556-557: “Particularly useful is to address the following question: what fraction of the chosen FEV, i.e., Ve(hT), is reduced by a particular flood-mitigation measure or policy?” It can be difficult, if at all possible, to answer this question, even in an approximate way.
Additional references for consideration:
Driessen, P. J. et al. (2018) Governance Strategies for Improving Flood Resilience in the Face of Climate Change. Water 10(11): #1595
Kron, W. et al. (2019) Reduction of flood risk in Europe - Reflections from a reinsurance perspective. Journal of Hydrology 576: 197-209.
Kundzewicz, Z.W. et al. (2017) Differences in flood hazard projections in Europe - their causes and consequences for decision making. Hydrological Sciences Journal. 62(1), 1-14.
Kundzewicz, Z. W., et al. (2018) Flood risk reduction: structural measures and diverse strategies. Proc. Nat. Acad. Sci. USA (PNAS) 115(49): 12321–12325, www.pnas.org/cgi/doi/10.1073/pnas.1818227115
Author Response
Responses to Reviewer 1
We thank the reviewer for the thorough and constructive review. More substantive changes in the revised manuscript have been highlighted in red, while smaller ones pertaining to tightening of the text have not.
It is a clear, well-written and logically built-up paper. It offers innovative, fresh, look into cost-effectiveness protocol for local flood-mitigation plans. The paper does not develop a general theory, but is practically useful, as a grass-root idea of a response to a practical need. It could be of value for local stakeholders: administration, municipal authorities, decision makers and inhabitants, in general.
We agree: our manuscript concerns an evidence-synthesis analysis/diagnostic and a roadmap/protocol for improving flood-mitigation design. It is important to highlight, however, that the roadmap we have introduced in section 4.1.1. is both novel and intended to be of practical use in the sense that its systematic approach to flood-mitigation design simulations is generally uncommon or not followed in current government and consulting-engineering practice in flood protection and design, despite the fact that this involves multi-million pound investments (as far as our experience goes within the UK), even though such a systematic approach may be more common in academia, where in general less, if any, money is at stake. Moreover, this type of evidence-synthesis paper is rather unusual in that it accommodates expertise from a multidisciplinary (and international) team of co-authors.
The assumptions of 2m excess depth (line 537) and the critical river-level threshold, hT, are of primary importance in the paper, being, in fact arbitrary. The authors may wish to offer advice as to the objective algorithm of determination of hT. In fact, any flood depth causes damage (e.g. water in cellars of buildings is also a problem, generating losses).
The precise choice, rather than assumption, of a 2m-depth is immaterial; however, reasonable flood depths of more serious floods range typically between 1m and 4m (see, e.g., contour plots of flood depth for the Houston floods). The figure of 1m was deemed too low in the sense that it neither lies in the middle of that 1m to 4m range nor instils a sense of peril on a human scale. We added a sentence as clarification at the end of section 3, and it was also explained in section 5, lines 6 and 7.
The choice of the critical threshold hT is neither arbitrary nor fixed but generally straddles a certain reasonable but not arbitrary interval. First, we do and did refer to the relevance of hT lying above 2.7m since that is the local first warning scale at Armley; second, we highlight the need of local knowledge which we estimated from photographic and eyewitness evidence to be around hT=4.1m, also referring to eyewitness observation, of near-flooding in mid-December 2015, at a height of 3.8m (which had not led to a major flooding situation at all). Finally, not only do we provide a graph of FEV as function of hT but also we discuss this in the penultimate paragraph of section 3: “While we have chosen hT=3.9m, its uncertainty will be around ±0.25m given that the first flooding occurred upstream of the Armley gauge station for a lower hT, per the gauge-station information, while the first water appeared in the streets for a higher hT=4.16m, per our photographic evidence. The resulting changes in FEV are, however, of the order of the uncertainty of 16% we have calculated. Note that the peak in mid-December was 3.8m, as discussed, and had not led to major flood damage. Hence, we continue with our Ve(hT=3.9m).” It is and was also discussed further both at the end of section 2 and in the summary of section 5.
Numbering of references should follow the order of appearance of references in the text. Hence, for example, it should start from [1] in line 25 and [2] in line 28, rather than the present numbers [32] and [21], respectively.
We used the MDPI standard style file, which did not work in, so in the end we ordered the references by hand.
The authors should support the statement: “though the amount of summer rainfall has decreased overall, it has contemporaneously intensified“ (l. 26-27) by an appropriate British reference.
To accommodate the remark, we have added the reference to Sanderson (2010), which already had appeared (later) in the original submission.
Damage estimates given in line 30 were exceeded by far in earlier floods in the UK (e.g. 2000, 2007), see Kron et al. (2019). Statement on flood generating mechanism in lines 30-32 is restricted to the UK conditions being examined (elsewhere, snowmelt runoff also occur, see Kron et al., 2019).
We have replaced (the word) rainfall by precipitation and added the reference with a remark.
When discussing flood risk reduction, in lines 32-36, see Driessen et al. (2018) and Kundzewicz et al. (2018).
Thank you: we chose the one reference we deemed most relevant, i.e., the one highlighting social dimensions, now even citing a statement therein.
Line 78 Replace: “river levels returning on average 1 : 10, 1 : 25, 1 : 100 or 1 : 200 years” by: “river levels with probability of exceedance in any one year being 1 : 10, 1 : 25, 1 : 100 or 1 : 200”
Done.
Lines 93-94: „the flood volume is a key parameter in the design of mitigation measures”. Yes, it is one of key parameters. Flood peaks stage is important for design of flood walls and levees. Flood duration also matters.
We agree and have discussed peaks, volumes in a paragraph of the introduction (lines 77-107 in the original text). Also note that flood-defence walls raise hT, such that peak levels and volumes communicate with another, also in our approach, cf. section 4. We also clearly indicate(d) that FEV is a complementary concept. See also the second bullet point in section 5 and, of course, the graphs in Figure 10, in which the higher-wall portions are achieved by raising hT closer or totally to the peak level corresponding with a peak discharge. Incidentally, we have removed the phrase while tightening the text, without loss of context.
How about flood hazards projections for the future? Climate track is mentioned in this paper via references to Hodgkins et al. (2017), IPCC (2013), Pacala and Socolow (2004). However, even if projections are uncertain (cf. Kundzewicz et al. 2017), flood hazard is likely to increase in much of Europe, including the UK. Hence, by referring to projections, the usefulness of the proposed approach can be strengthened.
We agree and did allude to uncertainty quantification using a more ensemble flood-prediction approach both in sections 4.1.1 (the roadmap) and the discussion in section 5 (the bullet points). We prefer to leave it at that for now not only because doing so aligns with the remarks of reviewer 3 to “make this manuscript clearly and concisely” but also because it will comprise future research and dissemination.
Lines 556-557: “Particularly useful is to address the following question: what fraction of the chosen FEV, i.e., Ve(hT), is reduced by a particular flood-mitigation measure or policy?” It can be difficult, if at all possible, to answer this question, even in an approximate way.
The roadmap discussed in section 4.1.1. implicitly includes a way to (start to) assess this question: by juxtaposing simulations (or an ensemble thereof) for a given level of flood protection (e.g. using return-period design floods & rainfall), for different scenarios with a varying suite of flood-mitigation measures, one can systematically assess the relevant fraction contributed to each flood-mitigation measure. Even though it is not always (cleanly) possible to distinguish these contributions from scenario to scenario and measure to measure, one would in doing so get an approximate sense of the contribution within a given square-lake target of a particular flood-mitigation measure, and one can and must also include the uncertainty therein, as we have illustrated. This discussion, albeit it very relevant and interesting, currently goes beyond the topics in this particular manuscript, so we leave it for future work, also because kernels for discussion can be found in the presented work, as acknowledged by the reviewer. Furthermore, note that we have added clarifying tables, one at the start of section 4 and one later in section 4.
Additional references for consideration:
Driessen, P. J. et al. (2018) Governance Strategies for Improving Flood Resilience in the Face of Climate Change. Water 10(11): #1595
Kron, W. et al. (2019) Reduction of flood risk in Europe - Reflections from a reinsurance perspective. Journal of Hydrology 576: 197-209.
Kundzewicz, Z.W. et al. (2017) Differences in flood hazard projections in Europe - their causes and consequences for decision making. Hydrological Sciences Journal. 62(1), 1-14.
Kundzewicz, Z. W., et al. (2018) Flood risk reduction: structural measures and diverse strategies. Proc. Nat. Acad. Sci. USA (PNAS) 115(49): 12321–12325, www.pnas.org/cgi/doi/10.1073/pnas.1818227115
Note that two of these references (Kron et al. 2019 and Kundzewicz et al. 2018) have now been included in the revised manuscript.
Reviewer 2 Report
I think this article has all the premises for publication
Author Response
Responses to Reviewer 2
I think this article has all the premises for publication.
We thank the reviewer for this support. No further comments required but we do refer to the response to the other two reviewers.
Reviewer 3 Report
General questions
The title of this manuscript, ‘A cost-effective protocol for flood-mitigation plans based on Leeds’ Boxing Day 2015 floods’, is interesting. However, this manuscript is more close to a technical report or manual than a scientific research manuscript. There are many unclear sentences and unnecessary details. To be an improved manuscript, authors should make this manuscript clearly and concisely. I strongly recommend authors should revise the manuscript not to include unnecessary details and some calculations which might be wrong.What is the cost-effectiveness protocol of flood-mitigation plans? Can you briefly explain?
A citation order in the text is little awkward. Typically, a citation number starts from [1] but here in the manuscript [32] is the first citation in the text.
Specific questions
(Line 31-32)Please rewrite the two sentences below.
‘intense substantial large-scale rainfall typical of winter’ and ‘intense but localized rainfall typical of summer’
(Line 73-75)
What bring you to write in italics?
(Figure 4)
I am not sure why you combined the three panels in Figure 4. Even a starting point of the graph is different in each panel. Also it is not comfortable to read the graphs.
(Line 6)
Can you show the derivation of the equation (9) from the equation (2)?
(Figure 8)
This figure is very low resolution, so you might want to replace this with a higher resolution.
(Line 396-402)
The significant digits are inconsistent such as 0.7Mm3, 1.5Mm3, 2.355Mm3 and 9.34. Please revise the significant digits considering consistency over the manuscript.
(Line 414, 417, 420)
Your calculation is ‘10M / 7.38% = 1.35M’. Don’t you think it is right? 7.38% is 0.0738, so the calculation should be ‘10M / 7.38% = 135.5M’.
14M / 11% is not 1.27M but 127.27M.
24M / 14% is not 1.51M but 171.43M.
Please double-check all calculations over the manuscript and revise them if needed.
Author Response
Responses to Reviewer 3
We thank the reviewer for the constructive review. More substantive changes in the revised manuscript have been highlighted in red, while smaller ones pertaining to tightening of the text have not.
The title of this manuscript, ‘A cost-effective protocol for flood-mitigation plans based on Leeds’ Boxing Day 2015 floods’, is interesting. However, this manuscript is more close to a technical report or manual than a scientific research manuscript.
Our manuscript concerns evidence-synthesis analysis/diagnostic and a roadmap/protocol for improving flood-mitigation design. Also, it is important to highlight, however, that the roadmap we have introduced in section 4.1.1. is novel in the sense that such a systematic approach to flood-mitigation design simulations is generally uncommon or not followed in current government and consulting-engineering practice in flood protection and design, despite the fact that this involves multi-million pound investments (as far as our experience goes within the UK), even though such a systematic approach may be more common in academia, where in general less, if any, money is at stake. Moreover, this type of evidence-synthesis paper is rather unusual in that it accommodates expertise from a multidisciplinary (and international) team of co-authors. We therefore prefer the wording protocol/roadmap rather than manual, while technical report is not an appropriate qualification given the social dimensions and dissemination purposes of our work.
There are many unclear sentences and unnecessary details.
Thank you. We have both addressed the specific comments below and reviewed and where necessary revised the text. However, without further specific comments, this generic remark made by the reviewer is difficult to deal with on a precise basis, not least because all other reviewers gave positive assessments on the clarity of the text. The abstract and introduction, sections 2, 3, 4, and 5 plus Appendices have been tightened where possible and in line with the remarks of the other reviewers. In doing so, we have reduced the length of the manuscript by over 5%, subsequently used to add two new and clarifying tables. We suspected that section 4 where the Leeds protection schemes, planned and updated, were presented in length could be considered “unnecessary details” by Reviewer 3. We would prefer keep these details to clearly demonstrate how an interested reader could reuse our framework on another case-study. To let him/her know what type of information and computations he/she would need. However, we agree that some readers could be more interested by the concept and general framework than by the details of our case-study. To help these readers getting the essence of the improvement without knowing all details and generally speaking to enhance clarity, as requested by the reviewer, we have added two new and clarifying tables at in section 4. Table 2 provides in a synthetic way what measures we are adding and with which objectives.
To be an improved manuscript, authors should make this manuscript clearly and concisely. I strongly recommend authors should revise the manuscript not to include unnecessary details and some calculations which might be wrong.
Thank you for the remark. It is unclear to which details and calculations the reviewer refers beyond the specific questions below, which we did address in detail. Motivated by this comment, we have had a fresh look at our manuscript and have as a result tightened the formulation in several places. Please, also see the response to the previous remark.
What is the cost-effectiveness protocol of flood-mitigation plans? Can you briefly explain?
Our cost-effectiveness protocol of flood-mitigation plans is a graphical analysis based on the square-lake representation of the excess flood waters that caused the (real or tentative) flooding, used as a pie/square/lake chart with each piece of the lake representing a flood-mitigation measure and a representation of its costs and uncertainty; several square-lake charts can represent different flood-mitigation scenarios, thereby facilitating quick and direct comparison. Such a graphical scenario presentation can either be done a priori to stimulate stakeholder and expert discussion or a posteriori, as a diagnostic to summarise detailed hydraulic/hydrological simulations, cf. the roadmap presented in section 4.1.1 and the discussion in section 5.
A citation order in the text is little awkward. Typically, a citation number starts from [1] but here in the manuscript [32] is the first citation in the text.
We used the MDPI standard style file, which did not work, so in the end we ordered the references by hand.
Specific questions
(Line 31-32) Please rewrite the two sentences below. ‘intense substantial large-scale rainfall typical of winter’ and ‘intense but localized rainfall typical of summer’
Done. The sentence has been changed and clarified.
(Line 73-75) What bring you to write in italics?
All citations are highlighted by quotes as well as being italicised to clearly distinguish them from the main text.
(Figure 4) I am not sure why you combined the three panels in Figure 4. Even a starting point of the graph is different in each panel. Also it is not comfortable to read the graphs.
The three-panel combination offers a unified graphical connection of the actually measured data (bottom-left panel), the rating curve (top-left panel) and hydrograph (top-right panel); such a graphical connection is less clear in the standard format of three graphs below each other. We add that our reconstruction has been generally well-received in related work, with the reviewer being the first person to comment on the perceived lack of clarify in our reconstruction. As a result of this point, we revisited the graphs and made some minor changes to improve clarity, around the origin.
Line 6) Can you show the derivation of the equation (9) from the equation (2)?
This is not possible since (9) is simply the numerical evaluation of the sum of Nm~128 terms in (2), in which the value of Nm accrues from quarter-hourly measurements taken over 32 hours—the flood-excess duration with Q>QT—at the Armley gauge station. As such, (9) has no derivation per se. We add that Nm, QT, the data from the river level (q.v. Fig.4) and discharge data files were obtained from the EA, and are available in the public domain.
Figure 8) This figure is very low resolution, so you might want to replace this with a higher resolution.
We have done this, so that the factual jpeg is now clear.
(Line 396-402) The significant digits are inconsistent such as 0.7Mm3, 1.5Mm3, 2.355Mm3 and 9.34. Please revise the significant digits considering consistency over the manuscript.
We have followed your suggestion and have also made some modifications.
Line 414, 417, 420) Your calculation is ‘10M / 7.38% = 1.35M’. Don’t you think it is right? 7.38% is 0.0738, so the calculation should be ‘10M / 7.38% = 135.5M’.14M / 11% is not 1.27M but 127.27M. 24M / 14% is not 1.51M but 171.43M.
This is a misunderstanding from the reviewer as the original text read: “GBP10M/7.38%=GBP1.35M per one percent of flood protection”, so the reviewer can only have missed the wording “per one percent of flood protection”, throughout. However, we have clarified it to “GBP10M/7%=GBP1.43M/1% (so GBP1.43M per one percent of flood protection)” (or just as 1.43M/% such as found in the original and current square-lake graphs), here and in the other cases.
Please double-check all calculations over the manuscript and revise them if needed.
Thank you for the remark. We have checked our calculations and made some modifications, but also note that the remark made above was based on a misunderstanding rather than an error on our part.
Round 2
Reviewer 3 Report
The manuscript is now ready for the publication because authors took care of the reviewer's comments well.
