A Cost Efficiency Analysis of Flood Proofing Measures for Hydraulic Risk Mitigation in an Urbanized Riverine Area
Round 1
Reviewer 1 Report
Comments on the paper:
Flood proofing measures cost-efficiency analysis for hydraulic risk mitigation in an urbanized riverine area.
Page 4, row 122
“Several points contributed to the flooding…”
I suggest to use: Several low land areas, or low banks of the river, or discontinuities in the hydraulic works, or several locations
A point cannot generate flooding unless it has an infinite velocity of the water.
Page 6, row 214
“ ….. by trying to optimize the reduction of the extension of the flooded area and the water depth in urban and agricultural areas.”
Urban and agricultural areas have the same degree of protection ? Flooding the agricutural areas has a benefic effect on the magnitude of the flood in urban areas, meaning less costs for their protection
Page 6, row 216
“…the frequency curve of maximum flow rates has been obtained…”
What statistical distribution was used ? Why for return periods in the range 50-500 years the curve has a zig-zag shape ?
Page 6, Fig. 4
May be, for uniformity puposes (see Fig. 10 – damages versus probability of exceedance), it would be good to represent maximum discharges versus probability of exceedance, too.
Page 7, Table 1
My suggstion is to present the values of the maximum discharges as round numbers. For instance Q5% = 250 m3/s (in fact it is not possible to measure a discharge like 249.5 m3/s with such a high accuracy).
Page 7, rows 229-231
“….for both regions a calibrated roughness coefficient was considered (0.028 m1/3/s for the river and 0.050 m1/3/s for the floodplain area).”
The roughness coeffcient is the same for agricultural areas and for urban areas ?
To discuss the model calibration. RMSE = ?; Nash-Sutcliffe coefficient =?
Page 7, rows 233-234
“…..the scenarios simulated for different flood proofing measures expected for a return period of 100 years.”
The simulations were run in steady state ? In this case, the effect of the flood (water level) is over-evaluated. If the simulations were in unsteady state, how the flood hydrograph looks like?
Please, make some comments on these issues.
Page 8, rows 257-258
“…it was therefore appropriate to perform an average of the water depth for the flooded areas…”
The damage functions put into evidence 3 classes of water depth: 0-1m; 1-4 m and >4m. For each class an average value of the water depth can be considered. A unique average value of the water depth for the whole flooded area is a rough approximation of the damages on agricultural areas. A simple GIS processing could be used to identify these classes.
What is the flooded area used for agricultural purposes ? What is the total damage for 1% probability of exceedance of the maximum discharge ? Is this damage of the same order of magnitude as the damages in urban areas or much lower ?
Page 10, row 292
“…This breach allows flood flows to inundate the eastern side of the river, resulting in damages…”
Please, indicate the North on the Figure 9. A Legend would be necessary (there are dark green circes and smaller light green circles).
Page 10, rows 298-300
“For the representation of the relationship Damage - Exceeding Probability, before the mitigations measures (Figure 10), the following steps have been carried out:
- Simulations with MLFP-2D model of inundation map for a return period of 100 years …”
There is a contradiction betewen the graph Damage-Exceeedance probability (which is defined for the following probabilities of exceedance of the maximum annual discharge: 0.5%; 1%; 1.5% and 2%) and the affirmation that the simulations were performed for a return period of 100 years (leading to only one value of the damages in Fig. 10).
Page 11, row 308:
“…of investment costs, the reduction of flood damage following the intervention and the total cost…”
In fact, the expected flood damage is represented in Fig. 11 (and not the reduction of the flood damage, which would be a derivative of the flood damage to 1 m of levee length).
Page 12, Figure 12 and 13
In Fig. 12 a combination of both types of flood barriers (250 cm and 150 cm) is presented. However, the cost-benefit analysis is made separately for the barrier of 250 cm and 150 cm respectively.
On the other hand, according to Figure 13 the total cost for flood barrier of 250 cm is always higher than the cost for the flood barrier of 150 cm. Any combined solution (partly a barrier of 250 cm and partly of 150 cm) will have the minimum between the two yellow curves. That means that the solution of a barrier of 150 cm is always better than any combined solution. The Fig. 12 should be corrected accordingly.
At the same time, the sentence from Page 13, rows 362-365 shouls also be revised:
“So in conclusion, the ideal concept would be a objective function of the two heights of barriers, focusing on the critical aspects of both, in order to make up for them, through the mix of the same and taking advantage of the benefits that they offer.”
Page 13, row 357:
“…..analysis show that it is not possible to look for the minimum of the function for the following case.”
In fact, a minimum on Fig. 11, can be identified: it is about 750.000 EUR (no concrete levee, which can be considered option zero: doing nothing).
On the other hand, if adding the total cost for putting into operation a flood barrier of 150 cm (which is about 350.000 EUR) and the total cost when using door-window barriers (approximately 600.000 EUR) one obtains 950.000 EUR > 750.000 EUR (option zero, doing nothing). To justify the preference for flood barriers and flood doors/windows something else should be added (like the stress of the population, the disruption of the economic and social life etc.).
Author Response
Answers in the attached pdf
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This paper deal with the evaluation of flood proofing measures by cost-efficiency analysis in urbanized riverine area. While I appreciate the effort of the work presented and the significance of flood proofing measures, I think the authors needs to improve the focus of the paper and provide more information on the methods used.
General comments
- Regarding flood control measures, the relationship between the countermeasure used in this study and the scale of floods (probability of occurrence) should be organized. Even if it is effective against a flood of a specific scale, is it a measure that is detrimental to floods of a larger scale? Evaluation of cost-efficiency of flood control measures for multiple scale floods should be conducted to decide the most cost-efficiency of flood control measures.
- Explanation of calculation method of cost is inadequate. How is total cost calculated? In addition, regarding the investment cost, the basis for calculating the unit cost of construction of the embankment should be explained.
- The manuscript should be reconstructed because it is confusing due to the mixing of the method, result, and discussion.
Specific comments
- Line 216-219
Method of hydrological calculation should be explained in the 3. Methodology. Especially, the method of sampling data (observation station, observation period) and calculation method of the return period should be described.
- Line 216-219
Calculation result from MLFP-2D should be evaluated by comparison with past inundation record.
- Line 223-234
Information of manning’s roughness and number of nodes should be described in the Methodology.
- Line 246-258
Calculation method of damage should be explained in the Methodology.
- Line 286-287 and Figure12
Rational of introducing position or length of flood barrier should be provided using the differences in the calculation result depending on the presence or absence of these structures.
Author Response
Answers in the attached pdf
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Comments on the paper:
Flood proofing measures cost-efficiency analysis for hydraulic risk mitigation in an urbanized riverine area.
The paper was seriously improved. However, there are some small issues to be solved.
Page 6, rows 194-195 and rows 199-201
“Regarding the Manning roughness coefficient, two calibrated values were considered, i.e. 0.028 m1/3/s for the river and 0.050 m1/3/s for the floodplain area, to be used in all simulations”
“Friction values, in terms of Manning’s n were randomly and uniformly distributed between 0.035 m1/3s-1 and 0.1 m1/3s-1 for the river, and 0.045 m1/3s-1 and 0.2 m1/3s-1 for the floodplain.”
The lower limit for randomly distributed Mannig roughness coefficient is 0.035, while the calibrated value is 0.028. Probably, the lower limit is 0.025 for the river instead of 0.035.
Page 6, Fig. 4
It is very unusual to represent the discharge values of the flood frequency curve on the Ox axis. In my previous review I noticed that the curve is not smooth as it usually is. Please, use the representation from the version 1 of the paper.
Page 7, Table 1
Please, give up the decimales. There are a lot of epistemic and aleatory uncertainties to pretend you are able to evaluate so accurately the maximum discharges. More than that, the numbers in the Table are slightly different if compared to the same discharges in the version 1.
Page 11, rows 315-318
“The new levee is supposed to be built with an extension of about 1300 m located on both sides of the stream to the mouth of the river, 550 m on the hydraulic right immediately downstream the railway line, in which there is, about 350 m further, the first village (Bastione). The remainder, about 600 m is in the left riverside where the second village (Caldà) is located (Figure 12).”
If we add 550+350+600 we obtain 1500 (>1300).
Page 15, rows 395-398
“So in conclusion, the ideal concept would be an objective function of the two heights of barriers, focusing on the critical aspects of both, in order to make up for them, through the mix of the same and taking advantage of the benefits that they offer.”
If the Objective function is min (Damages + Costs) the only optimal solution is based on barrier heights of 150 cm. The minimum is about 0.375 M EUR for about 250 m length of the 150 cm flood barriers, while the minimum for the 250 cm flood barrier is about 0.45 M EUR for almost the same length. Any mix solution will lead to a higer price than in the case of adoption of 150 cm barrier height.
Author Response
rev1
Author Response File: Author Response.pdf
Reviewer 2 Report
The manuscript has been much improved and is in a nice condition now. I think this manuscript will be acceptable.
Author Response
rev2
Round 3
Reviewer 1 Report
No comments or suggestions this time.
Best regards,
