Flooding Related Consequences of Climate Change on Canadian Cities and Flow Regulation Infrastructure

Round  1

Reviewer 1 Report

The work presented contains interesting topics, its abstract is accurate and informative including the study's purpose, main findings, and main conclusions; the interpretations and conclusions are justified by the results; the subject matter is within the scope of the journal. In my opinion some improvements are needed; in particular there are some unclear points in the paper.

The paper uses runoff projections from 21 GCMs for future change estimation, in order to obtain flood characteristics across Canada under four different emission scenarios; have the authors tested the reliability of the GCM projections?

Page 3, section 2 the authors should better explain the difference between the two investigated approaches: ‘robust GCM median’ and ‘all GCM median’;

Page 4, The CaMa-Flood model should be described in the paper;

Page 4, section 2.1.2, it is not clear what is the study area where the retrospective analysis was conducted; is it conducted on all the 100 most populated cities described in table 1?

The table at page 3 should be “table 1” and the table at page 5 should be “table 2”; so at page 5 - line 184 and at page 7 - line 192, should be “table 2” and at page 8 - line 199 should be “table 1”;

Page 4, line160, the authors should explain how the water levels are estimated;

Page 7, the sentence at lines 191-192 is repeated at page 5 lines 183-184;

Page 8, the authors should better explain how the projected future return periods of historical 100-year and 250-year return are estimated;

Page 9, line 237 is not clear, please rephrase it;

Page 11, in the table 5, it is not clear the information reported within the bracket next to the province; in the caption is explained that it refers to the projected future flood frequency magnitude in years, so is it the projected future return period?  

In the paper some further evaluations should be achieved in the context of the “Revisiting the Concepts of Return Period” considering the non-stationay analysis (e.g. Salas 2014).

The figures 2 is nor clear: the authors should improve the scale of representation; the same for the figures 4, 5, 6 and 7;

The introduction should be improved, including references to recent papers useful to introduce the proposed work in the context of the recent scientific research in this field; in my opinion the authors may introduce references regarding models for flood prediction and flood frequency analysis (Rossi et al., 1984, Blazkova and Beven 1997, Fiorentino et al., 2011, Gioia et al., 2014).

References

Blazkova S., Beven K., Flood frequency prediction for data limited catchments in the Czech Republic using a stochastic rainfall model and TOPMODEL. Journal of Hydrology 195 (1997) 256-278.

Fiorentino M., Gioia A., Iacobellis V. and Manfreda S. Regional analysis of runoff thresholds behaviour in Southern Italy based on theoretically derived distributions, Advances in Geosciences, ISSN 1680-7340, 26, 139–144, 2011-    www.adv-geosci.net/26/139/2011/, doi:10.5194/adgeo-26-139-2011.

Gioia, A., Manfreda, S., Iacobellis, V., Fiorentino, M. Performance of a theoretical model for the description of water balance and runoff dynamics in Southern Italy. Journal of Hydrologic Engineering 19(6), pp. 1113-1123, 2014.

Rossi F., Fiorentino M., Versace P. Two-component Extreme Value Distribution for Flood Frequency Analysis, WATER RESOURCES RESEARCH, VOL. 20, NO. 7, PAGES 847-856, JULY 1984

Salas, Jose D. and Jayantha Obeysekera. "Revisiting the Concepts of Return Period and Risk for Nonstationary Hydrologic Extreme Events." Journal of Hydrologic Engineering (2014): 554-568.

Author Response

Thank you for your feedback. Following describes how we have addressed your comments in the revised manuscript.

The work presented contains interesting topics, its abstract is accurate and informative including the study's purpose, main findings, and main conclusions; the interpretations and conclusions are justified by the results; the subject matter is within the scope of the journal. In my opinion some improvements are needed; in particular there are some unclear points in the paper.

Thank you for your encouraging comments. Please find below our responses to your queries, and details on how your comments have been addressed in the revised version of the paper.

The paper uses runoff projections from 21 GCMs for future change estimation, in order to obtain flood characteristics across Canada under four different emission scenarios; have the authors tested the reliability of the GCM projections?

Thank you for your comment. In this study, a multi-model ensemble approach has been adopted, and future projections of runoff from all GCMs have been assumed to be equally plausible in the future. We do acknowledge that studies specifically focussing on evaluating the reliability of GCM based projections have been performed, but this subject is identified as outside the scope of this study. This has now been more clearly documented in the lines 421-426 of the revised paper.   

Page 3, section 2 the authors should better explain the difference between the two investigated approaches: ‘robust GCM median’ and ‘all GCM median’;

Thank you for your comment. The text explaining the difference has been revised so that the difference between the two approaches is clearer to the readers. The revised text is available in the lines 129-138 of the revised paper.

Page 4, The CaMa-Flood model should be described in the paper;

Thank you for your comment. A description of the model has been added in the lines 143-164 of the revised paper.

Page 4, section 2.1.2, it is not clear what is the study area where the retrospective analysis was conducted; is it conducted on all the 100 most populated cities described in table 1?

Thank you for your comment. The retrospective simulation was performed for the entire Canadian landmass at 0.005º spatial resolution and does include the 100 most populated city locations summarized in Table 1. This has been discussed in the line 194 of the revised paper.

The table at page 3 should be “table 1” and the table at page 5 should be “table 2”; so at page 5 - line 184 and at page 7 - line 192, should be “table 2” and at page 8 - line 199 should be “table 1”;

Thanks for pointing this out. We have made necessary corrections so that the Table 1 and Table 2 are correctly placed in the text and correctly referenced.

Page 4, line160, the authors should explain how the water levels are estimated;

Thank you for your comment. Daily water levels are estimated in the CaMa-Flood model using the water storage, and floodplain elevation profile for each unit catchment. A description on CaMa-Flood model has been included in the lines 143-164 of the revised paper for providing clarity in this regard.

Page 7, the sentence at lines 191-192 is repeated at page 5 lines 183-184;

Thank you for pointing this out. The repeated sentences have been removed in the revised paper.

Page 8, the authors should better explain how the projected future return periods of historical 100-year and 250-year return are estimated;

An explanation on the calculation of future return periods of historical 100 and 250 year return period events is provided in the lines 152-160 of the revised paper.

Page 9, line 237 is not clear, please rephrase it;

The sentence has been rephrased and the revised sentence is provided in lines 172-180 of the revised paper.

Page 11, in the table 5, it is not clear the information reported within the bracket next to the province; in the caption is explained that it refers to the projected future flood frequency magnitude in years, so is it the projected future return period? 

Yes, the information provided within the brackets comprises of the province where the city is located, and the future frequencies of historical 100-year return period floods projected for the city. The caption has been revised so that it is able to more clearly convey this.

In the paper some further evaluations should be achieved in the context of the “Revisiting the Concepts of Return Period” considering the non-stationay analysis (e.g. Salas 2014).

Thank you for your insightful comment. The analysis presented in this work builds on the results from Gaur et al. (2018) and evaluates non-stationarity in flood characteristics between historical and future time-periods but non-stationarity within either of these time-periods has not been accounted for as advised in Salas and Obeyskera (2014). We acknowledge that more work can be performed in this direction in the future in line 426-430 of the revised paper.

The figures 2 is nor clear: the authors should improve the scale of representation; the same for the figures 4, 5, 6 and 7;

We have tried our best to revise the figures based on your comments. Thank you.

The introduction should be improved, including references to recent papers useful to introduce the proposed work in the context of the recent scientific research in this field; in my opinion the authors may introduce references regarding models for flood prediction and flood frequency analysis (Rossi et al., 1984, Blazkova and Beven 1997, Fiorentino et al., 2011, Gioia et al., 2014).

The introduction has been improved by including a discussion on flood modelling and flood frequency analysis in the lines 58-70 of the revised paper.

References

Blazkova S., Beven K., Flood frequency prediction for data limited catchments in the Czech Republic using a stochastic rainfall model and TOPMODEL. Journal of Hydrology 195 (1997) 256-278.

Fiorentino M., Gioia A., Iacobellis V. and Manfreda S. Regional analysis of runoff thresholds behaviour in Southern Italy based on theoretically derived distributions, Advances in Geosciences, ISSN 1680-7340, 26, 139–144, 2011-    www.adv-geosci.net/26/139/2011/, doi:10.5194/adgeo-26-139-2011.

Gioia, A., Manfreda, S., Iacobellis, V., Fiorentino, M. Performance of a theoretical model for the description of water balance and runoff dynamics in Southern Italy. Journal of Hydrologic Engineering 19(6), pp. 1113-1123, 2014.

Rossi F., Fiorentino M., Versace P. Two-component Extreme Value Distribution for Flood Frequency Analysis, WATER RESOURCES RESEARCH, VOL. 20, NO. 7, PAGES 847-856, JULY 1984

Salas, Jose D. and Jayantha Obeysekera. "Revisiting the Concepts of Return Period and Risk for Nonstationary Hydrologic Extreme Events." Journal of Hydrologic Engineering (2014): 554-568.

Gaur, A., Gaur, A. and Simonovic, S.P. (2018). Future Changes in Flood Hazards across Canada under a Changing Climate. Water, 10, 1441, doi: 10.3390/w10101441.

Reviewer 2 Report

This manuscript can be published as it is

Author Response

This manuscript can be published as it is.

Thank you very much for your positive recommendation.

Reviewer 3 Report

The state of the art needs to be reformulated. Some references are outdated and some information is missing.
the abstract need to be reviewed.
Too many tables are given. The information needs to be summarized.
the quality of the figures needs to be checked.
However, my major concern is related to the novelty and the differences between this work and the work of Gaur et al (2018). The authors need to justify the novelty and highlight the added value of this work related to the previous work.

Some additional considerations are given in the pdf file.

Comments for author File: Comments.pdf

Author Response

Thank you for your feedback. Following is a description of how we have addressed your comments in the revised manuscript.

The state of the art needs to be reformulated. Some references are outdated and some information is missing.

Thank you for your comment. The introduction has been reviewed, and revised. Reviews of some more recent references have been added and discussion on them has been expanded.

the abstract need to be reviewed.

The abstract has been reviewed and revised based on the comments provided in the pdf file.

Too many tables are given. The information needs to be summarized.

One of the tables (Table 1) providing the list of cities can be removed however we felt that it would provide the readers with more readily accessible information on the cities that have been selected for analysis if the Table was present in the paper. For this reason, we have chosen not to omit this table from the paper however if yourself and the editor find this necessary, please advise and we can remove this table (or provide it as a supplementary information).

We believe that other tables are necessary to showcase clearly our findings.  

the quality of the figures needs to be checked.

All the figures have been updated for this aspect in the revised paper. Please advise if further improvements are needed.

However, my major concern is related to the novelty and the differences between this work and the work of Gaur et al (2018). The authors need to justify the novelty and highlight the added value of this work related to the previous work.

Thank you for your insightful remark. This study does build on the results generated in Gaur et al. (2018) but presents valuable information and analysis that we believe will be useful for water resource managers and policy-makers in Canada to identify cities and flow regulation infrastructure (FRIs) that can be expected to be critically influenced due to climate change.

While spatially explicit maps of projected changes in flooding frequencies and flood timing as generated in Gaur et al. (2018) are useful to compare projected changes in flood-hazard from one region of Canada to another, how is that hazard translating to risk when overlaid with the distribution of the cities, and combined with the total flood-exposed population is one of the interesting aspects that has been explored in this paper.

This study answers question like these for Canadian water resource managers: which highly populated city in Canada can be expected to experience most increases in flood risk (flood hazard) in the future? What would be the magnitude of this increase? Similarly, what can be the extent of change in flood timing at the FRIs? Which FRIs can be expected to be worst hit by climate change? and so on.. We believe that the discussion and analysis presented in this study provides synthesized information that policymakers and readers can use to readily identify cities and FRIs highest at risk due to climate change influences. We have highlighted some of the novelty of this work in lines 107-113 of the revised paper.         

Some additional considerations are given in the pdf file.

Nearly all of the recommendations made in the pdf file have been incorporated in the revised version of the paper. If there are any major outstanding issues, do advice and we will address them. Thank you for providing your valuable feedback. 

Round  2

Reviewer 1 Report

page 2, line 73 is "Fiorentino et al., 2011";

the reference 'Gioia et al., 2014' reported at page 2, line 74, should be introduced in the references;

Author Response

Thank you for your feedback. Following describes how we have addressed your comments in the revised manuscript.

page 2, line 73 is "Fiorentino et al., 2011";

Above reference has been corrected and the correct citation can be found at line 64 of the revised paper.

the reference 'Gioia et al., 2014' reported at page 2, line 74, should be introduced in the references;

Above reference has been added to the list of references.

Reviewer 3 Report

The authors have improved the paper according to the reviewer’s suggestions. However, some corrections need to be done:

1. The introduction was updated, however, several references still outdated:

“Blazkova and Beven (1997) 71 used a stochastic rainfall simulator and TOPMODEL (Beven et al. 1984) to make reliable estimates of 72 flooding frequencies in three data sparse catchments in Czech Republic.””

“Rossi et al. (1984) used a two-component extreme value distribution to perform flood 76 frequency analysis in 39 Italian basins.”

Please make a critical reflection about the importance of these references.

2. L100-120: where these studies were conducted?

3.The differences between this study and the study performed by Gaur et al. (2018) was clarified. However, the innovation of this work in terms of methodology needs to be clarified.

4. L187: “been validated extensively for…” where?

5. L200-210: check the format

6. I think Table 1 is not necessary. Could be deleted or reduced (most 10 cities?)

7. The scale of Figure 1 is not readable.

Author Response

Thank you for your feedback. Following is a description of how we have addressed your comments in the revised manuscript.

The authors have improved the paper according to the reviewer’s suggestions. However, some corrections need to be done:

1. The introduction was updated, however, several references still outdated:

“Blazkova and Beven (1997) 71 used a stochastic rainfall simulator and TOPMODEL (Beven et al. 1984) to make reliable estimates of 72 flooding frequencies in three data sparse catchments in Czech Republic.””

“Rossi et al. (1984) used a two-component extreme value distribution to perform flood 76 frequency analysis in 39 Italian basins.”

Please make a critical reflection about the importance of these references.

We agree that both the above references are outdated. They were removed from the revised manuscript. A review of few more recent papers have added in the lines 65-71 of the revised paper.

2. L100-120: where these studies were conducted?

Information on the study area of the studies has been included in the lines 100-114 of the revised paper.

3.The differences between this study and the study performed by Gaur et al. (2018) was clarified. However, the innovation of this work in terms of methodology needs to be clarified.

The text provided in the lines 115-124 of the revised paper has been modified to explicitly mention the innovation of this work in terms of the methodology adopted.

4. L187: “been validated extensively for…” where?

Information on the catchments where model validation has been performed in the past has been added in the lines 170-171 of the revised paper.

5. L200-210: check the format

The formatting of these lines has been improved.

6. I think Table 1 is not necessary. Could be deleted or reduced (most 10 cities?)

Table 1 has been removed from the revised paper.

7. The scale of Figure 1 is not readable.

The Figure 1 has been revised so that the scale is readable.

Thank you very much for your insightful comments. We have tried our best to address them. Please do advise if we missed anything.