Assessing the contribution of extratropical cyclones to river floods that caused property damage in Quebec, Canada

Gagnon, Clarence; Nadeau, Daniel F.; Di Luca, Alejandro; Brault, Benoit; Hamon, Romane; Roy, Nicolas L.; Bourgault, Marc-André; Anctil, François

doi:10.5194/egusphere-2025-6192

Preprints

https://doi.org/10.5194/egusphere-2025-6192

Preprints

02 Feb 2026

| 02 Feb 2026

Assessing the contribution of extratropical cyclones to river floods that caused property damage in Quebec, Canada

Clarence Gagnon, Daniel F. Nadeau, Alejandro Di Luca, Benoit Brault, Romane Hamon, Nicolas L. Roy, Marc-André Bourgault, and François Anctil

Abstract. In the past few decades, the province of Quebec in eastern Canada has experienced widespread and costly flood events, yet the contribution of extratropical cyclones to these floods remains unquantified. This paper presents a methodology for analysing the contribution of extratropical cyclones (ETCs) to 498 fluvial floods that occurred in watersheds in southern Quebec, Canada, between 1991 and 2020, and highlights key characteristics shared by the 200 flood ETCs (storms that contributed to flooding) identified in the study. This analysis combined reconstituted river outputs, government financial aid claims following flooding, ETC tracks, and precipitation data to identify flood events and their relevant ETCs. ETC contribution was defined as the percentage of rainfall during a search window surrounding each flood event that was associated with the relevant ETCs. Most of these floods occurred in the spring (74.7 %). The majority (72.7 %) of flood events had a high (50–75 %) to very high (>75 %) ETC contribution, and only 2.6 % of events (mostly occurring in summer) had a negligible (< 5 %) ETC contribution. Flood ETCs had larger percentages of tracks originating from the Central US & Gulf of Mexico (+30.3 %) and lower percentages of tracks originating from Quebec & Maritimes (–16.7 %) and Western Canada & Pacific Ocean (–12.8 %) compared to non-flood ETCs (storms that did not contribute to flooding but spent at least one hour inside the Baseline Southern Quebec domain, N = 6,027). Flood ETCs also spent on average 18.2 h more in the domain and generally tracked over southern Quebec as opposed to over the Atlantic Ocean along the east coast of North America or over northern Quebec like non-flood ETCs. The percentages of financial aid claims allotted to flood ETCs were highly variable. Just five of the 200 flood ETCs were associated with 51.0 % of all financial aid claims filed during the 30-year period, indicating that a high to very high ETC contribution rarely resulted in widespread flooding. The method developed in this study could be applied to other regions or types of storms analysed (e.g. tropical cyclones and mesoscale convective systems) to further our understanding of the atmospheric causes of flooding and ultimately enhance flood preparedness.

Received: 12 Dec 2025 – Discussion started: 02 Feb 2026

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Preprint (PDF, 16772 KB)

Supplement (17886 KB)

Download & links

Preprint (16772 KB)
Metadata XML
Supplement (17886 KB)
BibTeX
EndNote

Clarence Gagnon, Daniel F. Nadeau, Alejandro Di Luca, Benoit Brault, Romane Hamon, Nicolas L. Roy, Marc-André Bourgault, and François Anctil

Status: final response (author comments only)

RC1:
'Comment on egusphere-2025-6192', Anonymous Referee #1, 23 Mar 2026
Title: Assessing the contribution of extratropical cyclones to river floods that caused property damage in Quebec, Canada
Authors: Clarence Gagnon, Daniel F. Nadeau, Alejandro Di Luca, Benoit Brault, Romane Hamon, Nicolas L. Roy, Marc-André Bourgault, François Anctil
Submitted to Hydrology and Earth System Sciences
Ref.: egusphere-2025-6192
Overview: This paper describes as assessment of the contribution of extra-tropical cyclones (ETCs) that contributed to floods and insured damages across populated watersheds of Quebec, Canada during 1991-2020. The study makes use of various datasets including ETC storm tracks, European Centre for Medium-Range Weather Forecasts reanalysis (ERA5) precipitation data, streamflow data generated by the Hydrotel model, and provincial insurance claims for damages related to flood events. Over the 30-year study period, three-quarters of all floods occurred during spring with 72.7% being associated with the passage of ETCs in proximity of the flooded waterways. Storms originating from the Gulf of Mexico and central United States were particularly damaging with lesser impacts from ETCs originating elsewhere in North America. This effort suggests that ETCs play a central role in flooding within highly populated watersheds of Quebec particularly during spring in concurrence with snowmelt freshets.
The paper is well-written with clear figures illustrating key findings from the analyses. The following report provides guidance for some revisions that would improve the paper.
General Comments:
The Hydrotel model is applied to generate streamflow simulations to assess 2-year recurrence flood values across the watersheds of interest, many of which are ungauged. Many waterways in Quebec are regulated by hydropower infrastructure and as such it is unclear how Hydrotel implements regulation within its model structure to accurately simulate flood patterns in regulated waterways. Are there differences in the impacts of ETCs to floods and damage reports in regulated vs. unregulated waterways?

It is unclear why the ECWMF fifth generation reanalysis (ERA5) at the relatively coarse of resolution of 0.25° ×25° was selected for total precipitation and snowfall data instead of its land surface product, ERA5-Land, available at the finer resolution of 0.10° × 0.10°. Given the relatively small size of many watersheds of interest and the high spatial variability of the precipitation field, applying the ERA5-Land product would likely produce higher accuracy results.

It is stated on p. 16: “The majority of these flood ETCs were identified in spring, primarily due to the increased number of flood events during this season due to the contribution of snowmelt, as well as the longer search periods (averaging 24.0 days in spring, compared to 8.1, 6.8, and 6.6 days in summer, fall, and winter, respectively).” This would suggest that a dominant generating mechanism for floods in the study area is spring snowmelt exacerbated by precipitation ETCs. Was there an effort made to assess the relative contribution of direct rainfall vs. snowmelt to the floods experienced across Quebec from 1991 to 2020? Perhaps the passage of an ETC at this time only coincides with the rapid onset of snowmelt, yielding only a modest contribution of the passing storm (if any) to the flood event.

Furthermore, as the principal mechanism for flood generation in most watersheds of southern Quebec is the spring freshet, perhaps the role of ETCs is under-stated in terms of their impacts to floods. ETCs during winter mostly deliver snowfall that accumulates in the seasonal snowpack. If snow accumulation is unusually high due to a succession of multiple ETCs during winter and then melts rapidly, one transient ETC during the spring freshet may only exacerbate the flooding.

Specific / Technical Comments:
p 1, line 18: Spell out “US” and replace the ampersand with “and” here and elsewhere in the text when referring to regions of interest.

p 2, line 56: Revise to “traveled” for consistency with use elsewhere in the paper.

p 2, lines 59-60: Should this be “Ottawa River basin” or the province of Ontario?

p 3, lines 90-91: “data” usually is considered a plural noun, please revise verbs to “provide”, “allow” and “depend”.

p 6, line 129: There is also a large concentration of Quebec’s population in the Gatineau area along the Ottawa River.

p 7, line 137: Write as “The watershed delineations”.

p 7, line 143: Spell out “ECMWF” and “ERA5”.

p 7, line 152: Here, “ECMWF” is spelled out but this should be shifted to line 143 instead, and henceforth “ECMWF” used in the text.

p 8, Figure 1: The legend above panel a) refers to the “BSQ domain” yet the figure caption does not describe which area this is. Panel d) outlines the basin affected by floods but the basin is not identified in the caption aside from the “Montérégie region”.

p 9, line 174: Change to “a crude”.

p 10, line 207: Delete “In order” and begin the sentence with “To quantify”.

p 11, line 218: Revise to “data were”.

p 11, line 236: Instead of “mild”, consider using “low” throughout the paper.

p 14, line 294: Instead of “worst year”, perhaps state “most active year”.

p 16, Figure 6: Replace “mild” with “low”.

p 16, lines 335-336: Spell out numbers less than 10 here, so “eight events”, “two events” and “three events”.

p 19, line 375: Here and elsewhere, replace the ampersand with “and” when referring to regions of interest.

p 19, line 389: Use the singular “storm” here.

p 20, line 403: Insert “(BSQ)” after “Baseline Southern Quebec” given the abbreviation is used in one of the figure panels.

p 24, line 471: Replace “mild” with “low”.

p 25, line 502: Delete the comma after “et al.”

p 25, line 517: Spell out “NOAA” when first introduced in the text.

p 26, line 532: The “Conclusions” section should be numbered like all other sections in the paper.

p 27, line 578: Change to “are not available”.

p 29, lines 609-611: Update the status of this reference, perhaps it is now published?

p 30, line 661: Replace “Dat” with “Data”.
Citation: https://doi.org/10.5194/egusphere-2025-6192-RC1
RC2:
'Comment on egusphere-2025-6192', Anonymous Referee #2, 27 Mar 2026
This manuscript attempts to quantify the contribution of extratropical cyclones (ETCs) to damaging fluvial floods in southern Quebec from 1991–2020. The study combines reconstituted river outputs, government financial-aid claims, ERA5 precipitation, and the NAEC storm catalogue to identify 498 local flood events and characterize the ETCs associated with them. The paper’s main findings are that most flood events occurred in spring, that most events had high to very high ETC contribution, and that flood ETCs differed from non-flood ETCs in their origins, lifetimes, and tracks.

Overall, I find the topic important and the dataset integration genuinely promising. The manuscript addresses a meaningful gap, particularly because it attempts to move beyond a few well-known case studies and to analyze damaging floods systematically across a broader set of watersheds, including ungauged areas. At the same time, I have substantial concerns about the robustness and interpretation of the attribution framework. In my view, the manuscript would benefit from major revision before it is suitable for publication. The key issue is that several central conclusions depend heavily on methodological assumptions whose justification and sensitivity analysis remain incomplete.

Robustness of flood timing definitions (SOF/EOF) needs to be addressed more directly.

A central concern is that the analysis relies on estimated Start of Flood (SOF) and End of Flood (EOF) dates derived from river sections exceeding the 2-year recurrence threshold, and the authors explicitly acknowledge that both SOF and EOF are crude estimates. These dates are then used to define the event search period and therefore directly affect the rainfall attributed to ETCs. The manuscript includes a sensitivity test for the spring SSP-to-SOF interval, but does not appear to test sensitivity to EOF despite the fact that ESP is set relative to EOF. This is a notable omission. The authors should either perform an analogous sensitivity analysis for EOF or provide a much clearer justification for why uncertainty in EOF is expected to have limited influence on the results. More broadly, the manuscript should discuss how uncertainty in SOF/EOF estimation propagates into ETC contribution estimates and event classification.

The spatial attribution framework requires stronger justification.

The manuscript defines local flood event locations by watershed centers and attributes flood-related precipitation to ETCs within a 1,000 km search radius around the affected watershed, assigning rainfall to the ETC closest to the watershed center when multiple ETCs are present. These are important modeling choices, but the rationale remains only partly convincing. It is not fully clear, for example, whether the relevant geometry is consistently defined relative to watershed centers or watershed edges across the different parts of the workflow. More importantly, the choice of a fixed 1,000 km radius seems somewhat arbitrary even if it is informed by prior ETC literature. A spring-only comparison with 750 km is useful, but it does not fully resolve the question of whether this radius is physically well matched to the flood attribution problem. The authors should justify more explicitly why this choice is preferable to alternatives such as watershed-only precipitation, an edge-based or distance-weighted storm attribution scheme, or some physically motivated storm-footprint approach. At a minimum they could present results from a wider range of buffers along with more justification as to why these buffers were chosen.

The manuscript is primarily descriptive, and the conclusions should be aligned more carefully with that scope.

The study contains many interesting descriptive patterns, but at present the discussion sometimes pushes toward stronger attributional or explanatory conclusions than the analyses can fully support. This matters because the manuscript itself states that no formal statistical attribution was performed. I therefore encourage the authors to distinguish more clearly between descriptive association, statistical difference, and physical attribution.

Statistical significance is emphasized more than effect size or practical importance.

Several comparisons are based on Kolmogorov-Smirnov tests or chi-squared tests. Those tests can be appropriate, but as currently presented they do not always communicate the magnitude or practical relevance of the differences. This is especially important for large samples, where null hypotheses are often easy to reject.

The manuscript could benefit from a tighter central narrative.

At present, the paper has several partially competing aims: identifying flood events, attributing rainfall to ETCs, classifying flood and non-flood ETCs, clustering cyclogenesis regions, and discussing implications for flood preparedness. Each component is interesting, but together they make the paper feel somewhat diffuse. I think the manuscript would be stronger if the authors more clearly prioritized one primary contribution.

Minor comments
Figure 1 is informative but overly complex.

The methods overview is useful, but the figure is visually dense and difficult to parse quickly. I recommend simplifying the flow diagram or moving some of the process detail to the supplement.

Figure 4 would benefit from a scale bar.

Because spatial buffers and domain definitions are central to the paper, the map showing the Baseline Southern Quebec domain would be easier to interpret with a scale bar.

Clarify terminology around flood “location.”

The manuscript states that local flood event locations are defined by watershed centers. Since floods are not point phenomena, this wording can be misleading. It would help to clarify consistently whether “flood location” refers to a representative centroid used for the search procedure, rather than the true geographic extent of flooding.
Citation: https://doi.org/10.5194/egusphere-2025-6192-RC2
RC3:
'Comment on egusphere-2025-6192', Anonymous Referee #3, 30 Mar 2026
This is a well-structured manuscript where authors have quantified contribution of extratropical cyclones to river flooding in southern Quebec region. The manuscript provided sufficient background information and reference studies in defining the knowledge gap and setting up the research question/objective of the study. I believe the time period is long enough and the types of data used and their sources were also adequate. The data including their sources were described well. The arguments made by the authors were supported by results shown through clear and well descriptive graphs and tables. However, my concert is in the methods used in the study. While authors claim, “quantification of contribution”, the adopted methods are quite imperial in nature and at times difficult to follow. The adopted method explains the data well individually (e.g. explaining characteristics of flood events, and ETCs), but when it comes to quantifying ETC contribution to flooding, the method becomes more description oriented than quantification, apology if I did not explain it well. Below are few other points to consider.
Ln 415-16: 5 most impactful floods account for 51% of the aid claim, I was wondering if the magnitude of those 5 floods were known? In general, 2-yr return period is used as flooding threshold, however, one can imagine these 5 floods were at significantly higher magnitude than 2-yr return period flooding. So that brings up the questions of what were the characteristics of the flood ETCs that were the cause of these 5 floods? Additionally, what made these ETCs unique enough to result in such extensive flooding? I believe explaining these would be beneficial to readers.

Although it is mentioned in section 3.1.1, it would be good to mention it in section 2.3 that “HydroBASINS level-7” is used in this study.

Ln 209: It would be beneficial to have the spring season defined in section 3.2.1 the same way it is defined for other seasons, since its is not defined anywhere else in the manuscript.

In section 4. Graphs and tables were well described as findings/results; however, it would be quite helpful if authors can briefly state what they implies to, e.g. the “why” part. While reading, I was hoping that section 5 would have that, but missed to find them adequately.
Citation: https://doi.org/10.5194/egusphere-2025-6192-RC3

Clarence Gagnon, Daniel F. Nadeau, Alejandro Di Luca, Benoit Brault, Romane Hamon, Nicolas L. Roy, Marc-André Bourgault, and François Anctil

Supplement

https://doi.org/10.5194/egusphere-2025-6192-supplement

Data sets

Data for "Assessing the contribution of extratropical cyclones to river floods that caused property damage in Quebec, Canada" Clarence Gagnon, Daniel F. Nadeau, Alejandro Di Luca, Benoit Brault, Romane Hamon, Nicolas L. Roy, Marc-André Bourgault, and François Anctil https://doi.org/10.5683/SP3/98HDVT

Clarence Gagnon, Daniel F. Nadeau, Alejandro Di Luca, Benoit Brault, Romane Hamon, Nicolas L. Roy, Marc-André Bourgault, and François Anctil

Viewed

Total article views: 270 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
137	110	23	270	41	15	28

HTML: 137
PDF: 110
XML: 23
Total: 270
Supplement: 41
BibTeX: 15
EndNote: 28

Views and downloads (calculated since 02 Feb 2026)

Month	HTML	PDF	XML	Total
Feb 2026	71	44	10	125
Mar 2026	50	46	11	107
Apr 2026	16	20	2	38

Cumulative views and downloads (calculated since 02 Feb 2026)

Month	HTML	PDF	XML	Total
Feb 2026	71	44	10	125
Mar 2026	50	46	11	107
Apr 2026	16	20	2	38

Viewed (geographical distribution)

Total article views: 257 (including HTML, PDF, and XML) Thereof 257 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 09 Apr 2026

Short summary

This article studies the contribution of extratropical cyclones to nearly 500 river floods that caused material damage in the province of Quebec, in eastern Canada, between 1991 and 2020. By bringing significant amounts of precipitation, extratropical cyclones played a key role in the majority of floods. During the period, over half of the flood-contributing storms originated in the central United States and the Gulf of Mexico, and just a handful caused most of the financial damage observed.


Total:	0
HTML:	0
PDF:	0
XML:	0