the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
A globally-applicable framework for compound flood hazard modeling
Abstract. Coastal river deltas are susceptible to flooding from pluvial, fluvial, and coastal flood drivers. Compound floods, which result from the co-occurrence of two or more of these drivers, typically exacerbate impacts compared to floods from a single driver. While several global flood models have been developed, these do not account for compound flooding. Local scale compound flood models provide state-of-the-art analyses but are hard to scale up as these typically are based on local datasets. Hence, there is a need for globally-applicable compound flood hazard modeling. We develop, validate and apply a framework for compound flood hazard modeling, which consists of the local high-resolution 2D hydrodynamic flood model SFINCS, which is automatically set up from global datasets and loosely coupled with a global hydrodynamic river routing and flood model, as well as a global surge and tide model to account for interactions between all drivers. To test the framework, we simulate two historical compound flood events, cyclones Idai and Eloise, in the Sofala province of Mozambique, and compare the flood extent to observations from remote sensing and to the global quasi 2D CaMa-Flood model. The results show that while the global and local model have similar skill in terms of the critical success index, they result in rather different flood maps. On the one hand, the local model has a higher hit ratio due to the representation of direct coastal and pluvial flooding (rain on grid) and a higher floodplain connectivity. It also shows a faster response to coastal drivers within the estuaries and more realistic flood depth maps. On the other hand, the local model has a higher false alarm ratio, which is partly explained by the inclusion of direct pluvial flooding without sufficient representation of small scale (subgrid) drainage capacity. To showcase a possible application of the framework, we also determine the dominant flood drivers and transition zones between flood drivers for both events. These vary significantly between both events because of differences in the magnitude of and time lag between the flood drivers. We argue that a wide range of plausible events should be investigated to get a robust understanding of compound flood interactions, which is important to understand for flood adaptation, preparedness, and response. As the model setup and coupling is automated, reproducible, and globally applicable, the presented framework is a promising step forward towards large scale compound flood hazard modeling.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(2546 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(2546 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2022-149', Anonymous Referee #1, 13 Jul 2022
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2022/egusphere-2022-149/egusphere-2022-149-RC1-supplement.pdf
-
AC1: 'Reply on RC1', Dirk Eilander, 05 Oct 2022
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2022/egusphere-2022-149/egusphere-2022-149-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Dirk Eilander, 05 Oct 2022
-
RC2: 'Comment on egusphere-2022-149', Anonymous Referee #2, 15 Jul 2022
The article presents a methodology for local-scale compound flood modeling using global input datasets and a newly developed hydrodynamic model (SFINCS). The framework is applied to a coastal catchment in Mozambique that recently experienced flooding from two tropical cyclones (Idai and Eloise). The proposed methodology for developing local-scale models based on global datasets/inputs is very interesting, and I believe the paper could be an important contribution to the compound modeling literature. However, I believe some more analysis/discussion on the model validation is needed before this work can be published. Therefore, I recommend a moderate revision.
My main concern/issue with the paper is that the results presented in section 4.1 are not compelling. It seems like the local-scale model and the global CaMa model perform similarly well for the two historical cases, which calls into question why someone should go through the trouble of setting up a high-resolution local model if similar accuracy can be achieved with an existing global model. To be clear, I believe there is a lot of value in using a high-resolution local model for flood hazard analysis, I just don’t think the results presented in section 4.1 do a good job of showing the additional benefit. Can any additional validation data, performance metrics, discussion, etc. be added to this section to show more clearly the benefit of using the SFINCS model? The ability to efficiently set up and run local-scale compound flood models for any catchment across the globe is really promising, but we need more confidence that the local-scale model will provide higher accuracy compared to existing global models.
I have some other specific comments below:
3.1.1
I wonder if an ocean model with 2.5 km coastal resolution can adequately capture peak storm tides from TCs, which tend to produce extreme storm surges over relatively small geographic areas (cite).
Lines 142-149: I was confused here as the sources of the storm surge, wind setup, and tide heights were not clear. The 5.0 m max water level (and 3.8 m for Eloise) reported here is based on what? Gauge, high water marks, reports, models? What is the max water level predicted by the author’s global model? I see 4.0 m as the max surge estimate, but what about the total modeled water level? Also, how is the “operational forecast” generated? Is this another global model that the authors compared with? In general I think this paragraph needs to be re-written to be clear about how their model results compare with the results from other models or other sources.
3.2
Figure 3: What does “actual event discharge” mean? The river discharge based on the gauge records or the CaMa discharge? If the latter, I would call it model-based discharge since it is not the “true” discharge.
3.3
In addition to simulated flood extent, can any comparison be made using simulated vs satellite-based flood depth? In low-lying regions, the flood extent could be similar between the model and satellite, but the depth could be significantly different. I’m not asserting that the author’s flood model is inaccurate, but just want to point out that a comparison based on flood extent alone does not provide a complete picture about whether flood dynamics are being accurately captured by the modeling framework.
4.1
It seems that although SFINCS simulates a larger extent of flooding than CaMa (due to incorporation of pluvial runoff), CaMa consistently predicts higher flood depths for both storms (except at location 5). I wonder if the authors have any ideas why CaMa would estimate higher flood depth than SFINCS?
Citation: https://doi.org/10.5194/egusphere-2022-149-RC2 -
AC2: 'Reply on RC2', Dirk Eilander, 05 Oct 2022
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2022/egusphere-2022-149/egusphere-2022-149-AC2-supplement.pdf
-
AC2: 'Reply on RC2', Dirk Eilander, 05 Oct 2022
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2022-149', Anonymous Referee #1, 13 Jul 2022
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2022/egusphere-2022-149/egusphere-2022-149-RC1-supplement.pdf
-
AC1: 'Reply on RC1', Dirk Eilander, 05 Oct 2022
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2022/egusphere-2022-149/egusphere-2022-149-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Dirk Eilander, 05 Oct 2022
-
RC2: 'Comment on egusphere-2022-149', Anonymous Referee #2, 15 Jul 2022
The article presents a methodology for local-scale compound flood modeling using global input datasets and a newly developed hydrodynamic model (SFINCS). The framework is applied to a coastal catchment in Mozambique that recently experienced flooding from two tropical cyclones (Idai and Eloise). The proposed methodology for developing local-scale models based on global datasets/inputs is very interesting, and I believe the paper could be an important contribution to the compound modeling literature. However, I believe some more analysis/discussion on the model validation is needed before this work can be published. Therefore, I recommend a moderate revision.
My main concern/issue with the paper is that the results presented in section 4.1 are not compelling. It seems like the local-scale model and the global CaMa model perform similarly well for the two historical cases, which calls into question why someone should go through the trouble of setting up a high-resolution local model if similar accuracy can be achieved with an existing global model. To be clear, I believe there is a lot of value in using a high-resolution local model for flood hazard analysis, I just don’t think the results presented in section 4.1 do a good job of showing the additional benefit. Can any additional validation data, performance metrics, discussion, etc. be added to this section to show more clearly the benefit of using the SFINCS model? The ability to efficiently set up and run local-scale compound flood models for any catchment across the globe is really promising, but we need more confidence that the local-scale model will provide higher accuracy compared to existing global models.
I have some other specific comments below:
3.1.1
I wonder if an ocean model with 2.5 km coastal resolution can adequately capture peak storm tides from TCs, which tend to produce extreme storm surges over relatively small geographic areas (cite).
Lines 142-149: I was confused here as the sources of the storm surge, wind setup, and tide heights were not clear. The 5.0 m max water level (and 3.8 m for Eloise) reported here is based on what? Gauge, high water marks, reports, models? What is the max water level predicted by the author’s global model? I see 4.0 m as the max surge estimate, but what about the total modeled water level? Also, how is the “operational forecast” generated? Is this another global model that the authors compared with? In general I think this paragraph needs to be re-written to be clear about how their model results compare with the results from other models or other sources.
3.2
Figure 3: What does “actual event discharge” mean? The river discharge based on the gauge records or the CaMa discharge? If the latter, I would call it model-based discharge since it is not the “true” discharge.
3.3
In addition to simulated flood extent, can any comparison be made using simulated vs satellite-based flood depth? In low-lying regions, the flood extent could be similar between the model and satellite, but the depth could be significantly different. I’m not asserting that the author’s flood model is inaccurate, but just want to point out that a comparison based on flood extent alone does not provide a complete picture about whether flood dynamics are being accurately captured by the modeling framework.
4.1
It seems that although SFINCS simulates a larger extent of flooding than CaMa (due to incorporation of pluvial runoff), CaMa consistently predicts higher flood depths for both storms (except at location 5). I wonder if the authors have any ideas why CaMa would estimate higher flood depth than SFINCS?
Citation: https://doi.org/10.5194/egusphere-2022-149-RC2 -
AC2: 'Reply on RC2', Dirk Eilander, 05 Oct 2022
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2022/egusphere-2022-149/egusphere-2022-149-AC2-supplement.pdf
-
AC2: 'Reply on RC2', Dirk Eilander, 05 Oct 2022
Peer review completion
Journal article(s) based on this preprint
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
825 | 435 | 19 | 1,279 | 16 | 9 |
- HTML: 825
- PDF: 435
- XML: 19
- Total: 1,279
- BibTeX: 16
- EndNote: 9
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1
Cited
6 citations as recorded by crossref.
- HydroMT: Automated and reproducible model building and analysis D. Eilander et al. 10.21105/joss.04897
- Connecting hydrological modelling and forecasting from global to local scales: Perspectives from an international joint virtual workshop A. Dasgupta et al. 10.1111/jfr3.12880
- A Hybrid Framework for Rapidly Locating Transition Zones: A Comparison of Event‐ and Response‐Based Return Water Levels in the Suwannee River FL R. Jane et al. 10.1029/2022WR032481
- Estimating nearshore infragravity wave conditions at large spatial scales T. Leijnse et al. 10.3389/fmars.2024.1355095
- Impact of Tides and Surges on Fluvial Floods in Coastal Regions H. Liang & X. Zhou 10.3390/rs14225779
- Towards a global impact-based forecasting model for tropical cyclones M. Kooshki Forooshani et al. 10.5194/nhess-24-309-2024
Anaïs Couasnon
Tim Leijnse
Hiroaki Ikeuchi
Dai Yamazaki
Sanne Muis
Job Dullaart
Hessel C. Winsemius
Philip J. Ward
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(2546 KB) - Metadata XML