the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 19 Aug 2024
Brief Communication: Rapid high-resolution flood impact-based early warning is possible with RIM2D: a showcase for the 2023 pluvial flood in Braunschweig
Abstract. In recent years, urban areas have been increasingly affected by more frequent and severe pluvial floods, attributed to climate change and urbanization. This trend is expected to continue in the future, underscoring the critical importance of flood warning and disaster management. However, pluvial flood forecasts on a communal level do not exist in practice, mainly due to the high computational run-times of high-resolution flood simulation models. Here, we showcase the capability of the hydrodynamic model RIM2D (Rapid Inundation Model 2D) to deliver highly detailed and localized insights into expected flood extent and impacts in very short computational processing times, enabling operational flood warnings. We demonstrate these capabilities using the case of the June 2023 torrential rain and resulting flood event in the city of Braunschweig, located in Lower Saxony, Germany. During this event, the city experienced intense rainfall of 60 liters per square meter within a short timeframe, resulting in widespread inundation, significant disruption to the residents' daily life, and substantial economic costs to the city. This study serves as a clear indication that different dimensions of flood consequences can be simulated at very high resolutions in extremely short computational times and that models such as RIM2D, along with the necessary hardware for their operation, have reached a level of maturity suitable for integration into operational early warning systems and impact-based forecasting systems for such floods.
- Preprint
(2111 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on nhess-2024-139', Anonymous Referee #1, 23 Sep 2024
This study shows a newly developed/combination technique that has the potential to enhance flood early warning systems based on the level of impacts. The topic is important, the writing is clear, the logic is sound, and the results appear to be significant. Overall, I suggest only minor revisions are needed before it can be considered for publication in NHESS as a Brief Communication paper. Specifically, I have a few comments:
1. While I understand the challenges of quantitatively validating the modeling results due to the lack of systematic data, such validation is crucial before asserting that the predictions of this technique are evident, as the authors claim in the conclusion. Could the authors explore indirect approaches to quantify the results? For instance, calculating the ratio of coherent versus incoherent grids based on the predicted and observed inundation areas (using only the grids with observations) could be akin to a confusion matrix analysis.
2. As the authors suggest, providing an uncertainty map for this tool is both important and useful. Why was this map not included in the exemplary case presented in the study?
3. The manuscript currently contains over 20 references, which exceeds the limit for a Brief Communication in NHESS. Additionally, there are a few typographical errors that need correction.
Citation: https://doi.org/10.5194/nhess-2024-139-RC1 -
RC2: 'Comment on nhess-2024-139', Anonymous Referee #2, 16 Dec 2024
The paper analyzes the capabilities of the RIM2D hydrodynamic model for flood impact forecasting, using the 2023 pluvial flood in Braunschweig, Germany, as a case study. The authors demonstrate that RIM2D produces accurate simulations of inundation areas, depths, and flow velocities with computational efficiency, supporting timely forecasts and early warnings. They evaluated computation times for a 48-hour flood episode and 2-hour design floods, showing that such analyses can be effectively integrated into real-time flood early warning systems. The study’s main scientific contribution is the evidence that this technology has reached sufficient maturity for operational use in early warning systems. While the results are specific to the studied location, the methodology could be adapted and applied to other areas to validate these findings further.
In my opinion, the research is relevant and deserves publication in Natural Hazards and Earth System Science. The topic is interesting and fits in the scope of the journal, the objectives are clearly defined, previous work is properly presented and acknowledged, the methodology is clearly explained, the results are adequately described and discussed, and the conclusions are useful for many scientists interested in the topic. Overall, the paper is clearly written, well structured, and correctly illustrated with figures.
I congratulate the authors on their good work and encourage them to incorporate a few suggestions for improvement.
1. Present the modeling domain. It would be interesting for the reader to know some basic data of the modeling domain: extension of the city of Braunschweig, number of inhabitants, population density, orography, climate, identified flood risks, drainage system, flood protection, type of economic activities that are developed, etc.
2. Describe the time scale of the flood being analyzed. If the phenomenon is primarily caused by pluvial flooding, the time available between the onset of precipitation and the occurrence of damage is likely to be relatively short. While the analyses presented seem to focus on the moment of maximum flood extent, an early warning system requires understanding how far in advance potential damage can be predicted. It would be valuable if the authors could provide insights or comments on this aspect.
3. Comment on the early warning system into which this tool could be integrated. While simulation accuracy and computation times are crucial, they are not the only requirements for incorporating such tools into early warning systems. To evaluate the utility of this tool within such a system, it is essential to thoroughly characterize the system by addressing several key aspects:
Define the system's objectives (monitoring, decision support, emergency warning, …).
Describe the frequency and cadence of radar data updates. The text specifies that radar data are available every hour. How long does it take to have rainfall data for the previous hour available?
Specify whether any nowcasting methods are in place to predict precipitation in advance. This may gain some lead time for the warning.
Outline the proposed real-time workflow, including data processing and integration. What is the actual time window for the 2D flood inundation model to allow for a timely warning? Account for the time required to generate results, understand them, and disseminate critical information.
In summary, assess whether the system as a whole can meet its mission goals within the required lead-times to effectively protect the population. A discussion of these factors would provide valuable insight into the practical integration and effectiveness of the tool in operational early warning systems.
Specific suggestions
Line 84-86: The authors indicate that buildings are supposed to act as reflective boundaries. What about other types of obstacles present in the city, such as street furniture, parked or circulating cars? In some floods these elements may cause important obstructions to the flow. Is that considered?
Line 253. I am not sure this information on running the model with multiple GPUs fits in the conclusions section, since it was not covered in the results or discussion sections. I would suggest presenting it earlier.
Typos:
From the formal standpoint, the paper is correctly organized and well written. However, I noticed a few typos:
Line 55 With-> with
Line 57 furthermore->Furthermore
Line 99 RADOALN->RADOLAN
Line 110 provides->provide
Line 192 were->where
Citation: https://doi.org/10.5194/nhess-2024-139-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 199 | 44 | 15 | 258 | 9 | 8 |
- HTML: 199
- PDF: 44
- XML: 15
- Total: 258
- BibTeX: 9
- EndNote: 8
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1