Japan Geoscience Union Meeting 2025

Presentation information

[E] Poster

A (Atmospheric and Hydrospheric Sciences ) » A-HW Hydrology & Water Environment

[A-HW29] Climate, Rivers, and Floods: Exploring Hydro-Geomorphological Interactions

Wed. May 28, 2025 5:15 PM - 7:15 PM Poster Hall (Exhibition Hall 7&8, Makuhari Messe)

convener:Laurence Paul Hawker(Organization Not Listed), Tomohiro Tanaka(Kyoto University), Stephen E Darby(University of Southampton)

5:15 PM - 7:15 PM

[AHW29-P01] Full-Scale Hydraulic Experiment on the Process of Cohesive Levee Breach and Geomorphological Changes in the Floodplain

*Yuzuno Kanbara1, Tomonori Shimada1, Shunichi Maeda1, Nobuyuki Hotta1, Tsuyoshi Miura2, Hisashi Kamei3 (1.Civil Engineering Research Institute for Cold Region Public Works Research Institute, 2.Hokkaido Regional Development Bureau, 3.Obihiro Development and Construction Department, Hokkaido Regional Development Bureau)

Keywords:levee breach, cohesive soil, floodplain, crevasse splay

1. Introduction
When a river overflows, inundation occurs in the floodplain. If a levee breaches, flood discharge increases, widening the inundated area. Furthermore, as flood discharge increases, a levee breach causes not only inundation but also erosion and sedimentation, exacerbating the damage. During the East Japan Typhoon in 2019, it was reported that sediment deposition on the floodplain increased the time and cost of restoration due to the need to remove the sediments even after the floodwaters receded.[1] However, the process of levee breach is not well understood, especially regarding cohesive levees, in comparison with sand and gravel levees. [2] Additionally, there are few quantitative studies on geomorphological changes in floodplains compared to river channels. Therefore, to clarify these phenomena, we conducted a full-scale cohesive levee breach experiment.
2. Methods
The experiment was conducted at the Chiyoda Experimental Channel in Hokkaido, Japan. [3] Figure 1 shows the test site. The levee was 30 m long, 2 m wide, and 2.5 m high, while the floodplain was 70 m long, 30 m wide, and with a slope of 1/500. The experiment was conducted as follows: The flow was initially constant and was then temporarily stopped for geomorphological measurements. Next, the flow rate was increased, and the levee was breached after 6 hours. The levee material was cohesive soil containing 40% silt and clay combined, and the riverbed material was sand and gravel. We measured water levels, flow velocity, and elevation, recording the flow regime on video.
3 Results
3.1 The process of the cohesive levee breach
We found that the process can be divided into the following three stages.
(1) Erosion of the back slope of the levee: When the river overflowed over the levee, the back slope was eroded stepwise.
(2) Levee breach: As the erosion of the back slope progressed, the top of the levee was eroded, reaching the shoulder of the front slope. As a result, the cross-sectional area of the levee became smaller and lost stability, and the front slope rapidly disappeared, resulting in a breach.
(3) Expansion of the breach width: Figure 2 shows the temporal change in breach width. Since the levee material was cohesive, little surface soil was transported by water. Similar to the lateral erosion of cohesive levees [4], it is considered that sand and gravel in the foundation eroded first, causing the levee to collapse every 1 to 2 m, widening the breach.
3.2 Geomorphological changes in the floodplain
Figure 3 shows geomorphological change and flow velocity. In the early stage of the experiment, sedimentation on the floodplain was a few centimeters. After the levee breach, a crevasse channel formed just below the levee. And sand and gravel were deposited around it, creating a tongue-shaped crevasse splay. Behind them, fine soil was deposited, and soil clods up to 1 m long were scattered. These results suggest sediment sorting due to differences in flow velocity.
4. Conclusions
In this study, we observed that the process of cohesive levee breach and geomorphological changes in the floodplain. In the future, we will analyze the hydrological conditions in more detail and compare the results with actual flood events for disaster prevention.
Acknowledgments
This research was supported by the Chiyoda Experimental Channel Advisory Committee and the Experimental Review Committee.
References
[1] Kawaike K., Takeda M., Toyota M., Yokawa H., Yamanoi K. and Nakagawa H.: Field survey and numerical simulation of fluvial inundation and sediment deposition due to dyke breach of Chikuma River, Journal of Japan Society of Civil Engineers, Ser. B1 (Hydraulic Engineering), Vol.76, No.2, pp.I_883-I_888, 2020. (in Japanese)
[2] ASCE/EWRI Task Committee on Dam/Levee Breaching: Earthen embankment breaching., J. Hydraul. Eng., 137(12), 1549–1564, 2011.
[3] Experiment on a Levee Breach from Overtopping at the World’s Largest Experimental Channel, http s://www.ceri.go.jp/news/2024GEWEX_Shimada.pdf (See 2025-02-06)
[4] K. Zhang, Z. Gong, K. Zhao, K. Wang, S. Pan, G. Coco: Experimental and numerical modeling of overhanging riverbank stability, Journal of Geophysical Research-Earth Surface 126(10), 2021.