Dam-Break Waves Over Rough Beds

Abstract

In the context of today’s climate change, with extreme events becoming more frequent and more intense, highly unsteady flows are a threat that can no longer be ignored in hydraulic and coastal engineering, since these can lead to human casualties and extensive damage. Impulse waves, storm surges, flash floods and tsunamis are among these unsteady flows, with tragic examples in the last years, including the Indian Ocean tsunami in 2004, Japan Tohoku in 2011 and Indonesia in 2018. These events showed that a deeper knowledge of the underlying physical phenomena is necessary to ensure safety to people and minimize expenses associated with recovery. Due to rarity and complexity of these flows, experimental approaches are often required and in laboratories unsteady flows can be reproduced using dam-break waves (Ritter 1892, Stoker 1958). However, most laboratory tests are conducted on (unrealistic) smooth inverts, hence rising the question on how the bed roughness affects the propagation and the hydrodynamic properties of these flows. Previous studies, including Dressler (1952), Wüthrich et al. (2019) and Nielsen et al. (2022) provided relevant information, but more research is needed to gain a better understanding. In particular, little knowledge is available on the behaviour of these highly unsteady flows propagating through Rigid Stagged Vegetation (RSV), which is representative of forests and other natural areas surrounding built environments.

Based on a large experimental campaign, this research studied the propagation of dam-break waves on rough beds, in the form of various configurations of RSV. Waves were generated in a 14 m long and 0.4 m wide horizontal flume, where a d₀ = 0.4 m impounded reservoir was released through the sudden opening of a gate, as shown in Figure 1. The waves propagated in the downstream horizontal flume, where different roughness configurations are installed. More specifically, the study analysed a smooth plywood configuration and 4 Rigid Stagged Vegetation (RSV) configurations (Figure 2), reproduced using nails with various grid densities and lengths, as detailed in Table 1. Tests were conducted on dry bed, as well as on an initial still water level h₀, which ranged between 7.5 and 50 mm (i.e. 0.0188 < h₀/d₀ < 0.125). Tests on dry bed were repeated 5 times, while tests on wet bed were repeated 10 times. Data were analysed using ensemble-average values. Six Acoustic Displacement Meters ADM (Microsonic TM mic+35/IU/TC, Dortmund Germany) were used to capture the wave profiles in time as well as the wave front celerity C between various ADMs. In this study only the celerities between ADM 5 and 6 are considered, since at this location the bore was fully developed (Buitelaar 2022). Wave propagation was also documented using videos and SLR high speed photographs.