Author(s): Pietro Beretta Piccoli; Youichi Yasuda

Linked Author(s): Pietro Beretta Piccoli, Youichi Yasuda

Keywords: Lternative installation; Gravel mounts; Refuge areas; Flood stages

Abstract: Channelized rivers are characterized by straight homogeneous fast flowing flow. The ecological value is poor, particularly during flood stages, when the riverine inhabitants are flushed downstream by the strong hydraulic conditions. This research aims to demonstrate how the introduction of alternative gravel mount on the riverbed could improve the flow heterogeneity, as well as generate refuge areas that may protect such inhabitants during floods. Data are collected inside a 0.8m wide experimental channel where a 5.4m long physical model is constructed to simulate gravel bed conditions (d50=1.65cm). For each conducted experiment, 7 gravel mounts are inserted every L=1.60m on alternative sides of the channel either using the same gravel as the bed or larger boulders stacked together (D50=6.28cm). The length of each gravel mount is kept constant at 0.6cm, while the height hgm is changed depending on the experiment. Discharge is Q=0.0588m3/s, while two different channel’s slopes are analyzed: I=0.01 (i.e. 1%) and I=0.002 (i.e. 0.2%). Gravel bed and water surface elevations are collected using a point gauge, while the horizontal and lateral velocity components are measured using an electrical-magnetic currentmeter. The definition of refuge area follows previous research on the preferred swim conditions of small-sized fishes. Two thresholds are here considered: mean velocity 10 cm/s and standard deviation of the velocity 7 cm/s (both defined as maximal acceptable limits). Focus is put on the ratio between the refuge areas’ volume Vref and the gravel mounts’ volume Vgm. The installation of gravel mounts with finer diameter (d50) causes the flow to meander significantly and the water surface to rise with strong waves forming in the centre of the experimental channel. The application of stacked boulders (D50) generally fails to deflect the main flow and the mean water surface is slightly higher than the previously described mount morphology, although the water surface is less wavy and more uniform across the channel’s cross section. For each experiment, the stability of the flow is confirmed with a fast Fourier transform on both measured velocity components. Suitable conditions for the definition of refuge areas are mostly found behind each mount close to the channel’s sides. The size of the refuge volume Vref appear to be much more dependent to the mounts’ height hgm than the channel’s slope steepness. The relationship between hgm and Vref is not linear, but a curve with the optimal relative volume Vgm/Vref found at hgm/L=0.03. For same mounts' height hgm, the mount volume Vgm is significantly larger when using finer corns d50. The same result is obtained confronting the refuge volumes Vref. Consequently, similar relative volumes Vgm/Vref=0.17..0.2 are found for both studied mounts morphology d50 (gravel mounts) and D50 (stacked boulders).

DOI: https://doi.org/10.3850/IAHR-39WC252171192022679

Year: 2022