Author(s): Arash Kanani; Ana Maria Ferreira Da Silva
Linked Author(s): Ana Maria Ferreira da Silva
Keywords: River flows; Horizontal turbulence; Coherent structures; Alternate bars; Meandering
Abstract: It has been recognized since many years that open-channel and river flows contain large-scale coherent structures of sizes comparable to the external flow dimensions. The large-scale structures can be vertical, in which case they rotate in the vertical planes of flow around horizontal axes, or horizontal, in which case they rotate in the horizontal planes of flow around vertical axes. Such structures are “born” , grow so as to eventually occupy the body of fluid, and finally disintegrate (or “die” ). Yalin (1992) and Yalin and da Silva (2001) advanced the hypothesis that the bed forms known as alternate bars occur as “imprints” on the movable bed of large-scale horizontal coherent structures (HCS’s). Recent experimental studies on the scales and dynamics of HCS’s in a shallow flow provided support for this hypothesis. In this paper, the writers extend the just mentioned studies with the aim of explaining why alternate bars cease to occur in relatively deep flows. For this purpose, a deep laboratory flow (width–todepth ratio≈7), incompatible with the occurrence of alternate bars, is investigated. The flow was conveyed in a 1 m wide, 21 m long, 0. 4 m deep straight channel, with vertical side walls. The channel bed was formed by a well-sorted coarse silica sand. The related instantaneous and time-averaged flow velocity measurements were carried out with a 2D Son Tek TM Micro ADV. Several independent techniques to treat the velocity signals, including filtering, autocorrelation, continuous wavelet transform and quadrant analysis, are used to gain insight into the length and time scales of the large-scale horizontal coherent structures, as well as aspects of their dynamics. Special attention is paid to the variation of turbulence structure over the flow depth. The results are compared with those obtained for the case of the shallow, alternate bar inducing flow previously investigated. The present work is used also to shed light on the relation between alternate bars and meandering and, in particular, on why the initiation of meandering in alluvial streams is not always preceded by the occurrence of alternate bars.