No abstract

No abstract

River confluences have a complex flow and sedimentation pattern that have vital influences on the hydraulic and bed morphology of river reach and the surrounding area. Confluences can be observed in waterways with various situations such as live bed conditions. This condition is a hydro-morphological situation with a high densimetric Froude number, i.e., bed load transport is supplied from upstream. According to the literature review, most of the experimental studies investigate the flow pattern on the flatbed and not on the developed riverbed, or/and in the low densimetric Froude number, or/and without supplying the sediment from upstream. Therefore, in the present study for the quantification of the flow pattern under these conditions, each developed bed was fixed with the cement blanket method in the laboratory. Then, the 3D velocity was measured at specific points at the confluence. The current study was designed to understand the flow pattern corresponding to the river bed behavior in the case of large floods. It is expected that the morphological features downstream of the confluence have a different pattern than the ones in the condition described in other literature. Thus, this paper describes briefly what are the different bed features and investigates the corresponding flow pattern. The results of the flow pattern on the developed bed show that all zones at the river confluence can be observed except the point bar due to the approximate equality of the mean longitudinal velocity of the separation zone and the main channel upstream of the confluence. Moreover, results show that by increasing the bedload ratio (sediment discharge to water discharge of the main channel of upstream of the confluence) from 0 to 3 × 10−4, flow deflection to the outer bank of the channel decreased down to 45%, the stagnation equivalent area decreased down to 2.5 times, and bed shear stress decreased down to 40%. Hence, the momentum of lateral flow decreased with increasing bedload. Besides, the recovery zone occurred at a longer distance after the confluence compared to the case without bedload. Hence, the location of the maximum velocity zone, vortices, and secondary flows changed downstream of the confluence, by changing the bed load value.

<p>Channel confluence is one of the important sections of channel networks which is also common encountered in nature. Six different zones exist at a channel confluence: 1) stagnation zone, 2) flow deflection zone, 3) flow separation zone, 4) maximum velocity zone, 5) flow recovery zone and 6) shear layers between combining flows zone. Due to the complexity of flow pattern at channel confluence, this location is always interesting among researchers. Although there are a number of studies on the flow and sediment pattern at confluences, there are still some gaps to be studied. Hence, a calibrated numerical model should be a good tool for evaluating the various effective parameters on flow and sediment patterns. The numerical 2D shallow-water model used in this paper is SFLOW which was developed by NTNU. Besides, the model calibration part of the current study is done by using a set of data from laboratory experiments.</p><p>This study attempt to simulate bed changes at channel confluences with a 2D shallow-water modeling under non-hydrostatic pressure, and show the applicability of the SFLOW model for this complex flow pattern. SFLOW solving the depth-averaged Navier-Stokes equations which is equipped with cutting-edge solvers. Besides, SFLOW modeled turbulency with depth-averaged two-equation RANS. In comparison with other codes, one of the interesting features of SFLOW is solving the non-hydrostatic pressure besides of hydrostatic part. This leads to a more realistic representation of the complex flow and sediment patterns of channel confluences, and consider less computational power than full 3D models.</p>

No abstract