Depth-averaged flow computation at a river confluence

Abstract

A depth-averaged elliptic computational model in curvilinear coordinates is presented for velocity and depth computations in shallow water river confluences of any geometry. The pressure correction equation has been used with SIMPLE or SIMPLE-like procedures in depth-averaged models to date; here this is replaced by a depth correction equation to improve the convergence. This is a significant enhancement to the SIMPLE procedure when applied to depth-averaged flow computations of river elements with irregular bed topography. First, the model is validated using laboratory measurements from a flat-bed confluence with flow recirculation. Second, it is validated using laboratory measurements from a confluence with irregular bed topography derived at the dynamic equilibrium of a movable bed experiment. Next the model is employed to simulate the complex confluence flow at a bend of a river with a floodplain. The model predictions give satisfactory agreement with field observations. Though the flow in the scour hole region of a river confluence is three-dimensional, if proper parameters are used to represent bottom friction and mixing, a depth-averaged model can be used to obtain sufficiently accurate velocity and water surface elevation information for the design of flood control measures in shallow water river confluences.

Hydro‐Thermodynamic Processes at a Large Confluence Under Reservoir Regulation

“…The shear layer indicate a high level of turbulence, or a large‐scale coherent structure, which is the main mechanism of the mixing interface in the scenario with no density difference (Rhoads & Sukhodolov, 2008). The separation zone is labeled as region A to the left of the shear layer zone, which is a typical characteristics related to the dynamic process in the mainstream‐tributary confluence found both in laboratory experiments (Hua et al., 2009; Rhoads & Sukhodolov, 2004) and field surveys (Baranya et al., 2010; Konsoer & Rhoads, 2013; Rhoads & Sukhodolov, 2004; Weerakoon et al., 2003). However, during the post TGD period, the separation zone almost disappeared (see Figure 6b), because the momentum of the tributary flow seriously reduced (Mr = 0.07) under the high water level condition, and thus the inertial effect of the lateral flow is too small to form a full separation.…”

Modelling Hydraulics and Sediment Transport at River Confluences

“…This is particularly the case for natural junctions. Some attempts were made to use two-dimensional (Khan et al, 2000;Weerakoon et al, 2003;Zanichelli et al, 2004) or pseudo three-dimensional models (Wang et al, 1996) for confluence modelling. However, Lane et al (1999) show that the predictive ability of a three-dimensional model is markedly increased over a two-dimensional model at confluences, particularly if the two-dimensional model is not corrected for the effect of secondary circulation.…”

“…Furthermore, three-dimensional velocity data collected at confluences indicate large variations from the bed to the water surface in the flow field (De Serres et al, 1999;Rhoads and Sukhodolov, 2001), which would not be adequately simulated in a two-dimensional model. However, for large confluences where, for example, the objective is to investigate flood-control measures rather than detailed mixing processes (Weerakoon et al, 2003), depth-averaged models still represent a reasonable compromise if secondary circulation corrections are available.…”

Flow Modelling of Mahaweli River Reach with the Riverbed Intake of Udunuwara-Yatinuwara Water Supply Scheme

“…The applications of depth-averaged models in Cartesian coordinates to natural rivers have been presented by ASCE [1]. Also, Wijbenga [9], Wenka et al [8], Ye and McCorquodale, Weerakoon [6], Weerakoon et.al [7] presented applications in -curvilinear coordinates which are more suitable for applications to rivers.…”