Evaluating the Impact of Turbulence Closure Models on Solute Transport Simulations in Meandering Open Channels

Abstract

River meanders form complex 3D flow patterns, including secondary flows and flow separation. In particular, the flow separation traps solutes and delays their transport via storage effects associated with recirculating flows. The simulation of the separated flows highly relies in the performance of turbulence models. Thus, these closure schemes can control dispersion behaviors simulated in rivers. This study performs 3D simulations to quantify the impact of the turbulence models on solute transport simulations in channels under different sinuosity conditions. The 3D Reynolds-averaged Navier-Stokes equations coupled with the k − ε , k − ω and SST k − ω models are adopted for flow simulations. The 3D Lagrangian particle-tracking model simulates solute transport. An increase in sinuosity causes strong transverse gradients of mean velocity, thereby driving the onset of the separated flow recirculation along the outer bank. Here, the onset and extent of the flow separation are strongly influenced by the turbulence models. The k − ε model fails to reproduce the flow separation or underestimates its size. As a result, the k − ε model yields residence times shorter than those of other models. In contrast, the SST k − ω model exhibits a strong tailing of breakthrough curves by generating more pronounced flow separation.

Analysis of Storage Effects in the Recirculation Zone Based on the Junction Angle of Channel Confluence

“…The recirculating flow in a confluence can form a storage zone that traps the solute inflow from a tributary. Such a storage zone increases the residence time of solutes and generates a long-tailed concentration curve [43]. The mixing characteristics in rivers change as a result of these storage zones, which affects far-field mixing results.…”

Comparisons of Two Types of Particle Tracking Models Including the Effects of Vertical Velocity Shear

“…The main features of measured concentration curves are that the concentration curves have a steep, rising limb and long tail. These skewed concentration curves can be observed in the initial period [32] or the Taylor period, including a recirculation zone [43]. As the curvature of the meandering channel is not sharply curved enough to cause the recirculating flow [44], it can be said that the skewed concentration curves are not attributable to the storage effects.…”

Lagrangian Observations and Modeling of Turbulence Along a Tidally Influenced River

“…Here, we use high‐resolution Lagrangian field observations to validate a numerical model and to explore the flow‐following distributions of turbulence along a river. Turbulence in meandering rivers has also been studied via modeling approaches, including models based on the RANS equations (e.g., Almeida & Ota, 2020; Fischer‐Antze et al., 2009; Gholami et al., 2014; Kim et al., 2020; Rameshwaran & Naden, 2004; Wormleaton & Ewunetu, 2006), which rely on the performance of the turbulence models. More recently, the impact of turbulence closure schemes on the flow in curved channels has been investigated (e.g., Almeida & Ota, 2020; Kim et al., 2020).…”

“…Turbulence in meandering rivers has also been studied via modeling approaches, including models based on the RANS equations (e.g., Almeida & Ota, 2020; Fischer‐Antze et al., 2009; Gholami et al., 2014; Kim et al., 2020; Rameshwaran & Naden, 2004; Wormleaton & Ewunetu, 2006), which rely on the performance of the turbulence models. More recently, the impact of turbulence closure schemes on the flow in curved channels has been investigated (e.g., Almeida & Ota, 2020; Kim et al., 2020). However, few studies have used high‐resolution Lagrangian field observations to validate a numerical model and to explore the flow‐following distributions of turbulence along a river.…”

“…These studies have also shown that different turbulence models predicted slightly different simulated flow; for instance, Almeida and Ota (2020) demonstrated that the k-ϵ closure scheme could reliably reproduce the velocity all along the domain in their 2D-simulation of a curved channel whereas the other model they tested, the Elder Model (an algebraic model), could not reproduce the velocity field both in sections of high curvature and in straight portions of the channel. Similarly, other studies have found that 3D velocity structure (and hence predictions of solute transport) are sensitive to the selected turbulence closure scheme, mainly due to the difficulty in perfectly reproducing the flow separation (Kim et al, 2020). While previous studies have focused on the across-river structure of turbulence around a single bend and reveal complex and often contradictory results (Engel & Rhoads, 2017), the along-river distributions of turbulence are relatively less explored-particularly for longer sections covering tens of kilometers.…”