Bar dynamics and bifurcation evolution in a modelled braided sand‐bed river

Abstract

Abstract.Morphodynamics in sand-bed braided rivers are associated to simultaneous evolution of mid-channel bars and channels on the braidplain. Bifurcations around mid-channel bars are key elements that divide discharge and sediment. This, in turn, may control the evolution of connected branches, with effects propagating to both upstream and downstream bifurcations. Recent works on bifurcation stability and development hypothesize major roles of secondary flow and gradient advantage. However, this has not been tested for channel networks within a fully developed dynamic braided river. A reason for this is a lack of detailed measurements with sufficient temporal and spatial length, covering multiple bifurcations. Therefore we used a physics-based numerical model to generate a dataset of bathymetry, flow and sediment transport of an 80-km river reach with self-formed braid bars and bifurcations. The study shows that bar dissection due to local transverse water surface gradients is the dominant bifurcation initiation mechanism, although conversion of unit bars into compound bars dominates in the initial stage of a braided river. Several bifurcation closure mechanisms are equally important. Furthermore, the study showed that nodal point relations for bifurcations are unable to predict short-term bifurcation evolution in a braided river. This is explained by occurrence of non-linear processes and non-uniformity within the branches, in particular migrating bars and larger-scale backwater-effects, which are not included in the nodal point relations. Planform morphology, on the other hand, has predictive capacity: bifurcation angle asymmetry and bar-tail limb shape are indicators for near-future bifurcation evolution. Remote sensing data has predictive value, for which we developed a conThis article is protected by copyright. All rights reserved.ceptual model for interactions between bars, bifurcations and channels in the network. We conducted a preliminary test of the conceptual model on satellite images of the Brahmaputra.

Network response to disturbances in large sand-bed braided rivers

“…This approach is similar to our earlier work in Schuurman et al (2013) and Schuurman and Kleinhans (2015). The hydrodynamics were computed in three dimensions by applying conservation of momentum (Eqs.…”

“…At the same time, discharge and sediment distribution are controlled by the bifurcation topography and local flow pattern. For example, the branch with the smallest angle to the approaching flow is likely to experience the least amount of sedimentation and to become the dominant branch (Mosselman et al, 1995;Schuurman and Kleinhans, 2015). Bar migration and reshaping might change the local flow pattern, thus affecting the nearby upstream bifurcations through the backwater effect and nearby downstream bifurcations by changing the approaching flow direction.…”

“…These bed level changes are expected to modify the discharge and sediment division over nearby bifurcations and the bifurcations become instable ("bifurcation instability"). Then, the network aspect described by Schuurman and Kleinhans (2015) is expected to emerge: bars are reshaped and migration directions are altered by the bifurcation instability, which again affects both the upstream bifurcation through the backwater curve and the downstream bi-28 F. Schuurman et al: Response to disturbances in braided rivers furcation through redirection of the approaching flow. Thus, it is hypothesized that a single disturbance within, along, or upstream of a braiding river reach causes a local bed level change that triggers a cascade of morphological changes in the network, eventually affecting the entire reach.…”

“…We adopted the river dimensions and conditions from Schuurman and Kleinhans (2015) for the default scenario conditions (Table 1): a straight initially plane bed with 3200 m width, 80 km length for scenario series A and 40 km for scenario series B, an initial bed slope of 9.3 × 10 −5 , uniform fine sand (D 50 = 200 µm), and a constant discharge of 40 000 m 3 s −1 (which is close to the effective discharge of the Brahmaputra River). Fixed channel walls were applied by default.…”

“…2. We used two series of simulations: series A to simulate a "realistic" situation starting with a "realistic" bed topography constructed by Delft3D as described by Schuurman and Kleinhans (2015), and series B, which is a simplified situation starting with a regular symmetrical bar pattern. The focus of our study is on series A, as it provides a better representation of reality than series B.…”