Quasi-two-layer morphodynamic model for bedload-dominated problems: bed slope-induced morphological diffusion

TL;DR

This paper introduces a two-layer depth-averaged sediment transport model that accounts for bedload and morphological diffusion effects, validated against empirical data and improving predictions of bed evolution.

Contribution

A novel two-layer model incorporating bedload transport and bed slope-induced diffusion, validated with empirical data and applied to mining pit evolution.

Findings

Model accurately predicts sediment transport rates.

Inclusion of morphological diffusion improves bed evolution predictions.

Validated against empirical and experimental data.

Abstract

We derive a two-layer depth-averaged model of sediment transport and morphological evolution for application to bedload-dominated problems. The near bed transport region is represented by the lower (bedload) layer which has an arbitrarily constant, vanishing thickness (of approximately ten times the sediment particle diameter), and whose average sediment concentration is free to vary. Sediment is allowed to enter the upper layer, and so total load may also be simulated, provided that concentrations of suspended sediment remain low. The model conforms with established theories of bedload, and is validated satisfactorily against empirical expressions for sediment transport rates and the morphodynamic experiment of a migrating mining pit by Lee et al. (1993). Investigation into the effect of a local bed gradient on bedload leads to derivation of an analytical, physically meaningful expression for morphological diffusion induced by a non-zero local bed slope. Incorporation of the proposed morphological diffusion into a conventional morphodynamic model (defined as a coupling between the shallow water equations, Exner equation and an empirical formula for bedload) improves model predictions when applied to the evolution of a mining pit, without the need either to resort to special numerical treatment of the equations or to use additional tuning parameters.