Qiong Zhang, Eric Deal , J. Taylor Perron , Jeremy G. Venditti , Santiago J. Benavides , Matthew Rushlow, Ken Kamrin

Bedload sediment transport is ubiquitous in shaping natural and engineered landscapes, but the variability in the relation between sediment flux and driving factors is not well understood. At a given Shields number, the observed dimensionless transport rate can vary over a range in controlled systems and up to several orders of magnitude in natural streams. Here we (1) experimentally validate a resolved fluid-grain numerical scheme (Lattice Boltzmann Method - Discrete Element Method or DEM-LBM), and use it to (2) explore the parameter space controlling sediment transport \change{}{in simple systems}. Wide wall-free simulations show the dimensionless transport rate is not influenced by the slope, fluid depth, mean particle size, particle surface friction, or grain-grain damping for gentle slopes (0.01~0.03) at a medium to high fixed Shields number. (3) Examination of small-scale fluid-grain interactions shows fluid torque is non-negligible for the entrainment and sediment transport near the threshold. And (4) the simulations guide the formulation of continuum models for the transport process. We present an upscaled two-phase continuum model for grains in a turbulent fluid and validate it against bedload transport DEM-LBM simulations. To model the creeping granular flow under the bed surface, we use an extension of the Nonlocal Granular Fluidity (NGF) model, which was previously shown to account for flow cooperativity from grain-size-effects in dry media. The model accurately predicts the exponentially decaying velocity profile deeper into the bed.