Abstract Bedload transport in mountain streams occurs in highly turbulent flows over rough beds with immobile macroroughness elements such as boulders. These large particles generate complex three‐dimensional flow structures that change considerably the dynamics of turbulence near the bed and redistribute shear stresses, reducing the momentum available for sediment transport. Despite their importance, the effects of macroroughness on sediment fluxes remain highly uncertain, in part due to the challenges of measuring and modeling complex and intermittent transport regimes. To understand how turbulence in low relative submergence conditions controls sediment motion, we perform large‐eddy simulations (LES) over a fixed rough bed with an array of boulders, coupled with a Lagrangian particle model that tracks the motion and interactions of individual grains. This high‐resolution approach resolves the flow at scales relevant to particle entrainment and transport, allowing us to examine both local and large‐scale behavior. Our results show that boulders introduce spatially coherent patterns in bedload flux that persist across averaging scales. Through a multi‐scale analysis, from grain‐resolved Lagrangian statistics to depth‐averaged and reach‐scale transport, we identify the flow variables that strongly correlate with particle transport and clarify the features interacting with grains at the bed. We also evaluate the limitations of threshold‐based formulations and test the performance of classical bedload transport predictors in strongly nonuniform turbulent flows. These findings provide new insights into the mechanisms of bedload transport in these conditions and offer a physically based foundation for improving transport models in natural streams.
Barros et al. (Fri,) studied this question.