<p>Exchange flows at the water-sediment interface control river water quality and carbon cycling through microbial respiration. However, accurate quantification of these exchange flows and microbial respiration is still challenging in field surveys due in part to the dynamic turbulence generated by streambed topography. Using a framework that combines Structure-from-Motion (SfM) photogrammetry with a fully-coupled surface-subsurface computational fluid dynamics (CFD) model, this work studies the effects of streambed sediment structure on riverbed turbulence and its impact on exchange flows and microbial respiration. Specifically, the SfM photogrammetry is first applied to obtain mm- to cm-scale resolution riverbed topography over a meter scale domain at four sites; these high-resolution riverbed topography data are then used to generate meshes for use in hyporheicFoam, a fully coupled surface-subsurface model developed in OpenFOAM. Simulated time series of water depth and average flow velocity from a previously-developed 30-kilometer scale CFD model will be used to set the water depth and mean flow velocity conditions for high-resolution CFD models of the SfM-characterized locations. The modeling results will be used to investigate the dependence of riverbed exchange flows, concentration gradients, and the concentration profile from the water surface to riverbed on water depth, mean velocity, roughness size, sediment distribution, bed porosity, and subsurface permeability. The relative importance of flow advection, turbulence dispersion, and microbial reaction in both streambed and surface water will also be evaluated.</p>