Propagules go with the flow: Near‐field particle dispersion in reaches with different hydrodynamic conditions

We examined the effects of riverbed roughness and turbulence (shear velocity, ) on propagule dispersion in the near‐field region (< 100 m) by releasing microbead models of larval and juvenile unionid mussels in tributaries of the Grand River (Ontario, Canada). The Conestogo River had the roughest bed and highest mean , followed by the Grand and Speed rivers. We predicted more downstream transport with higher velocities and that longitudinal dispersion coefficients (Kx) would match the patterns in roughness and . The Conestogo River had the highest downstream particle flux and Kx as predicted by simple empirical equations. Inconsistent with model predictions, however, the Grand River had the lowest particle flux and Kx. These differences were greater than expected based on the small differences in reach‐averaged mean velocities between the Grand and Conestogo rivers. This mismatch between Kx predicted by simple empirical models and those fit to an advection–diffusion model was related to the inertial properties of the flow in the advective zone (i.e., near field) of the reaches. Streamwise, lateral, and especially vertical velocities at drift nets were spatially heterogeneous within and among reaches, demonstrating the influence of the flow direction on particle flux. Although bulk fluid statistics provided a useful indication of how dispersal distances vary among rivers, our results suggest that near‐field dynamics can be complex, requiring high‐resolution bathymetry and velocity data for the development of improved advection–diffusion models. Care should be exercised in predicting the dispersal of particles at ecologically relevant spatial scales in rivers.