Mechanistic understanding of colloid retention and transport in porous media is important in many environmental processes and applications. In this study, we characterized and quantified colloid retention under unfavorable attachment conditions through modeling coupled with pore-scale experiments, focusing on the effects of solution ionic strength and interstitial flow speed. A computational approach was developed to simulate the motion of colloids suspended in pore-scale flow through a network of grain packing in a bounded channel. Simulation results showed that colloids could only be retained at the secondary energy minimum (SEmin) due to the presence of the high repulsive energy barrier (above 1500 kT). The fraction of colloids that can move into the SEmin well and be subsequently retained by the attractive van der Waals force is controlled by the competition of hydrodynamic transport along the streamline and Brownian shifting across the streamline. The tangential hydrodynamic force could slowly drive retained colloids toward the rear stagnation region along the surface, leading to accumulation of retained colloids. The retention at SEmin is dynamically irreversible when the SEmin depth reaches about −4 kT. These mechanistic insights explain well the dependence of the retention ratio on flow speed at a given ionic strength as well as the saturation of the retention ratio with ionic strength at a prescribed flow speed. Furthermore, for a given ionic strength, there is a critical flow speed below which Brownian motion dominates the colloid retention rate, leading to a very strong dependence of surface coverage on flow speed. These simulation results were confirmed by experimental confocal-microscopy observations in capillary porous channels as well as by other published results, supporting the mechanistic findings.