<p>Coarse bed load transport is a crucial process in river morphodynamics, but it is difficult to monitor in mountain streams. Predicting bed load is a difficult task especially in steep step-pool streams, where the critical dimensionless shear stress is affected by local channel slope and relative submergence, and only part of the flow energy is available to entrain and transport sediments as some is dissipated in local hydraulic plunging and jumps. Here we present a new sediment transport dataset obtained from two years of field-based monitoring (2014-2015) at the Estero Morales, a high-gradient stream in the central Chilean Andes. This stream features step-pool bed geometry and a glacier-fed hydrologic regime characterized by abrupt daily fluctuations in discharge. Bed load was monitored directly using Bunte samplers and by surveying the mobility of passive integrated transponder (PIT) tags. We used the competence method to quantify the effective slope, which is the fraction of the total slope responsible for bed load transport. This accounts for only 10% of the total slope, confirming that most of the energy is dissipated on macroroughness that characterize step-pool stream. We used the displacement lengths of PIT tags to derive the statistics of flight and resting times, observing that the average length of a flight scales inversely with grain size.</p>