Abstract

Bedload transport rates in mountain streams may exhibit wide fluctuations even under constant flow conditions. A better understanding of bedload pulses is key to predict natural hazards induced by torrential activity and sediment issues in mountainous areas. Several processes such as bedforms migration, grain sorting and random particles’ trajectories are evoked as the driving agents of pulse formation and development. Quantifying the effects of these processes is a difficult task. This work aims to investigate the interactions between bedload transport and bedform dynamics in steep gravel-bed rivers. Experiments are carried out in a 17-m long 60-cm wide flume inclined at an angle of 2.7%. The bed is initially flat and made of homogenous natural gravel with a mean diameter of 6 mm. We imposed 200 identical hydrographs (of 1 hr duration) at the flume inlet (the bed surface was not flattened out during these cycling floods). The input hydrograph and the input sediment discharge are nearly triangular. Bed topography is measured after each flood using ultrasound sensors while the bedload transport rate is steadily monitored at the outlet using accelerometers (accelerometers fixed on metallic plates record the impacts of the grains flowing out of the flume). For the sake of comparison, a similar experiment consisting of 19 floods of 10 hours is carried out under constant supply conditions. We show that accelerometers are a cost effective technique to obtain high-frequency bedload discharge data. Spectral analysis of the bedload timeseries is used to highlight the different time scales corresponding to different bedload transport processes. We show that long timeseries are necessary to capture the different processes that drive bedload transport, including the resilience time after a perturbation of the bed. The alternate bars that develop and migrate along the flume are found to significantly influence bedload transport rate fluctuations.

Details

Actions