Are Bedload Transport Pulses in Gravel Bed Rivers Created by Bar Migration or Sediment Waves?

Abstract

Bedload transport exhibits considerable spatial and temporal variability, as reflected by large fluctuations of transport rates. Among the various mechanisms proposed for explaining this variability, bedform migration is often cited as the main cause. We took a closer look at this issue by running long‐duration experiments in a gravel bed flume using constant water discharge and sediment feed rates. We monitored bed evolution and measured bedload transport rates at the flume outlet using high‐resolution techniques. The bed was initially flat, but within a few hours bedforms consisting of alternate bars and pools had developed. The bars exhibited a stick‐slip motion: they were stable for long periods but moved episodically (every 10 hr on average). Their downstream migration produced 36% of the sediment volume transported, mostly in the form of intense pulses. Much of the transport was caused by the displacement of sediment waves from pool to pool.

Impact of Vegetation on Bed Load Transport Rate and Bedform Characteristics

“…Multiple bed load transport rates were measured, and the flow was assumed to reach equilibrium once the cumulative average of all bed load measurements varied by less than 10%, which took 6 to 40 hr depending on the flow condition. We note that in previous studies of very low bed load transport rate (less than 10 grains per (m · s)), temporal fluctuations were observed over much longer time scales, such that the time needed to reach equilibrium may be much larger than 40 hr (Ancey et al, 2015;Dhont & Ancey, 2018). The equilibrium bed load transport rate and its represents the experimental duration in hours, which is the time from the beginning of the experiment to when the last bed load transport measurement was made.…”

The Effect of Episodic Sediment Supply on Bedload Variability and Sediment Mobility

“…Long‐term persistence or long memory is characteristic of processes compatible with the presence of fluctuations on a range of time scales, which could reflect the long‐term variability of several influential factors, and it can be conceptualized as a tendency of clustering in time of similar events (Koutsoyiannis & Montanari, ). In the case of bedload, the scales of fluctuations could relate to the scales of bedform evolution and changes in sediment supply (e.g., bar migration [Dhont & Ancey, ] or step collapse [Saletti et al, ]). Under constant feed, Ma et al () found that bedload rate time series were autocorrelated only at intermediate T but not at very short and very long T .…”

“…If the probability for bedload rate to exhibit a given value is time-independent (no autocorrelation, e.g., white Gaussian noise), or if it depends only on the recent history (short-term autocorrelation, e.g., Markov process), the fluctuation decreases significantly with T. The decrease gets milder as the autocorrelation persists over longer time scales, and if bedload rate depends upon the entire history either because it is nonstationary or because it exhibits long-term persistence, the fluctuation would remain constant regardless of T. Long-term persistence or long memory is characteristic of processes compatible with the presence of fluctuations on a range of time scales, which could reflect the long-term variability of several influential factors, and it can be conceptualized as a tendency of clustering in time of similar events (Koutsoyiannis & Montanari, 2007). In the case of bedload, the scales of fluctuations could relate to the scales of bedform evolution and changes in sediment supply (e.g., bar migration [Dhont & Ancey, 2018] or step collapse [Saletti et al, 2015]). Under constant feed, Ma et al (2014) found that bedload rate time series were autocorrelated only at intermediate T but not at very short and very long T. We expect that large sediment pulses, which introduce significant long-term variability in bedload rate (Elgueta-Astaburuaga & Hassan, 2017), will increase the strength and persistence of autocorrelation in bedload rate time series.…”

Quantification of bedform dynamics and bedload sediment flux in sandy braided rivers from airborne and satellite imagery

“…This represents a significant issue in terms of limiting the availability of data required to understand key links between flow, morphology and sediment transport rates, which are essential for parameterising and validating numerical models. While such experiments have demonstrated that sediment flux in braided rivers may correlate with the migration rate of bars within the channel (e.g., Wickert et al, 2013) or the cyclic erosion and filling of pools (Dhont and Ancey, 2018), this observation remains to be properly validated in the field. Helley-Smith type samplers) or fixed location pit-type bedload traps have been used.…”

“…The most detailed work on sediment flux has thus necessarily been limited to physical experiments where detailed topographic surveys of the entire bed are possible at high temporal and spatial resolution. While such experiments have demonstrated that sediment flux in braided rivers may correlate with the migration rate of bars within the channel (e.g., Wickert et al, 2013) or the cyclic erosion and filling of pools (Dhont and Ancey, 2018), this observation remains to be properly validated in the field. The new technologies described above now provide an opportunity to quantify the spatial distributions of bar and bedform migration rates in the field and thus advance knowledge of how flow, morphology and sediment transport are linked.…”