Statistical description of flume experiments on mixed-size bed-load transport and bed armoring processes

The objective of this paper is to describe the statistical properties of experiments on non-uniform bed-load transport as well as the mechanism of bed armoring processes. Despite substantial effort made over the last two decades, the ability to compute the bed-load flux in a turbulent system remains poor. The major obstacles include the poor understanding of the formation of armor lays on bed surfaces. Such a layer is much flow-resistible than the underlying material and therefore significantly inhibits sediment transport from the reach. To study the problem, we conducted a flume study for mixed sand/gravel sediments. We observed that aggregated sediment blocks were the most common characters in armor layers - the largest sizes resist hydraulic forces, while the smaller sizes add interlocking support and prevent loss of fine material through gaps between the larger particles. Fractional transport rates with the existing of armor layers were measured with time by sediment trapping method at the end of flume. To address the intermittent and time-varying behavior of bed-load transport during bed armoring processes, we investigated the probability distribution of the fractional bed-load transport rates, and the underlying dynamic model derived from the continuous time random walk framework. Results indicate that it is critical to consider the impact of armor layers when a flow is sufficient to move some of the finer particles and yet insufficient to move all the larger particles on a channel bed.