The temporal and gradational trends of sand infiltration in a gravel bed

Numerical modeling of sand and gravel river beds for fish habitat studies requires understanding and simulating bed mixing processes at finer scales than traditional engineering studies. Despite significant research on sand pulses introduced into a gravel bed, very little experimental data exists with sufficient spatial or gradational resolution to validate a numerical simulation. There are no data on the temporal succession of sand infiltration. Twenty-four flume experiments were conducted in which sand pulses varying in size, gradation and, sometimes color, were introduced into gravel substrates. The gravel beds also varied in size, gradation and mobility. After each experiment, cores were collected and vertically sampled to measure infiltrated sand content and gradation with depth. After nine of the experiments additional cleansing flow phases were also conducted. The sand source was discontinued and clean water was introduced into the flume to remove the sand. Cores were collected to evaluate the efficiency of the cleansing flows. A gradational threshold separated bridging and unimpeded static percolation processes. Samples with filter ratios (d15gravel/d85sand) less than 8 tended to bridge while unimpeded static percolation was observed in samples with filter ratios greater than 12. Regardless of whether the experiments bridged or not, the interstitial deposits of most cores fined with depth due to a variety of processes including granular sorting, bedload sorting and hydraulic sorting. Analysis of the colored sand results demonstrated that bridge layer formation did not definitively preclude subsequent infiltration. Introducing a sand pulse increased the gravel transport capacity and mobile gravel generated more interstitial sand deposition than static gravel experiments with the same materials. Cleansing flows that did not move gravel particles only removed sand from the surface pores and was insensitive to flow increases that did not agitate gravel particles. Minor gravel agitation, however, increased the efficiency of the cleansing flows and sand content was reduced throughout the active layer (3 to 5 d90s) of the mobile gravel experiments