Fine particle deposition to initially starved, stationary, planar beds

Abstract

Reported here are results from new flume experiments examining deposition and entrainment of inert, silt-sized particles (with spherical diameters in the range from 20 to 60 lm) to and from planar, impermeable and initially starved beds underlying channel flows. Bed surfaces comprised smooth or fixed sandsize granular roughness and provided hydraulically smooth to transitionally rough boundaries. Results of these experiments were analysed with a simple model that describes the evolution of vertically averaged concentration of suspended sediment and accommodates the simultaneous delivery to and entrainment of grains from the bed. The rate of particle arrival to a bed diminishes linearly, and the rate of particle entrainment increases by the 5/2 power, as the value of the dimensionless Saffman parameter S = u * 3 /g'm approaches a threshold value of order unity, where u * is the conventional friction velocity of the turbulent channel flow, g' is the acceleration due to gravity adjusted for the submerged buoyancy of individual particles and m is the kinematic viscosity of the transporting fluid. This transport behaviour is consistent with the notion that non-cohesive, silt-sized particles can neither reach nor remain on an impermeable bed under flow conditions where mean lift imposed on stationary particles in the viscous sublayer equals or exceeds the submerged weight of individual particles. Within the size range of particles used in these experiments, particle size and the characteristic size of granular roughness, up to that of medium sand, did not affect rates of dimensionless arrival or entrainment to a significant degree. Instead, a new but consistent picture of fine-particle transport is emerging. Silt-sized material, at least, is subject to potentially significant interaction with the bed during intermittent suspension transport at intermediate flow speeds greater than the value required for initiation of transport (ca 20 cm sec )1 ) but less than the value (ca 50 cm sec )1 ) required by the Saffman criterion ensuring transport in fully passive suspension or, equivalently, 'wash-load'.

Quantifying bed‐related suspended load in gravel bed rivers through an analysis of the bedload‐suspended load relationship

“…Similar observations were made at the reach scale in the Isabena catchment where changes in fine sediment stocks were correlated with the maximum flood discharge [Buendia et al, 2016]. The strong interactions observed in these field studies between suspended load and the river bed were also demonstrated in several flume experiments [Glasbergen, 2014;Hamm et al, 2009;Krishnappan and Engel, 2006;Mooneyham and Strom, 2018]. Navratil et al (2010) estimated these stocks in a typical mountainous braided river by considering stocks in dried braided channels, bars and vegetated bars in addition to sediments stored in the wetted channel.…”

Suspension flow and sedimentation in self-affine fractures

“…Typical experimental studies focus on breakthrough curves rather than the details of transport. 8,9 There are some recent studies for the case of fine colloidal particles which have similarities to the present work, for example, in the continuum modeling of Hamm et al 10 and Chen et al, 11 the experimental measurements of the effects of varying bed roughness by Mazumder et al, 12 and the molecular dynamics calculations of combined sedimentation and aggregation by Whitmer and Luitjen, 13 there is little similar work for larger particles. Although we are able to relate some particular features of our results to previous work below, we particularly regret the absence of detailed particle-level experimental studies which would allow us to directly test our results.…”

A laboratory study of dynamic sorting of fine, non-cohesive particles by steady, unidirectional currents

“…Silt-sized material thus in motion was in well-mixed yet dilute suspension transport {but, after Hamm et al (2009), with significant ongoing bed exchange and potential for sorting by way of selective deposition and selective-re-entrainment}. To demonstrate graphically that our study suspensions were vertically well-mixed, we show in Fig.…”

“…In a more recent laboratory study of the transport behavior of non-cohesive, well-sorted, unimodal silt suspensions in recirculating channel flows, Hamm et al (2009) observed that the rate of particle arrival to a bed diminishes linearly, and the rate of re-entrainment of particles newly arrived at the bed increases nonlinearly and to a dramatic degree, with bed shear stress. Importantly, in the context of sortable silt, the characteristic size of silt in dilute suspension and the characteristic size of granular roughness of the bed, up to that of medium sand, did not affect rates of particle arrival or re-entrainment (once adjusted to accommodate the characteristic still-water settling velocity of particles used in a given experiment) to a significant degree.…”

“…On the other hand, there is equally compelling laboratory evidence that flow-driven sorting of silt by size during deposition from dilute suspension and re-entrainment of newly-deposited particles does not occur. In the interest of full disclosure, the experiments of Hamm et al (2009), which failed to detect dynamic sorting of silt, were not conducted with the sortable-silt hypothesis in mind. Indeed, the lack of dynamic sorting of silt by grain size in simple channel flows came as a surprise to us.…”