The nature of saltation and of ‘bed-load’ transport in water

Abstract Owing to observational difficulties the distinction between a ‘suspended’ load of solids transported by a stream and a ‘ bed-load ’ has long remained undefined. Recently, however, certain critical experiments have thrown much light on the nature of bed-load transport. In particular, it has been shown that bed-load transport, by saltation, occurs in the absence of fluid turbulence and must therefore be due to a separate dynamic process from that of transport in suspension by the internal eddy motion of a turbulent fluid. It has been further shown that the forward motion of saltating solids is opposed by a frictional force of the same order as the immersed weight of the solids, the friction coefficient approximating to that given by the angle of slip. The maintenance of steady motion therefore requires a predictable rate of energy dissipation by the transporting fluid. The fluid thrust necessary to maintain the motion is shown to be exerted by virtue of a mean slip velocity which is predictable in the same way as, and approxim ates to, the terminal fall velocity of the solid. The mean thrust, and therefore the transport rate of saltating solids, are therefore predictable in terms of the fluid velocity close to the bed, at a distance from it, within the saltation zone, of a ‘centre of fluid thrust’ analogous to the ‘centre of pressure’. This velocity, which is not directly measurable in water streams, can be got from a knowledge of stream depth and mean flow velocity. Thus a basic energy equation is obtained relating the rate of transporting work done to available fluid transporting power. This is shown to be applicable to the transport both of wind-blown sand, and of water-driven solids of all sizes and larger than that of medium sand. Though the mean flow velocity is itself unpredictable, the total stream power, which is the product of this quantity times the bed shear stress, is readily measurable. But since the mean flow velocity is an increasing function of flow depth, the transport of solids expressed in terms of total stream power must decrease with increasing flow depth/grain size ratio. This considerable variation with flow depth has not been previously recognised. It explains the gross inconsistencies found in the existing experimental data. The theoretical variation is shown to approximate very closely to that found in recent critical experiments in which transport rates were measured at different constant flow depths. The theory, which is largely confirmed by these and other earlier experiments, indicates that suspension by fluid turbulence of mineral solids larger than those of medium sands does not become appreciable until the bed shear stress is increased to a value exceeding 12 times its threshold value for the bed material considered. This range of unsuspended transport decreases rapidly, however, as the grain size is reduced till, at a certain critical size, suspension should occur at the threshold of bed movement.