Intense Transport Of Non-cohesive Bedload Sediments By Steady Currents Or Asymmetric Waves

Intense transport of non-cohesive bedload sediments often occurs in rivers in flood, in estuarine and coastal waters in storm conditions, causing significant morphodynamic evolution, rapid siltation of harbors, navigation channels and other problems. Therefore, it is of great importance to be able to predict bedload transport rate in strong currents or big wave regime with high accuracy for the coastal engineering community.Since Du Boys in 1879 suggested drag-force model and related bedload transport rate formula, investigators have proposed possibly over one hundred bedload formulas using various approaches. Nonetheless, the majority of these formulas are empirical and semi-empirical with fixed coefficients that were regressed from a limited number of experimental and field data. This has seriously limited the application scope of such formulas. Some investigators tried to build mathematical models by treating the intense transport of mixture consisting of bedload particles and inter-particle fluid as Newtonian fluid or other types of fluid based upon continuum mechanics, and derived formulas for calculating particle velocity, bed thickness, etc. However, most of the results could only be claimed qualitatively correct and the few transport rate formulas derived in this way have not been accepted by the engineering community.The reasons are, on one hand, these theoretical models only analyzed the force exerted on a single particle and follows its movement tract from a microscopic perspective with the molecular kinematic theory, this led the models to be unable to take account into the effect of the intense interparticle collisions upon transport behavior of particle assemblage; on the other hand, these models often incorrectly treated Bagnold's (1954) stress relation as relation between water flow shear stress and the normal dispersive pressure arising from interparticle collisions.In addition, the existing empirical, semi-empirical transport rate formulas could not effectively take into account the effect of bed slope, wash-load concentration in driving water, and other factors on bedload transport rate.In view of these, the present study attempts to build a theoretical model to describe the intense transport of non-cohesive bedload sediment using system theory, holism theory and continuum mechanics and fully considering the forces acting upon the upper and lower boundary surface of the sediment-water mixture as it moves.The present study is conducted through theoretical analysis supplemented with bedload transport flume experiment and numerical experiments. The major results are:1. Has built a theoretical model (partial differential equations, PDEs) to describe the intense transport of bedload sediment-water mixture. The model consists of a mass conservation equation and a momentum equation, belonging to a nonlinear hyperbolic conservation system.2. Has sought the special solution to the model that corresponds to the "plane-bed regime" in bedload transport processes. Using a 'UCH' approach this solution is transformed into a 1D theoretical bedload transport rate formula. Its coefficient appears to be a dynamic function that could generically take account into the hydrodynamic and sediment property, in particular of the effects of bed slope (positive and negative), wash-load concentration in the dilute water column upon the bedload transport rate. After specifying the particle velocity and concentration profiles in the mixture layer and how to calculate the dynamic frictional angle of particles, etc. the 1D theoretical formula can be used to predict bedload transport rate. It compares very well with large flume experiment and field observation datasets. Worthy mentioning, the present study suggests a formula to determine the particle velocity at the top of bedload layer on the basis of logarithmic velocity distribution in a current boundary layer. It is calibrated with extensive flume experimental data.3. Using finite difference numerical scheme-Weighted Average Flux method (WAF) that is proficient in solving Riemann discontinuity problem, the present study successfully solved the PDEs. Through prescribed numerical experiments in which a steady flow of sediment-fluid mixture is disturbed by a hump upon an incline, the present study discusses the stability of the mixture flow. It finds: the surface of mixture driven by a steady current will lose stability due to the local disturbance, and dynamic waves appear and evolve downstream with growing wave height and wave length but diminishing wave speed until they break. The numerical experiments disclose the occurrence of interface wave or internal wave in stratified flows consisting of shearing water flow and the high-concentration mixture flow once the latter is disturbed locally. The numerical code was executed in MATLAB environment.4. Assuming a logarithmic velocity distribution in the wave boundary layer, the present study obtains a formula for determining the wave friction factor that is used to compute the Shields parameter. This formula has advantages of solid physical basis, simple structure and easy usage in comparison with other existing formulas. Based upon the special solution of the PDEs the present study obtains a 1D bedload formula to predict the net transport rate in a wave cycle in sheet-flow regimes driven by asymmetric waves. This bedload formula compares well with the experimental data obtained from a large-scale wave flume (Grober WellenKanal, GWK of the ForschungsZentrum Küste, Hannover) and a large-scale oscillatory water tunnel (Delft Hydraulics) as well as field observation in the Myalup beach, Australia.