Effect of suspended sediment size distribution on Lower regime bedform length in alluvial channels Open Access
Downloadable ContentDownload PDF
AbstractIn the present thesis, an attempt is made to develop and apply an analytical scheme based on computational fluid dynamics to advance the understanding of certain aspects of the interaction between suspended sediment load and bedform geometry. A large part of the thesis is concerned with the solution of the boundary-value problem associated with the convection-diffusion of heterogeneous suspended sediments over two dimensional ripples and dunes. The numerical model of Haque and Mahmood for the solution of steady state convection-diffusion equation over 2D bedforms, for uniform suspended sediments, was modified for application to the case of non-uniform suspended sediments. The Rotational Invicid Flow (RIF) model is used for the solution of flow problem. Numerical simulations were made, with a fixed dune geometry and flow rate for various combinations of sediment size distribution and mean size value of the sediment mixtures. The convection diffusion solution is obtained for each size fraction of sediment within the mixture and later on combined to get a concentration field for the mixture. The diffusion coefficient (E*), pertaining to the marginal solutions for non-uniform suspended sediments, are numerically computed (based on finite element method) by solving steady state convection-diffusion equation with complex boundary conditions, over ripples or dunes. To investigate the influence of sediment size distribution on the bedform length, non-dimensional fall velocity (W) versus plots have been developed. A parametric study was also carried out for various combinations of D/L ratio, sediment size distribution and mean size of the mixture. Based on the data obtained from the numerical experimentations a model is developed relating bedform length (L), mean dimensionless fall velocity (W), and standard deviation ( ) of the mixture. For validation purposes the model equation is applied for computing bedform length, and the predicted results are compared with the field and laboratory data. In a comparative analysis involving several existing relations, the proposed method was found to give better agreement with the observed bedform lengths, except in the case of laboratory data.