Doctor of Philosophy
Department of Civil and Mining Engineering
Moerwanto, Arie Setiadi, A riverine finite element model with sediment transport sub-processes, Doctor of Philosophy thesis, Department of Civil and Mining Engineering, University of Wollongong, 1997. http://ro.uow.edu.au/theses/1220
The purpose of this study is to design a one-dimensional numerical morphological model of a river. A study of the coupled and the decoupled solution methods used to link the hydrodynamics and the sediment morphology is also carried out. This study culminates in the development of both model formulations, in which the finite element method is implemented to solve the governing equations. The models are coded in a modular format to allow modifications to the computer program to be made relatively easily.
In developing a model which is capable of simulating various sediment transport sub-processes, the following aspects are included in the present model:
i) the adoption of the dynamic wave approach by retaining all the terms in the momentum equation, ii) the explicit separation of the bed load and the suspended load transport, iii) the vertical exchange of sediment and armoured layer development and a means of handling non-uniform distribution of bed material, and iv) incorporating the spatial and temporal lag effects of bed load transport as a consequence of the dynamic wave approach.
The following characteristics of the present model are evident in the test results:
i) The present model is simple and robust. It can cope with a wide range of field problems. ii) The formulation of the present model is simpler and requires less computer storage than the Holly and Rahuel morphological model. However, the results from simulating the deposition and redistribution of the deposited sediment down the channel are comparable to those from the Holly and Rahuel model. iii) The upwinded FE scheme in the present model is used to solve the advection-dispersion equation for the suspended sediment but is restricted to a maximum Courant number of about 2 in spite of its fully implicit formulation
The present study demonstrates that the temporal lag effect of bed load transport is a phenomenon that significantly reduces the celerity of bed disturbances during the flashy flows. Due to the disparity between the hydrodynamic celerity and the morphological celerity, this result widens the applicability of the decoupled solution approach.