Numerical Modelling for Flow Simulation of Large Alluvial River Systems

Abstract

The complex morphological nature of alluvial rivers and their effect on the well, being of the human race settled along and around the system has necessitated a wide range of research interest; Research interests that are driven by a desire to solve and harness the erratic and sometimes destructive nature of these rivers. These desires: to understand the physical processes occurring in rivers have lead to the development of physical and mathematical modelling methods. The concept of numerical modelling for flow simulation has found large acceptance over the past decades as a means to simplify and solve these problem. An essential and well-known feature of alluvial channels is that their morphology, flow-resistance, and sediment-transport characteristics adjust in response to hydrodynamics of stream flow and sediment transport. Alluvial streams carry extremely varying discharge and sediment loads. Effective analysis of river problems requires recognition and understanding of the governing processes in the river system. In the present study, it is attempted to model and simulate the Brahmaputra River with multidimensional mobile bed numerical models. HEC-RAS 4.1 is used for one dimensional modelling and.CCHE2D for two dimensional modelling. The study area covers 622km reach of BrahmaputraRiver from cross section 2 (Dubhri) to cross section 65(Kobo). Hydrographic data of the year 1988, 1992, 1993 and 1997 and simulation period of Nov, 2003 to Sept, 2007 is adopted. The drainage-area ratio method has been found out to be useful tool in transferring gauged flow data to ungauged flow stations. Sediment transport analysis performed with HEC-RAS has broadly indicated areas with aggradation and degradation. Taking threshold values of ±5cm thalweg level change for the simulation period of Nov, 2003 to Sept, 2007; it has been observed that, generally, 65% of the' 64 cross sections analyzed are exhibiting fairly moderate bed change over the over the study period, with sizeable areas depicting aggradation, notwithstanding that the mobile bed variation of HEC-RAS 4.1 is oriented to simulate sediment transport capacity for a given L __- river cross-section. As per known information on the prototype, there is general trend river of bed aggradation and river bank erosion. It has been clearly observed that when there is relatively lower sediment transport capacity compared to sediment inflow, at given cross section during a computational time step, occurrence of aggradation the most likely phenomena. Likewise for higher sediment transport capacity degradation is resulting phenomena. This phenomena is also been seen to have brought about complex interaction of stream power, shear stress of the movable portion of the bed and velocity among others. High resolution mesh is generated for 622km reach of the river that is necessary for two dimensional numerical model simulation platforms. The two dimensional model better shows the transverse aggradation and degradation phenomena compared to one dimensional simulation; but cannot be validated easily as compared to one dimensional model. Hybrid modeling is attempted in the Gandak River study. One dimensional HECRAS 4.1 model and PHYSICAL MODEL are used. The study is supplemented by the application of. satellite based spatio-temporal variation of channel migration in the vicinity of study area. Major channel migration occurred between the year 1972 to 1989 The channel has shifted its course from north to south ward direction, where the subsequent shifting and bifurcation are occurring. The physical model is run for two scenarios, one with clear water condtion and the other for sediment concentration of 2000ppm. Model dimension on the ground is 25m by 12m. Geometric scale of Lr =400 and ZI= 50. With clear water simulation the model is verified against observed stage and discharge. The study shows from that deposition has occurred in the middle of the channel between the abutments and erosion while scouring occurred at the toe of the guide bank. And it is also evident that the left guide bank is under more attack due to the secondary flow generated at the upstream section of the guide bank. To prevent the damage that may occur on the, guide bank and to channelize the flow to the centre of the river, submerged vanes are proposed.