I have been trying to understand morphological dynamics of the Braided rivers (Brahmaputra) using 2D/3D models. The main hurdle is the unavailability of detailed bathymetry information for braided reaches of the river. When I use 2D models such as CCHE-2D, it frequently crash when simulated with coarse bathymetry data. Is there any way I can improve my modeling strategy to get better results over the Braided reaches using remote sensing or any other information.
If you consider that the trench of stream is in dynamic equilibrium , there is certain general conditions that may be stated by means of tracers. Thouhg this view is not detailed it can help to guide some hydraulic and morphologic insights to model it properly.
With the help of remote sensing technique you can comfortably carry out this model either in 2 or 3 dimensions. First obtain a natural color composites satellite imageries of the area from which you can trace the bathymetric shape of the river. Also a 3-D digital elevation model (in case of a 3D model) of the area can help you determine the elevations across the river (this can be obtained from google earth on special request ). You then divide your area of interest into grids and pick the elevations (levels) base on the color chart at various interceptions. Reduce the levels based on your local datum assuming you have a reference point within the study area. With the help of the spot heights (levels), grid numbers (coordinates) you can plot a map show the contour of the area and also pick the bathymetric shapes at different water levels/layers (depending on how many layers you want to mesh for a 3D model). After meshing, you can assume some boundary conditions like upstream and downstream velocities, slope and others if they are not available. Using finite difference method you can couple the model in 2 or 3 dimensions without visiting the site. Although you can update when you eventually have access to the required information and or parameters required. The result using this procedure can give close to 80% correlation with field data.
River cross-sections across the Brahmaputra from Pasighat to Dhubri taken at 10 years' interval may be available with the Water Resources Department of Govt. of Assam. That may help in developing the mesh of desired resolution required by your software. If 3/4 sets of data spanning 3/4 decades for your concerned stretch of the river can be collected, it can be helpful to understand the complex morphological dynamics of the Brahmaputra. I am not sure if by limited bathymetry information you mean that you are already in possession of the said data. From the data-availability perspective, bathymetry data is more critical than the other two elements affecting the morphological dynamics of a braided river viz. flow and sediment.
The remote sensing alternative would be to go for Bathymetric LiDAR data which, I am afraid, may not be readily available.
The availability of sufficiently detailed bathymetrical data is often a problem in morphodynamic modeling. At the same time, many braided rivers are very dynamic and any bathymetrical data are outdated within short time. In a recent study of the Yellow River, and in recent publications, we only defined the outer bank lines and an initially flat bed. The model (Delft3D) constructed a detailed bathymetry with bars and channel branches. This approach worked very well, and gave good insight in the river system and its dynamics. If you like I can give you more details about it.
- Badges
- Science topic
- Similar topics
- Computer Communications (Networks)
- Simulators