Geospatial description of river channels in three dimensions

Data describing the three-dimensional structure of river channels are required for hydrodynamic simulation, water quality modeling and flood inundation mapping. However, unlike land surface terrain, there is no standardized three-dimensional geospatial representation of river channel morphology. In addition, the bathymetry data that are used to describe the threedimensional structure of river channels are collected only on short reaches, and are usually not available for the entire stream network at a regional scale. The research presented in this dissertation deals with the standardized geospatial description of river channels in GIS at both local and regional scales. At a local scale, where the bathymetry data are available, a procedure is developed that uses the curvilinear orthogonal coordinate system approach to create a threedimensional structure of river channels in the form of cross-sections (ground profiles transverse to flow) and profile-lines (ground profiles parallel to flow). To describe the channel bathymetry at regional scale, an analytical model is developed that uses locally collected field data on the Brazos River in Texas and readily available regional scale data such as the hydrography data. The analytical model, called the River Channel Morphology Model (RCMM), is based on deriving relationships among different channel characteristics such as the channel planform (shape of the channel), the thalweg location, and the cross-sections. Then, using only the channel planform, in conjunction with the derived relationships, it is possible to create a three-dimensional channel morphology that comprises a mesh of cross-sections and profile-lines. The three-dimensional mesh at the regional scale, however, provides only a mean surface for the channel bed. Although the results from RCMM are promising, it is limited by its ability to model all the variables that are involved in the development of complex river geometry. Bathymetry data collected on the Brazos, Guadalupe, and Sulphur Rivers in Texas are used to demonstrate the applicability of the theories presented in this dissertation. It is learned that creating an accurate description of river channel geometry is a complex problem, which requires knowledge about channel planform, discharge, geology, geomorphologic features such as pools and riffles, and small-scale stochastic variations to solve.