CSDMS 2015 annual meeting poster YunchengXu
Morphological impact of large woody debris: Numerical and Experimental Modeling
Xiaofeng Liu, Department of Civil and Environmental Engineering, Institute of CyberScience, Pennsylvania State University, Pennsylvania, United States.
Yong Lai, Technical Service Center, U.S. Bureau of Reclamation, Colorado, United States.
David Smith, U. S. Army Corp of Engineers, Mississippi, United States.
David Bandroski, Trinity River Restoration Program, U.S. Bureau of Reclamation, California, United States.
Abstract:
Sound decisions in stream restoration practices and eco-hydraulics should be based on solid quantitative analysis, either from carefully designed physical modeling or high fidelity computational study. Due to the complexity and uniqueness of each river and stream, realistic representation of the channel and the hydraulic structures in the modeling effort is of critical importance to the creditability of the quantitative analysis result. This paper presents a method to faithfully represent the geometries in both physical and computational modeling using 3D scanning and printing. Large woody debris (LWD) is one of the popular stream restoration designs and it is chosen as an example in this paper. The method demonstrated can be used for other structures too. The trees are first scanned using high accuracy laser scanner and the surface geometry is reconstructed. Then the trees are printed using a 3D printer. The printed trees are consequently placed in the flume to test the hydraulic performance and their effect on flow field, turbulence, air entrainment, etc. In particular, the flow velocity at specific locations around the trees will be measured using an acoustic Doppler velocimeter(ADV). The flow patterns around the trees will be also captured using colored tracer. In parallel, the scanned trees are imported into a 3D computational model which can simulate 3D open-channel flows around complex geometries. The computer model also has the capability to physically and realistically position the artificial trees in the flume. The numerical simulation uses large eddy simulation (LES) for turbulence, and with a combined Volume of Fluid(VOF) method and Level Set(LS) method for free surface. With the precise match of geometries in the physical test and computational model, the numerical results are compared with the experimental data. Ideally, the numerical model should match with the experiment since they are duplicates. Results from this rigorous mutual validation have high confidence level and provide accurate description of the physical process around the reintroduced trees.
* Please acknowledge the original contributors when you are using this material. If there are any copyright issues, please let us know and we will respond as soon as possible.
