The Mechanism of Large Woody Debris Initial Entrainment in a Flume Experiment

• Main Authors: Ci-Rong Wang, 王啟榮
• Other Authors: 陳樹群
• Format: Others
• Language: zh-TW
• Published: 2010
• Online Access: http://ndltd.ncl.edu.tw/handle/98707309248305712100

碩士 === 國立中興大學 === 水土保持學系所 === 98 === There has thus far been relatively little research into large woody debris (LWD) motion. The research was designed as a fixed bed flume experiment to explore the state of LWD critical entrainment, and discussed the initial flow condition was influenced by different length, diameter, orientation and bed roughness. Additionally, the theoretical model was developed to predict wood entrainment and compare predictor with experiment results. The experiment results show that different orientations of wood will lead to different types of wood movement. For wood parallel to flow, the mechanism of motion features started by semi-floating and semi-sliding, because flow was gradually raised with the buoyant force increased and the friction force decreased until the wood moved to downstream. For wood oblique or perpendicular to flow, the mechanism of motion features started by rolling. Both the models and the experiments indicate that stable wood is significantly associated with wood angle relative to flow direction, the density of wood, wood diameter, channel slope and bed roughness. The wood stability is less sensitive to the choice of the apparent drag coefficient and wood length. Although previously reported as the most important factor in wood stability, wood length did not significantly affect the threshold of movement in the experiments or the model predictions, for wood shorter than channel width. This research consists of sliding and rolling equilibrium equations to establish a prediction model of wood entrainment: where dw is the flow depth for wood incipient motion, Llog is the wood length, Dlog is wood diameter, ρlog and ρw are the densities of wood and water, respectively, S is the channel slope (S=tanα), CD is the drag coefficient of the wood in water, ds is the bed grain size, θ is the angle of the wood relative to flow, and is the friction angle between wood and channel bed. This prediction model is helpfully used for multiple angles of the wood relative to flow, and the flow depth for wood incipient motion could be predicted under the wood and channel characteristics are given. To compare model predictors with experiments, the bias account for about 10% of the wood diameter. Moreover, for wood oblique or perpendicular to flow, the bias only account for about 5% of the wood diameter. According to the results, the prediction model is relative successfully at predicting depths for wood incipient motion.