Rill Erosion Dynamic Processes Studied

STUDY ON THE DYNAMIC PROCESS OF RILL EROSION ZHANG Qingwen Directed by Prof. LEI Tingwu (State Key Laboratory of Soil Erosion and Dryland Fanning on the Loess Plateau, Institute of Soil and Water Conservation, CAS and Ministry of Water Resources, Yangling, Shaanxi 712100) Abstract The process-based prediction model is one of the most effective ways to research soil erosion, and lays scientific foundation for soil and water conservation practice. It is of importance to study on the physical mechanism of nh erosion under different hydrodynamic conditions in order to help physical process-based prediction model. The existing process-based soil erosion models are conceptually but not fully in reality based on the physical process. Many parameters, such as 77~ (transport capacity), are just conceptual expressions, with no strong physical proof, and nor experimental verification. To address the process-based prediction model full physically, parameters in the models should have clear physical meaning, and be presented by a mathematic functions, or be determined experimentally. In considering the fact that, under steady flow, sediment yield increases with nil length and finally approaches its maximum value the sediment transport capacity, a laboratory flume experimental method for transport capacity determination through sediment yields at variable slope lengths was advanced. A functional relation of the sediment loads with nh lengths was obtained by measuring the sediment loads of steady-state flow water at (a series of) various nil lengths, in order to determine the transport capacity and the quantitati~抏 relationship of the detachment rate in eroding nh. Using a typical silt-clay soil from the Loess Plateau, a series of 405 flume experiments were conducted, on slopes of 50 , 100 , f50 , 20~ and 250 , flow rates of 2, 4, 8 L/min and nIl lengths of 0.5, 1, 2, 3, 4, 5, 6, 7, 8 m, with 3 replicates for each. The sediment loads of steady-state flow water at (a series of) different nIl lengths were measured and a corresponding relation between these two was then after determined. A proper mathematical function to fit the relationship between sediment loads and rill lengths in the experiments was founded to be: C A (1 e~x)B A method and the mathematical expression were advanced to estimate transport capacity with thus obtained experimental data. With a given slope and discharge rate, sediment concentration could still not come to a stable value even at the flume length. In cases like this, mathematical function can be used to fit the relationship between sediment loads and nIl lengths. And the sediment transport capacity S can thereafter be estimated by the limit of the fitted function within some given error range, such as 5% or 10%. Transport capacity increased with in bed slope and less significantly increased with increase in flow rate. Computed transport capacity showed that slope played more important role in sediment transport capacity than inflow rate did, and 200 was to be a critical slope of the bess used. The transport capacity increases slightly at slope over 200. Detachment rates showed a good exponential relationship with nil length and linear relationship with sediment load.