Theoretical Study And Application On The Framework Of Digital Watershed Model

The digital watershed model can simulate the process of rain-runoff, soilerosion and sediment transportation in a large drainage basin, which is veryimportant in flood control, water regulation and the conservation of soil and water.Up to date, the research of the digital watershed model is still in the initiatory periodand the range of research is restricted into the small drainage basin. The aim of thepaper is to construct a holistic framework of digital watershed model system.The framework of digital watershed model is based on the model base, which isconstructed by means of distributed hydrological model theory. Under theframework, the whole river basin is classified into three levels: the overall watershedmodel, the sub basin model and the small catchment model. The overall watershedmodel is composed of some sub basin models and one sub basin model is composedof some small catchment models. The small catchment model is the fundamental unitin the framework. The models of different levels are connected through the commondatabase instead of the direct data exchange, which makes every model independentfrom each other. The independent model could easily be modified and upgraded. Themodels are developed by means of object oriented programming (OOP) method,which makes the model have good expansibility. There are four kinds of objects: thesmall catchment, the drainage trenches, the reservoir and the irrigated area. Thewhole river basin model can be built based on these four kinds of objects. Forinstance, this thesis has built the rain-runoff model in Xiaolangdi-Huayuankou basinof Yellow River.The rain-runoff model has been validated in Luohe River basin by using theobserved hydrological data from July 20th to August 20th in 1980. The applicationof the developed digital watershed model in the Xiaolangdi to Huayuankou Reach ofthe Yellow River shows that the simulation time-step of the model can be adjusted tomeet with different computation requirements: short time-step (6 min) forevent-based streamflow simulation and long time-step (1 hour) forcontinuous-precipitation-based streamflow simulation. Combined with therainfall-predict module, the model can be used for flood prediction. In addition, themodel can also be used for the sediment yield simulation in the small catchment ofthe sandy and grit area in Loess Plateau. As an example, the suggested modelsimulates the hydrograph of the water and sediment discharge rate in Chabagoubasin, which is a small catchment in the sandy and grit area. The well fit simulationresults can supply decision support for check-dams planning.