Study On Hydrological Characteristics Skeleton Data Model And Distributed Spatial Discretization In Plain River Network Region

Distributed watershed model is widely used in hydrological process simulation, which is an indispensable tool of scientific understanding and reasonable solution to the water environmental problems and water resources. But the distributed watershed model is greatly restricted in plain river network region. The main problem is that the existing methods are difficult to achieve resonable division of the catchment and construction of topological relations between catchments in plain river network region. Solving this bottleneck can effectively improve the applicability of the distributed watershed model and be significant to the rational use of water resources and water environmental governance in plain river network region.Western area of Taihu Lake Basin is selected as the study area. In consideration of the reticular river network, dotted lake and reservoir and numerous hydraulic facilities in plain river network region, the dissertation build a hydrological characteristics skeleton data model by using of multi-source data of remote sensing image and thematic maps. Then a new spatial discretization method by considering the typical hydrological characteristics such as cross, looping river network as well as lakes, reservoirs and polder is proposed. And a spatial discretization system which can be used for distributed watershed modeling in plain river network region is developed.The main contents and results are as follows:(1) A data model to express the plain river network region hydrological characteristics skeleton was constructed. The hydrological structure is abstracted into point, polyline and polygon based on the analyzing of the hydrological structure and characteristics of plain river network region. Point, includes hydrological node such as reach nodes, inlets and outlets of lake and reservoir, hydrologic stations, etc. Polyline, includes reach and dyke. Polygon, includes lakes, reservoirs and polder. Object-oriented approach is designed to express the data structure of the spatial topological relations of hydrological structure. And Geodatabase is chosed to store the hydrological characteristics skeleton data model.(2) A new method of distributed spatial discretization based on hydrological characteristics skeleton data model in plain river network region was proposed. River flow direction detection algorithm, multiple flow direction coding algorithm and local terrain flow correction algorithm are designed to divide the discrete units include subbasin, lake, reservoir and polder and build the spatial topological relations of the discrete units by using real river and water area data in plain river network region. This method takes into account the non-uniformity of plain river network region and complex confluent and separated flow in river network. (3) Plain river network region distributed spatial discretization system is developed by ArcGIS secondary development platform. The system implements the construction of the hydrological characteristics skeleton data model, grid direction extraction based on DEM and hydrological characteristics skeleton data model, spatial discrete units division and construction of basin topological relations in plain river network region which provides an important foundation to the distributed watershed modeling.