Spatial Variability Of Hydrological Processes And Model Parameters Based On Digital Basin

On the basis of raster DEM data in the Baohe River basin, this paper presents an application of DEDNM to generate drainage network automatically, build digital basin, and extract drainage network to establish the topological river network structure. On the platform of digital basin derived from raster DEM data, - a grid-based hydrological model and a subcatchment-based hydrological model were used to study the effects of spatial variability on runoff process on such aspects as precipitation, model parameters and soil evaporation.Project I, project II and project III that are different in processing precipitation data as the input of hydrological model were built to model day hydrograph and flood hydrograph, so as to analyze the effect of spatial precipitation heterogeneity on runoff process. Project I is the conventional subcatchment method. Project II is modified subcatchment method by interpolating rain-gauge data to each grid, then averaging rain values over all grids within a subcatchment as the input of the ~ subcatchment. Project III is the grid method by interpolating rain-gauge data to a grid element as the input of that element. Those three methods were utilized to simulate daily runoff and flood discharge, respectively. It is concluded that the more sufficiently the spatial rain heterogeneity is taken into consideration, the higher the accuracy of flood modeling is.Raster-based digital elevation data with a spatial resolution of 30 secondscovering the Hanjiang River Basin, were matched spatially with the one-kilometer-grided land use/cover data generated from remote sensing information. The parameter in the Xin'anjiang Model, IMP, was extracted directly from above land use/cover data. Another parameter, SM, in each subcachment and in each raster grid was obtained from the relation between SM and the ratio of forest land area to subcatchment area. Thus, a semi-distributed hydrological model and a distributed hydrological model were established to analyze the effect of some parameters in Xin'anjiang on runoff process according to the spatial variability of land surface characteristics.By introducing a parameter into the three sources evaporation module of Xin'anjiang model, the whole layers nonlinearity method and the middle layer nonlinearity method were applied to evaporation computation to study the effect of spatial soil evaporation variability on runoff simulation. The result shows that not only spatial precipitation heterogeneity and spatial land use/cover variability but also spatial soil evaporation heterogeneity influences flood simulation.As remarked above, spatial variability is very crucial to hydrological modeling. Case Study in the Baohe River basin has shown that considering the spatial variability in study basin is an effective way to improve the performance of hydrological modeling. The impact of human activity on the hydrological cycle is taken into consideration. On the platform of digital basin, the quantitative response of land-use change on hydrological processes is implemented with the help of the correlation between model parameters and land-use within subcatchments, so as toserve for land planning and sustainable development within the Basin.