Research On Hydological Modeling In High Mountian Basins

High mountain basins are the “water tower” of the word. Most of the longest rivers in the word are derived from thesebasins.Water resource in high mountain basins is very important to the global water resource management.However, hydrological processes in such basins owns lots of special characteristicsdue to the complex hydrometeorological and topographic conditions.This study focuses on the below four characteristics of hydrological processes in high mountain basins: The spatial variability of precipitation is rather strong; Runoff generation conditions are complex, the spatial variability of runoff components in basin discharge is significant; Snowmelt greatly contributs to the total runoff; The landscape characteristic varies strongly in space, and produce diverse runoff generation conditionsacross the basin.And then we develop corresponding methods to improve the modeling performance.Combining remote sensing precipitation product with gauged data based on the spatial pattern of precipitation in high mountain basins: TRMM satellite precipitation products provide the spatial pattern of precipitiaton. We used the differencesin the precipitation values between pixels to describe the spatial variability of preicitation. By combining the gauged data from the sparse network with the TRMM precipitation product, we developed a method to obtain better estimation of precipitation field. The combining method for precipitation product helped to improve the simulation of peak flows in the Lancangjiang basin.Step-wise calibration of model parameters based on hydrograph partitioning: According to the decreaingof temperature with the increasing of elevation and the different lowest elevation bands for snow and glacier cover, we developed some indices to identify the seasonal runoff components in high mountian basins, and partitioned the hydrograph into several subsets based on the switch of the dominant runoff components. Each subset was related to a subset of model parameter on the basis of the parameters’ physical basis. And model parameters were calibrated on different hydrograph subsets in a stepwise way. The proposed stepwise calibration method had been applied in the Tailan basin, Xinjiang, China and yielded considerable good performance.Estimating degree-day factor for snowmelt based on the process information during the accumulation and ablation of snow cover: According to the controlling of temperature on the seasonality of snow cover, we divided the snow cover series into three sub-periods:the accumulation period, the ablation period and the combination period. We estimated the snow density in the accumulation period based on the snowfall water equivalent gauged by precipitation data. The estimated snow density was then applied in the ablation period to estimate the snowmelt water equivalent for the estimation of degree-day factor subsequently. The estimatingmethod had been applied to the Lienz basin in Austria, and improved the simualiton of snow cover pattern significnatly.A vari-structure hydrological model based on the topgraphic index: Based on the calculation method of runoff generation in the Xinanjiang model and the in the HBV model, we developed ahydrological model to contain both the two calculation methods. And the actual calculation method of runoff generation in each subcatchment is determined by its mean topographic index: when its mean topgraphic index is higher than a threshold value, runoff genetaion in this subcatchment is calculated by Xinanjaing method, and otherwise, its runoff generation is calculated by HBV method. The flexible model structure had been applied in the Lancangjiang basin and improved the simulation of peak flows significantly.