开都河流域山区径流模拟及降雨输入的不确定性分析

Distributed Hydrological Modeling in Kaidu Basin:MIKE-SHE Model Calibration and Uncertainty Estimation

选取塔里木河源区的开都河流域为研究区,将流域内气象水文站点数据与遥感数据相结合,利用气象、土壤类型、土地利用和地表覆盖、数字高程(DEM)和降雨等资料,模拟流域水文过程,并在出山口实测径流数据的基础上对模型进行率定和验证;对降雨输入所带来的径流模拟不确定性进行分析,探讨降雨输入的空间异质性对水文预报结果的影响机制.结果表明:MIKE-SHE模型能在水文、气象站点稀少,土壤及水文地质数据缺乏的条件下,模拟开都河流域的日径流过程,但精度上仍有待提高;降雨输入的空间分布程度对径流模拟有重要影响.FY-2C遥感估算降雨资料能够更好地表达降雨时间的空间异质性,相应地对径流模拟精度也有一定程度的提高.

The distributed watershed model is the important tool for decision-making in water resources management.For a large extent,the accuracy of models in predicting the hydrologic process and erosive behaviors is highly affected by the knowledge in respect of the spatial precipitation.Particularly in the arid and semi-arid areas,where water is short,the continuous assessment and monitoring of the hydrological system components are necessary.A GIS-aided MIKE-SHE model was applied to simulate the stream flow in an ungauged Kaidu River basin（19 012 km2）,one of the most important sources of the Tarim River in Northwest China,located in the arid areas.Further,the rainfall variability and its effect on the flood prediction was analyzed by using 4-year rainfall observation data collected at the watershed and 9-month remote sensing precipitation data of the whole region.Two different parameter sets were calibrated via traditional meteorological records and remote sensing satellite data.The observed discharges at Bayinbuluke and Dashankou stations were used as reference for calibration and validation.It is clearly shown that a well-calibrated MIKE-SHE model with five ＂ree＂ parameters is able to produce consistent results with correlation coefficients greater than 0.7;the spatial variability of precipitation has a significant improvement on the model performance.