甘肃梨园河流域SWAT径流模拟与预报

SWAT runoff simulation and predictions in Liyuan River basin of Gansu Province

以甘肃梨园河流域为研究对象,收集相关资料组织数据库,建立SWAT分布式流域水文模型,并模拟1990-1999年月地表径流过程。通过比较模拟值与实测值得相对误差Re小于10%,相关系数R2和Nash-Sutcliffe系数Ens均大于0.80,模拟效果较好,认为SWAT模型可适用于梨园河流域的地表径流模拟。而后结合自回归模型和时域法预测的气象数据,加载到已率定的SWAT模型中预测2009-2018年地表径流。结果表明:预测期径流深呈下降趋势,分析径流对气候变化的水文响应,认为径流对气温的水文响应的敏感性较径流对降水的敏感性高。

Data regarding to Liyuan River basin were collected to establish a data-base.Then,a physically based distributed hydrological model（SWAT） was used to simulate the surface runoff from 1990 to 1999 on the basis of this data-base.After that,the analog value was compared with the actual one,we could get the results that the relative error is less than 10%,the correlation and Nash-Sutcliffe coefficients are both above 0.80.It indicated that the established SWAT model can be applied to the surface runoff simulation of Liyuan River basin.Lastly,meteorological data predicted by auto-regressive model and time domain analysis is added to SWAT model to predict surface runoff of the years（2009-2018）.The results show that there is an apparent downtrend of runoff during this period.Through analying the response of the runoff to climate changes,it is found that the response of runoff to temperature is more sensitive than that of runoff to precipitation.