基于扰动分析方法的NAM模型参数敏感性分析

Parameter Sensitivity Analysis of NAM Model Based on Perturbation Analysis Method

降雨-径流模型中的主要参数需通过率定方法得到,为提高参数率定精度,找出参数的敏感性等级,选取NAM模型中9个主要参数,以2002年美国Upper Sangamon流域的Fisher观测站径流和气象数据为研究对象,利用扰动分析法分析9个重要模型参数对于模拟结果的敏感性。结果表明,Lmax是比较敏感参数;Umax、CQOF和TG是一般敏感参数;CK1,2是不敏感参数;其他参数为弱敏感或不敏感参数。通过选取比较敏感和一般敏感参数为率定对象,以该流域2000~2008年降雨-径流数据为基础,对模型进行率定和检验,模拟精度均达到良好水平,提高了NAM模型的模拟精度。

The main parameters of rainfall runoff model were calculated by calibration.In order to improve the accuracy rate and find the sensitivity level of different parameters, 9 main parameters in NAM model were selected and based on the runoff and meteorological data of Fisher observation station in Upper Sangamon basin in 2002, and then the sensitivity of the 9 important main parameters to the model＇ s simulation results were analyzed by the perturbation method.The results showed that Lmax was relatively sensitive to the daily runoff simulation results;Umax,CQOF and TG were generally sensitive;CK1.2 was insensitive;the other parameters were weakly sensitive or insensitive.Through selectingthe relatively sensitive and generally sensitive parameters for calibration object, based on the rainfall runoff data from 2003 to 2008 for calibrationandvalidation,the simulation accuracy of calibration and verification period reached a better effect,and the simulation accuracy of the NAM model were improved.