中文摘要：

降水、气温的空间分布是影响流域水量平衡模拟的关键因素,运用距离权重反比法（IDW）、梯度距离权重反比法（GIDW）、样条函数法（Spline）和克里金插值法（Kriging）对青海湖流域及周边地区43个气象站1995—2009年逐日气温和降水进行了空间插值,并以气象要素空间插值数据驱动模型,进行布哈河流域径流模拟.选用布哈河口月平均流量,以Nash-Suttclife系数（Ens）、相关系数（R）和相对误差（RE）为评价指标,进行校准期（2000—2004年）和验证期（2005—2009年）的径流模拟效果比较.结果表明：径流模拟精度较高,GIDW和IDW更适合于布哈河流域的气象要素空间化,并且气象要素空间插值数据误差是引起模型模拟不确定性和参数据不确定性的原因之一.

英文摘要：

Spatial distributions of precipitation and air temperature are key factors affecting water balance modeling in a watershed.In this paper,the inverse distance weighting（IDW）,gradient plus inverse distance weighting（GIDW）,spline and Kriging methods were used to spatially interpolate the daily precipitation and air temperature data from 43 weather stations around the Buhachu basin from 1995 to 2000.Then the four interpolated weather data were used to drive Soil and Water Assessment Tool to conduct runoff modeling,taking 2000-2004 as the calibration period and 2005-2009 as the validation period.The simulated and observed monthly runoffs at the Buhahekou Hydrometrical Station were compared through Nash-Sutcliffe Efficiency（Ens）,coefficient of correlation（R） and relative error（RE）.The results show that the performance of runoff model driven by GIDW and IDW interpolated data is better,and GIDW is most suitable for weather data interpolation in Buhachu basin,judging by snowmelt and runoff coefficients.It is found that errors in the spatial interpolated weather data cause uncertainties in hydrological process modeling and parameter estimation.