基于逐步聚类分析的统计降尺度模型（SCADS模型），在多GCM模型集合的9个大尺度气象变量与开都河流域6个气象变量之间，建立统计降尺度关系，并进行开都河流域未来气候变化的预估。结果表明，SCADS模型生成的开都河流域各气象变量的模拟值与实测值拟合较好。各气象变量在率定期（1961年－1990年）和验证期（1991年－1999年）的NSE系数均大于0.55，精度较高。此外，利用SCADS模型进行开都河流域各气象变量的预估。发现在三个不同时期内（2011年－2040年，2041年－2070年和2071年－2100年），月均气温升高，月均蒸发量、降水量、日照时数增加，月均相对湿度升高。

A statistical downscaling model that based on the stepwise-cluster analysis(SCADS), was employed to establish a statistical downscaling relationship between the large-scale climate variables from the multi-ensemble GCMs, and the regional climate variables of the Kaidu River basin, as well as to calculate the prediction of climate change in the future. The results indicated that the outputs of SCADS could model the climate variables of Kaidu River basin with a satisfactory. The NSE for all climate variables in the calibration (1961-1990) and validation (1991-1999) periods were larger than 0.55, indicating a good precision of SCADS. Besides, in terms of the prediction of climate change in the future of Kaidu River basin, it was shown an upwards of monthly average temperature, and a larger amount of evaporation, precipitation, sunshine-hour, and relative-humidity in three different periods (2011-2040, 2041-2070 and 2071-2100).

国家自然科学基金项目（51379075）