中文摘要：

利用2008年1月1日至2009年9月31日黑河流域阿柔冻融观测站的气象和土壤水分数据,采用基于方差的多参数敏感性分析方法研究通用陆面模型（Common Land Model,CoLM）模拟的土壤水分对土壤质地（砂土和黏土）的敏感性,进而采用SCE-UA参数优化算法分别优化土壤质地和土壤水力参数,分析不同优化策略对土壤水分模拟结果的影响。研究结果表明,浅层土壤水分对土壤质地较为敏感,敏感性系数达到了0.45以上,并且砂土含量对土壤水分的影响更为显著;利用SCE-UA算法优化土壤质地或土壤水力参数都可以有效地提高土壤水分的模拟精度,优化土壤水力参数易产生＂异参同效＂现象,而优化土壤质地能够使土壤水力参数的取值范围更加合理。

英文摘要：

The sensitivity analysis of soil moisture to soil texture （sand and clay） was studied by using the common land model and meteorological forcing data from January 1, 2008 to September 31,2009 at the Arou Observation Station in Heihe river basin. The SCE-UA algorithm was utilized to optimize the soil texture and soil hy- draulic parameters separately and the impact of the resuh of soil moisture with different strategies was analyzed. The result shows that surface moisture is quite sensitive to soil texture, the sensitivity coefficients are larger than 0.45, and the sand percentage is more significant to the soil moisture; SCE-UA algorithm is used to optimize the soil texture or soil hydraulic parameters which can effectively improve the accuracy of soil moisture simulation. However, optimizing soil hydraulic parameters will likely lead to the apperance of the phenomenon of equifinality for different parameters, while optimizingsoil texture can constrain the range of soil hydraulic parameters and make them more reasonable.