中文摘要：

针对估算山区土壤水分过程中使用的数据和辐射传输模型都十分类似、精度不高的问题，对遥感环境因子和土壤水分做相关性分析以提高估算精度。利用Landsat-5TM计算水分指数，获取MODIS植被指数、地表温度数据，基于黑河上游148组土壤表层（0，10cm）水分实测数据，提取148组海拔高程、地表温度、水分指数、植被指数，采用线性回归法得到土壤水分与环境因子的相关性。结果表明，土壤水分与海拔高程、水分指数、植被指数为正相关，且相关性依次减小，与地表温度为负相关，相关性介于植被指数和海拔高程之间，且全部通过0．01的Pearson相关系数双尾检验。

英文摘要：

Soil moisture is of great importance to regional climate, ecology, hydrology and other systems. Understanding the relationship between the soil moisture and the environmental factors by remote sensing is essential, particularly in high elevation,and hard to access mountainous watersheds such as the Heihe River Watershed in Northwest China. This paper correlated several indices with the soil moisture content in the upper reach of the Heihe River Watershed. Normalized Difference Water Index （NDWI） was calculated by using Landsat 5 TM images, and then MODIS NDVI, LST data were obtained from NASA and USGS sites. Based on the 148 sets of measured surface soil moisture （0-10 cm） in the Upper Reach of the H eihe River Watershed,we extracted 148 sets of NDWI, NDVI, LST, and DEM data, and conducted correlation analysis between the surface soil moisture and those indices. The results indicate that surface soil moisture is in significant positive correlation with NDVI, NDWI,and DEM negative correlation with LST. All these correlations are significant at the 0.01 level （2-tailed） ;vegetation has the largest influence on soil moisture and surface temperature,while altitude and the surface humidity have relatively smaller influence on the surface soil moisture in the study area.