A sets of soil moisture data produced from CLM3. 5 driven by the China Meteorological Forcing Dataset, and two widely-used re-analysis soil moisture datasets ( ERA40 and NCEP - 2) were used to systematically and comparably analyze the spatial-temporal variations of surface ( 0 ~ 10 cm) soil moisture in the growing season ( May to September) over the Taihang Mountains from 1980 to 2008,and matched them with observational data within the same study area. Results show that CLM3. 5 data and ERA40 data are generally consistent well with the observations, and both exhibits a relatively consistency of spatial distribution of soil moisture,namely dry in the north while wet in the south,in which spatio-temporal variability of soil moisture simulated by CLM3. 5 is larger than that of ERA40 data. However,NCEP - 2 doesnt describe the spatio-temporal distribution of soil moisture well. Moreover, soil moisture has significantly positive correlation with precipitation and relative humidity in the Taihang Mountains,wheraes negative correlation with air temperature. In real condition,the mechanism of interaction between soil moisture and climate change is more complex.