The Community Land Model version 4.0 (CLM4.0) driven by the atmospheric forcing data of Princeton University was used to simulate soil moisture (SM) in China from 1961 to 2010. The simulated SM data were compared with observation data, National Centers for Environmental Prediction (NCEP) Reanalysis data, and SM data retrieved from the Advanced Microwave Scanning Radiometer-Earth Observing System (EOS; AMSR-E) data. The results showed that although CLM4.0 simulation is capable of capturing the characteristics of spatial distribution and temporal spatial variation of measured SM, the simulated SM was systematically higher than the observations in every layer in northeast China, the Jianghuai basin, and the Hetao region. The spatial patterns of the simulated SM and the NCEP SM demonstrated reasonable consistencies, and the spatial patterns of simulated SM and AMSR-E retrieved SM showed consistencies in the northern area of 35°N. On the basis on the simulated SM from 1961 to 2010, it can be concluded that the spatial distributions in every layer were characterized by a gradually increasing pattern from the northwest to southeast. Dry regions were located in Xinjiang, Qinghai, Gansu, and western Inner Mongolia, whereas the most humid regions were located in the Northeast China Plain, the Jianghuai region, and the Yangtze River basin. The SM generally increased from surface layer to deeper layers. The variation trends essentially showed consistencies in all layers. The SM mainly decreased in the northern area of 35°N except western Xinjiang and partial regions of Northeast China, and increased mainly in the Yangtze River basin, southern and southwestern China. In the global warming background, the simulated SM of summer demonstrated various responses to the variation in precipitation. SM decreased in typical arid and semiarid regions and increased in humid regions. The humid regions responded significant to precipitation, whereas the semiarid and arid regions were ranked second.