[Objective] The applicability of soil water assimilation products at different depths was evaluated to provide a scientific basis for the selection of soil water data for studies on climate change and ecological environment construction in the Western Sichuan Plateau.[Methods] Taking the soil moisture value of the automatic meteorological observation station as a reference, the accuracy of the soil moisture products of CMA land data surface assimilation system (CLDAS-V 2.0) and global land data surface assimilation system (GLDAS-Noah) in the Western Sichuan Plateau was evaluated and the response of soil moisture products to the precipitation process and the effect of precipitation on the soil were analyzed.[Results] ① GLDAS and CLDAS data had an extremely significant correlation with the reference value (p<0.01), and the correlation between the GLDAS data and reference value was generally better than that of the CLDAS data. ② The errors of two sets of soil moisture products in the surface and shallow layers of the study area were small. The error of surface soil moisture was greater than that of the shallow layer, and the error of the GLDAS data was smaller than that of CLDAS data. Both sets of data overestimated soil moisture and the overestimation of soil moisture in the case of no precipitation was more prominent than that in the case of precipitation. ③ The two sets of data showed the same change process as precipitation in the surface layer and shallow layer, and the change in soil moisture lagged behind the change in precipitation. The response of the GLDAS data to precipitation change was more sensitive than that of the CLDAS data, but the CLDAS data had better spatial resolution than the CLDAS data in the spatial distribution characteristics of soil moisture in the Western Sichuan Plateau.[Conclusion] The applicability of the two sets of soil moisture products in surface and shallow layers in Western Sichuan Plateau is good. The quality of GLDAS data is generally better than that of CLDAS data, but CLDAS data have more advantages in the performance of spatial differentiation of soil moisture.