Soil moisture is a key variable affecting agricultural production and ecological environment. It is therefore of great significance to acquire accurate spatial distribution information of soil moisture. The Cosmic Ray Neutron Probe (CRNP) can realize the continuous observation of soil moisture at field scale, but its ability to observe soil moisture in different geographical environments and vegetation cover types still needs to be further understood. In this study, data from 16 representative observation stations worldwide were selected, and a unified data processing process and soil water estimation method were adopted to carry out a systematic and comprehensive comparison of soil water observation capabilities of CRNP under different climatic zones and vegetation cover conditions. The results indicate that, in general, CRNP observation was highly sensitive to soil water dynamic changes and can clearly respond to precipitation events. CRNP outperforms over grasslands, the results of one site were relatively poor, and the other sites showed very high accuracy of soil moisture estimation, with the optimal RMSE and ubRMSE reaching 0.03cm3 /cm3, the optimal Bias reaching -0.01cm3 /cm3, and the R2 reaching 0.93. In agricultural fields and forest cover, the quality of observation results is still high despite the slight decrease of R2. The optimal RMSE is 0.05 cm3/cm3, the optimal ubRMSE is 0.03 cm3/cm3, the optimal Bias is 0.006 cm3/cm3, and the optimal R2 is 0.88. In shrub areas, The CRNP observation effect of selected sites is strongly affected by the local environment, and the accuracy of soil water estimation is quite different. This study helps us better understand the capabilities and potentials of CRNP in observing soil moisture, and provide a scientific basis and reference for applying CRNP observation to capture the spatio-temporal changes of soil water at field scale under different vegetation cover and climate conditions.