Knowledge of the spatiotemporal dynamics of soil water content (SWC) and soil water stable isotopic compositions (SWSIC) is critical for understanding water exchanges across the atmosphere-land interface. However, compared with those of SWC, studies on the spatiotemporal characteristics of SWSIC are still scarce, which limits their use for more accurately characterizing terrestrial ecohydrological processes. To examine whether SWSIC and SWC share similar spatiotemporal features, their spatiotemporal patterns along with relevant controls were investigated at shallow soils (0-10 cm) on a karst hillslope in Southwest China, based on seven sampling campaigns conducted over a two-year period. The results revealed considerable spatiotemporal variations in SWSIC. Compared to SWC, 6D values exhibited smaller spatial variations, while line-conditioned excess (lc-excess) values exhibited considerably larger ones, suggesting that soil evaporation heterogeneity might not be the primary reason for spatial variations in 6D of soil water in this study. The spatial structures of SWSIC were less temporally stable than those of SWC due to the combined impact of land surface processes and temporal variations in isotopic compositions of input water on SWSIC. Moreover, the spatiotemporal patterns of SWSIC were less explained by environmental variables that were also differed from those of SWC, suggesting that the knowledge gained from SWC studies might not be directly transferable to understand SWSIC spatiotemporal characteristics. Nevertheless, it should be highlighted that the temporal stability analysis method initially proposed for studying SWC could be extended to select representative sites for monitoring areal average SWSIC, due to the temporal persistence of the SWSIC spatial structures. In addition, a negative correlation between the spatial mean and standard deviation of lc-excess was found, indicating that the associated spatial variability increased with areal mean kinetic fractionation signals caused by soil evaporation. These findings have important implications for designing SWSIC monitoring schemes, which in turn can improve the interpretation of SWSIC data for various application purposes.