The karst region of southwest China is characterized by surface soils low in moisture and nutrients, low vegetation coverage and high microhabitat diversity. Knowledge of the temporal and spatial variability of surface soil moisture content is crucial for hydrological modeling and understanding soil water dynamics at different scales. This study aimed to characterize the spatial variability of surface soil (0-16cm) moisture content on two slopes, one dominated by shrubs and the other by a shrub-grass mixture, using a grid (10 m×10 m) sampling scheme and geostatistical methods. The study was conducted in the dry season (October, 2011 to March, 2012) in a typical karst area located in northwest Guangxi, southwest China. Surface soil moisture content was found to vary moderately, with coefficients of variation from 10% to 100% on the two slopes. Surface soil moisture content and associated coefficients of variation on the shrub slope were significantly higher than those on the shrub-grass slope. This was mainly attributed to the irregular distribution of shrubs on the shrub-grass slope. Soil moisture content was positively correlated with rainfall but negatively correlated with coefficients of variation. The spatial distribution of surface soil moisture content on the two slopes differed. The spatial range of soil moisture content was negatively correlated with the magnitude of the nugget effect. For the shrub slope, surface soil moisture content was negatively correlated with mean soil moisture at the initial and middle stages of sampling, but was positively correlated mean soil moisture at the later stage of sampling. For the shrub-grass slope, surface soil moisture and mean soil moisture were positively correlated at all sampling stages. Differences between the two slopes were likely due to the combined effects of vegetation, soil type, topography and other factors. The coefficients of variation, spatial range and the nugget effect of soil moisture varied with season on the two typical slopes. These results suggest that mean soil moisture has a substantial influence on the spatial and temporal variability of surface soil moisture in this area;this could help design suitable soil sampling strategies to match future research objectives. The spatial distribution patterns of soil moisture on the two slopes were much different from those found in previous studies in karst areas. This was probably due to the geological differences between our study sites and sites of the previous studies. Kriged maps based on the selected variogram models showed a regular distribution of soil moisture on the two slopes. Soil moisture had a negative relationship with altitude on the two slopes. However, the maximum soil moisture was observed at the right downhill corner, probably due to the gentler slope, greater soil depth, and lower gravel content in that location. In summary, by investigating the spatial and temporal variation of surface soil moisture on two slopes with different vegetation types, this study helps to improve our understanding of hydrological processes, which is useful for rehabilitating vegetation in the karst region of southwest China.