Spatial autocorrelation analysis of soil water content in a karst region of Guangxi Province

Mashan county, located in the middle Guangxi Zhuang Autonomous Region, southwestern China, was selected as the study area. Based on the plentiful information from field surveys, soil sampling and laboratory analysis, we were studied the spatial autocorrelation coefficients, correlation distances and spatial patterns of soil water content in topsoil (0–20 cm) using semi-variances and Moran’ s Istatistics. The results show that the mean value of soil water content is 16.97%. Soil water content shows a moderate spatial autocorrelation within the distance of 78.8 km, which is affected by the constitutive and random factors. (2) Moran index of soil water content in the study area is 0.43, suggesting that the soil water content possesses spatial autocorrelation. In the ranges of 0-21.7 km and 31-34 km, the values of Moran′s Iof soil water content are greater than 0, implying positive spatial autocorrelation; while in the ranges of 21.7-31 km and 34-45 km, the values are negative, indicating negative spatial autocorrelation. Lisa cluster maps show that there are spatial aggregation areas and spatial isolated areas of the soil water content. The “high-high” spatial aggregation areas cluster in the northeast of Mashan county and “low-low” spatial aggregation clustered in the southeast. There are bigger risk of short of soil water content in the “low-low” spatial aggregation and “high-low” spatial isolated areas.