Soil moisture is one of key factors that influences crop growth and the construction of ecological environments. Research regarding the spatial heterogeneity and driving factors of soil moisture is crucial for the efficient utilization of water resources and management of the ecological environment. To determine the causes of the spatial heterogeneity of soil moisture under four land-use patterns (cropland, forestland, grassland, and bare land) in the ecologically degraded red bed area of the Nanxiong basin under the humid conditions of South China, we used geostatistics and classic statistical methods to analyze the spatial variation of soil moisture, and then discuss the primary influencing factors. A good fit with high determination coefficients (0.876, 0.853) was obtained for cropland and forest when using a Gaussian model in the geostatistical analysis, while the soil moisture in bare land exhibited intermediate spatial autocorrelation which was affected by the structure and random factors. Through cross-validation, the results of inverse distance weighting were determined to be less than those of ordinary kriging, while ordinary kriging could more accurately reflect the spatial distribution of soil moisture in cropland, forestland, grassland, and bare land. Overall, the spatial heterogeneity of the soil moisture was primarily affected by natural and anthropogenic factors, including the structural characteristics of the soil, soil type, vegetation, climate, and changes in land-use patterns. Human intervention had the most prominent role in determining the spatial heterogeneity as well as the driving factors of soil moisture under the four land-use patterns in this typical red bed area.
 

Red beds desertification,Land use patterns,Soil moisture,Spatial variation,Spatial distribution