Regional pattern of soil organic carbon density and its influencing factors for the plow layers of cropland

Understanding regional spatial pattern of soil organic carbon density (SOCD) and its potential influencing factors in cropland is of paramount importance when evaluating soil quality and assess the carbon sequestration potential. However, little information is available regarding the regional SOCD for the plough layer of the cropland soils with high-density sampling sites. In this context, 14,058 soil samples were collected at a regional scale (21 x 10(4) km(2)) to investigate the spatial variation in SOCD and to identify the dominant factors influencing this variation using classical statistics, principal component analysis (PCA), residual maximum likelihood (REML), and geostatistical methods, such as semivariogram and ordinary kriging methods. The results indicated that soil organic carbon (SOC) and SOCD both demonstrated moderate variability ranging from 2.61 to 35.67 g kg(-1) and 2.61 to 35.67 kg C m(-2), respectively. Geostatistical analysis showed that the nugget-to-sill ratios for SOC and SOCD were 0.59 and 0.61, respectively, which demonstrated moderate spatial dependence. The PCA and a minimum data set method identified that organic nitrogen fertilizer, accumulated temperature >= 0 degrees C, rice straw incorporation, precipitation, and soil pH were the major factors affecting the spatial variability in SOCD. The REML analysis showed that geomorphologic features, soil types, and irrigation conditions also exerted substantial influence on SOC and SOCD. This study provided reliable estimates of spatial variation in SOCD and thus is helpful in the formulation of strategic sustainable SOC management and regional policy decision making.