County-Scale Spatial Variability of Soil Nutrient Distribution and Determination of Reasonable Sampling Density

Spatial variations of soil organic matter, available nitrogen, phosphorus and potassium were discussed based on geostatistics combined with GIS technology in Wugong County. Spatial interpolation results from different sampling densities of organic matter were analyzed and compared by testing interpolation accuracy under different densities with the Root Mean Square Error(RMSE) and correlation coefficient(r) in order to determine a reasonable number of samples in the county. The results showed that there existed moderate spatial variability in the soil nutrient. The order of coefficient of variation of soil nutrient was available phosphorusavailable potassiumorganic matter available nitrogen. The soil nutrients, with positive base effect, showed moderate spatial variation and moderate spatial autocorrelation. Among them, the spatial variability of organic matter and potassium was little affected by human factors, but related with the soil parent material, soil type, climate conditions, respectively. Nevertheless, the spatial variability of available nitrogen and phosphorus was much influenced by human factors, and related with the tillage and fertilizer agricultural practices. With the increase of sampling density, kriging interpolation accuracy was improved. It was possible to reduce the number of samples to satisfy the need of the research regarding spatial variability of soil nutrients. Taking spatial prediction accuracy into consideration, more than 2213 samples should be the suitable number to evaluate the spatial variability of soil organic matter on county scale, and the maximum sampling unit was 17.8 hm2.