Study on the Spatial Variability of Farmland Soil Nutrient Based on the Kriging Interpolation

Abstract

Studies about the spatial variability and elevation influence of soil nutrient including soil total N (TN), available P (AP) and available K (AK) could provide a theoretical basis for the scientific management of farmland and protection of ecology and environment of lands. Based on the analysis of one hundreds more soil samples of farmland soil in Ninghai County, Zhejiang province, China, the spatial variability of soil TN, AP and AK was studied employing Kriging interpolation and geostatistical analysis. The elevation affected significantly the distribution of TN and AK. The semivariogram model for soil TN and AK in ordinary Kriging interpolation was well fitted by exponential and spherical model, respectively, in lag size of 1000 m with directional influence. The ratio of Nug/sill of soil TN is 70.8%, which means moderate spatial autocorrelation. The ratio of Nug/sill of soil AK is .81.86% which means weak spatial autocorrelation. Soil AP was well fitted by the spherical model in lognormal Kriging at the lag size of 1000 m without directions influence and the ratio of Nug/Sill is 57.43% which showed moderate spatial autocorrelation. The spatial autocorrelation of soil TN, AP and AK showed that random factors such as fertilization, soil management and land use weak the spatial variability of soil nutrient in studied area. In addition, Kriging could successfully interpolate soil TN, AK and AP. The spatial distribution pattern of soil TN presents the clear zonal shape from east to west while the content of soil TN is less and less from north to south. The spatial distribution of soil AK is more and more from north to south and presents the clear zonal shape from northeast to southwest. The spatial distribution of AP had no apparent geographical rule. The spatial distribution of soil TN, AK and AP validates the conclusion of elevation influence.