1Agronomy Department, Tobacco College, Henan Agricultural University, National Tobacco Cultivation, and Physiology and Biochemistry Research Centre, Zhengzhou-450002, China
2Pingdingshan Tobacco Company of Henan Province, Pingdingshan-467000, China
Chongqing Tobacco Science Research Institute, Chongqing -400715, China
*Corresponding author's: e-mail: liugsh1851@163.com
Online published on 29 August, 2013.
Soil samples (from topsoil 0-20-cm) were taken from 111 pointson an approximately 20-m grid in March 2009. Using a geographic Information System Software Platform (ArcGIS), the soil sampling points from the primary scheme were regularly deleted to create three sampling scenarios. Twelve chemical properties were analyzed by geostatistical techniques at three sampling intervals. Results indicated that the twelve properties showed strong spatial variability at the three sampling intervals. The reasonable sampling intervals of soil available phosphorus (AP), available potassium (AK), cation exchange capacity (CEC), Fe, Cu, and Zn were 20-m. In the 20× 40-m sampling interval, soil alkaline hydrolyzable nitrogen (AN), total nitrogen (TN), organic matter (OM), and active soil OM (ASOM), had the least interpolation errors. Soil pH and Mn had the least interpolation errors at the 40-m sampling interval.
Kriging, Precision agriculture, Spatial prediction, Sampling interval, Tobacco
