Geostatistical Analyses of Soil Salinity in a Large Field

Estimating spatial variability of soil salinity is an important issue in precision agriculture. Geostatistical methods provide a means to study the heterogeneous nature of spatial distributions of soil salinity. In this study, geostatistical methods, kriging and cokriging, were applied to estimate sodium adsorption ratio (SAR) in a 3375 ha agricultural field. In cokriging, more easily measured data of electrical conductivity (EC) were incorporated to improve the estimation of SAR. The estimated spatial distributions of SAR using the geostatistical methods with various reduced data sets were compared with the extensive salinity measurements in the large field. The results suggest that sampling cost can be dramatically reduced and estimation can be significantly improved using cokriging. Compared with the kriging results using total SAR data, cokriging with reduced data sets of SAR improves the estimations greatly by reducing mean squared error and kriging variance up to 70% and increasing correlation of estimates and measurements about 60%. The sampling costs for SAR estimation can be reduced approximately by 80% using extensive EC data together with a small portion of SAR data in cokriging.