Within-Field Spatial Variability of Surface Soil Properties at Three Dairies in North-Central Texas

Abstract

Soil is always not uniform, even across small areas. Furthermore, the spatial variability of soil is seldom considered in agricultural practices. This study evaluated pastures at three dairies in Erath County, TX. Composited surface soil samples representing pastures of 194, 234, and 284 ha were collected for Dairy A, B, and C, respectively, at a density of 2.5 per hectare. Soil pH, organic carbon, calcium carbonate equivalent, clay, and extractable phosphorus were determined. The within-field spatial variability of the properties was evaluated geostatistically, as well as via descriptive statistics, given the equidistant sample distribution. The semivariogram model provides information to quantify the spatial dependency. For some properties, however, it contradicts the information delivered by classic statistics. Optimal sampling density, which defines the relationship between the expected error level and sampling density, was proposed in this study to improve the comparison capability and the interpretation reasonability of the geostatistical information. The proposed method of calculating optimal sampling density may provide guidance for designing a sampling scheme for larger areas with similar environmental settings, or for areas with known semivariograms of certain soil properties.