Combining Nested and Linear Sampling for Determining the Scale and Form of Spatial Variation of Regionalized Variables

The scale and pattern of variation of continuous spatial variables are often difficult to identify. The semi‐variogram of regionalized variable theory provides a precise solution once the approximate scale of spatial variation is known. The latter can be determined economically over several orders of magnitude by a nested analysis of variance where stages incorporate spatial scale. The method can become cumbersome if many stages are required, and an unbalanced design is described that enables many stages to be examined with a reasonable number of sampling points. The methods are illustrated with examples from a soil survey in the Midlands of England.