ABSTRACT
A statistical model includes a random variable, that is, very roughly speaking, a quantity that is sampled and whose values are distributed according to some probability distribution. It is often convenient to model a spatial random field as the sum of a collection of separate components, each with its own properties. A spatial random process whose properties do not vary by location is said to be stationary. The properties of stationarity and isotropy in spatial data must always be presented as assumptions, which can be examined in an exploratory way but cannot be subjected to rigorous hypothesis testing. Data that obey the first law of geography are said to be spatially autocorrelated. N. Cressie presents a model for spatial variability in which the data consist of the sum of four components with differing scales of variability.
