Diggle and Ribeiro (Model-Based Geostatistics. Springer, New York, 2007) and Mase (Geostatistics and kriging predictors. In: International Encyclopedia of Statistical Science. Springer, Berlin, 2010) describe geostatistics as a branch of spatial statistics that deals with statistical methods for the analysis of spatially referenced data with the following properties. Firstly, values Y i , i = 1, …, n, are observed at a discrete set of sampling locations x i within some spatial region \(\mathcal {S}\subset \mathbb {R}^d\) , d ≥ 2. Secondly, each observed value Y i is either a measurement of, or is statistically related to, the value of an underlying continuous spatial phenomenon, \(Z\left (\boldsymbol {x}\right )\) , at the corresponding sampling location x i . The term model-based geostatistics refers to geostatistical methods that rely on a stochastic model. The observed phenomenon is viewed as a realization of a continuous stochastic process in space, a so-called random field.