Geostatistical interpolation differs from the procedures described in Section 6.6 in that it assumes the data point values represent a sample from some underlying true population. By analysing this sample it is often possible to derive a general model that describes how the sample values vary with distance (and optionally, direction). This model may be then used to interpolate, or predict, values at unsampled locations, in much the same way as with deterministic interpolation. If the samples meet various additional conditions it may also be possible to provide estimated confidence intervals for these predictions. Geostatistical interpolation methods attempt to address questions such as:

·         how many points are needed to compute the local average?

·         what size, orientation and shape of neighbourhood should be chosen?

·         what model and weights should be used to compute the local average?

·         what errors (uncertainties) are associated with interpolated values?

A substantial number of specialist terms are used in the field of geostatistics and many of these are explained in subsection 6.7.1.1, before we proceed to describe specific geostatistical models. These include:

·         Geostatistics

·         Semivariance

·         Sample size

·         Support

·         Declustering

·         Variogram

·         Stationarity

·         Sill, range and nugget

·         Transformation

·         Anisotropy

·         Indicator semivariance

·         Cross-semivariance

·         Comments on geostatistical software packages

·         Semivariance modelling

·         Fractal analysis

·         Madograms and Rodograms, and

·         Periodograms and Fourier analysis