An introduction to prediction methods in geostatistics

Abstract

In this survey we present various classical geostatistical prediction methods with a focus on interpolation methods that are known as Kriging. For this, we introduce basic concepts in spatial statistics, such as random field, stationarity, and variogram. Then, the main types of Kriging interpolation methods such as simple, ordinary, and universal Kriging are derived as best linear predictors in the mean squared sense. We further comment on multivariate and nonlinear generalizations such as cokriging or indicator Kriging and their aspects of application. Finally, we demonstrate the performance of Kriging prediction with the help of synthetic data.