Ordinary Kriging

To utilize Ordinary Kriging, select Ordinary Kriging from the Parameters Window when the Interpolation methods step is selected. This interpolation method requires that a grid, a search neighborhood, and spatial correlation model be defined first.

Then, press Show the results on the Steps Window to display the applicable result in the Graphics Window.

Ordinary kriging (OK) is a geostatistical approach to modeling. Instead of weighting nearby data points by some power of their inverted distance, OK relies on the spatial correlation of the data to determine the weighting values. This is a more rigorous approach to modeling, as correlation between data points determines the estimated value at an unsampled point. The concept of spatial correlation and how to measure and model it in your data set is briefly described in Spatial Correlation and Modeling Spatial Correlation. Furthermore, OK makes the assumption of normality among the data points. (See Setting Normality/Lognormality Assumption.)

In addition to the spatial correlation structure, SADA requires a definition of the neighborhood around estimation points. The issue of neighborhood definition is important to inverse distance and indicator kriging, as well. A discussion of neighborhood definitions is consolidated in Defining A Neighborhood.

Because OK is a statistical framework, a kriging variance is also produced for each block that can be viewed with the Draw a variance map interview. The OK estimate and OK variance are the parameters of a normal distribution located at the estimation point that can serve as a measure of uncertainty about the estimated value. This serves as an important foundation for decision frameworks that determine cost and boundaries of the remedial process. See Overview of Decision Frameworks.

A full explanation of ordinary kriging is beyond the scope of this manual. It is assumed that the reader is familiar with the process. For information on ordinary kriging see GSLIB Geostatistical Software Library and Users Guide by Deutsch and Journel (1992) or An Introduction to Applied Statistics by Isaaks and Srivastava.