Summary

It is often difficult to fit a model semivariogram to both the principle-axis and the minor-axes using traditional methods with anisotropy ratios. Traditionally, models (possibly nested) are modified with anisotropy factors; these represent the relative range of the semivariogram for three orthogonal axes. This technique is restrictive, and this discussion presents a method for relieving this limitation by defining different semivariogram models, independently for each axis. These are referred to as directional semivariograms. The process increases the kriging processing time by 80% to 200%, but the method offers the modeler greater flexibility, and simulation or estimation results that are more representative of the site, because the spatial variation of the data can be more precisely defined.

AAPG Search and Discovery Article #90937©1998 AAPG Annual Convention and Exhibition, Salt Lake City, Utah