Geologic terrains commonly contain trends arising from structural and depositional causes. Geological mapping algorithms often suppress or distort such trends (anisotropy) because interpolation schemes generally assume spatially uniform (isotropic) influences of nearby data points in all directions. In such cases, the resulting map can look unrealistic and therefore be unusable unless edited. Proper assessment of the data prior to mapping can determine the direction and strength of anisotropy. Incorporation of this information in the gridding algorithm permits a more faithful representation of the underlying geological reality. Improvement in the areas of sparse control arises because the structural analysis coupled with other geological insights puts important constraints on the estimation procedures. In addition, drainage area configurations (Voronoi polygons) more closely conform to the geologic model, when anisotropic properties are incorporated in their construction.
Two methods for incorporating directional trends into surface models are presented, a regional approach where the entire dataset is adjusted before gridding, and a local approach where directional bias is built into the weighting function of each data point. Because the regional approach is built into the gridding algorithm, it is more efficient than some techniques commonly used, which require the user to perform a series of steps. The local approach allows different directions and magnitudes of bias to be applied to different areas of a map more easily. Also, the surface at boundaries between areas of varying trend is smooth, overcoming a major problem with methods currently used.