The 2D Variogram Matix is one of the reporting options within the Kriging modeling settings. It is used to generate a detailed plot that summarizes the variograms, their statistics, and more. You can think of this as a summary report of all of the different model types and directions that you can view separately in the interactive variogram editor.
Some definitions:
- Variogram: See What is a Variogram for basic information about variogram plots and models.
- Major Axis: The bearing, if any, along which samples are more related at greater distances. (Is bi-directional, so the major axis includes 180 degrees opposite.)
- Minor Axis: The bearing that is perpendicular to the Major Axis. (Is bi-directional, so the minor axis includes 180 degrees opposite.)
- Range: The point-to-point distance beyond which which samples are no longer considered related.
- Sill: Where the variability curve flattens out.
- Nugget: If the variogram does not go through the origin, that jump up the Y axis is called the "nugget". It indicates that even at very close distances there is some variability.
- Correlation: This correlation coefficient reports how well the variogram model fits the data. 1 = perfect direct relationship, 0 = no relationship, -1 = perfect inverse relationship.
- Anisotropy: This reports the directionality of the data, and equals the minor-axis-range / major-axis-range. The closer to 1, the less directional (or anisotropic) the data.
- Spoke Spacing: Azimuth angle (in degrees) between point-pair sampling spokes.
Since a variogram compares point values based on distance, it relies on a bunch of point pairs to do this comparison. Instead of using all possible point pairs (a potentially huge sample), the points are sampled along spoke and distance (or lag) increments.
- Spoke Tolerance: Azimuth angle (degrees) to either side of the spokes within which samples will be pulled.
- Lag Spacing: Distance increment (in map units) along the spoke bearings at which samples will be pulled.
- Lag Tolerance: Distance along either side of the lag increments within which point pairs will be pulled.
Back to Kriging (Grid Models)
Back to Kriging (Solid Models)
RockWare home page