Performing declustering to adjust for preferential sampling

There are two ways to decluster your data: by the cell method and by Voronoi polygons. Samples should be taken so they are representative of the entire surface. However, many times, the samples are taken where the concentration is most severe, thus skewing the view of the surface. Declustering accounts for skewed representation of the samples by weighting them appropriately so that a more accurate surface can be created.

Learn more about declustering

Steps:

1. Choose Kriging/CoKriging, select a dataset and field, then click Next.
2. Choose either Probability, Disjunctive, or Simple as the kriging type; set Transformation type to Normal Score and the Decluster before transformation parameter to True; then click Next.
3. Click Next on the following dialog box.
4. Click the tabs to toggle between the Cell size, Anisotropy, and Angle charts.
5. Change Cell size , Anisotropy, Angle, and Shift to find the minimum in the graph.

   Alternatively, click the Declustering method arrow and choose Polygonal to switch to a polygon declustering display.

6. Click Next.
7. Choose a method type from the Approximation Method drop-down menu, set Number of bins, then click either the Density or Cumulative tab on the Normal Score Transformation dialog box.
8. Click Next.
9. Specify the desired parameters on the Semivariogram/Covariance Modeling dialog box and click Next.
10. Specify the desired parameters on the Searching Neighborhood dialog box and click Next.
11. Examine the results on the Cross Validation dialog box and click Finish.
12. Click OK on the Method Report dialog box.

Tip:

Declustering is only used when you choose normal score transformation as the transformation type. Choose Probability, Simple, or Disjunctive as the kriging method to access the Normal Score option.