Using disjunctive kriging with declustering to create a prediction map
Etapes :
-
Click the Geostatistical Analyst Wizard button
on the Geostatistical Analyst toolbar.
- Choose Kriging/CoKriging, choose a dataset and attribute field, then click Next.
- Choose Disjunctive, set Transformation type to Normal Score, then set Decluster before transformation to True.
- Specify the desired parameters on the Declustering dialog box and click Next.
- Specify the desired parameters on the Normal Score Transformation dialog box and click Next.
- Specify the desired parameters on the Semivariogram/Covariance Modeling dialog box and click Next.
- Specify the desired parameters on the Searching Neighborhood dialog box and click Next.
- Examine the results on the Cross Validation dialog box and click Finish.
- Click OK on the Method Report dialog box.

If you use a normal score transformation and your data has been preferentially sampled (higher densities of points in some areas), declustering the data is a good option to include in the model.
Rubriques connexes
7/10/2012