How to expand prediction maps

Discussion created by thomazz on Jul 22, 2010
Latest reply on Nov 12, 2010 by bartkowalski

I'm currently predicting the soil contamination by cadmium (in mg/kg) in the Campine area by creating a prediction map by means of ordinary kriging. The predictions are made based on a database of more than 1000 sample points. The problem I'm currently confronted with is that the size of the prediction map doesn't cover the entire area that I would like to predict. The reason for this is that the frontiers of the prediction map are determined by the outer sample values, but that for a small part of the area I would like to predict I don't have any sample values. So, only a prediction map is created for the area that is between the outer sample values.
However, I suppose that with the sample points that are close to the outer sample values, it is possible to predict the contamination of a larger area than the area that is limited by the outer sample points.

I hope my explanation of the problem is clear. Is there anyone who can help me?
Thanks in advance!