CHAPTER 5: Kriging
SUMMARY OF MAJOR POINTS
1. We can evaluate the accuracy of any linear estimator if we have a model for
the semi-variogram.
2. We can produce the minimum variance
unbiased linear estimator using the kriging technique, if we have a model for
the semi-variogram.
The standard litany of the advantages of kriging can be
found in numerous publications. The points of major importance are:
a) Given the basic
assumptions, no trend, and a model for the semi-variogram, kriging always produces the Best Linear Unbiased Estimator.
b) If the proper models are
used for the semi-variogram, and the system is set up correctly, there is always a unique solution to the kriging
system.
c) If you try to ‘estimate’
the value at a location which has been sampled, the kriging system will return
the sample value as the estimator, and a kriging variance of zero. In other
words, you already know that value. This is
usually referred to as an ‘exact interpolator’.
d) If you have regular sampling, and
hence the same sampling/block setup at many different positions within the
deposit, it is not necessary to recalculate the kriging system each time.