What's wrong with geostatistics?
Geostatistical fiction
Statistical fact
The Bre-X fraud
Bait borehole
Documentation
Sampling and Statistics Explained
Spreadsheet templates
Statistical facts
Geostatistical textbooks
 
About my work
 
Contact me
 
 
 

Bre-X's salted boreholes and kriged boreholes
in three lines between SEZ44 and SEZ49

Test for spatial dependence


The "geostats" (kriging variances of sets of kriged estimates) for each of the distance-weighted average show that kriging does create an illusion of spatial dependence where it does not exist. Fisher's F-test cannot be applied simply because each kriged estimate is a functionally dependent value of a set of independently measured values. After all, only independently measured values give degrees of freedom even measured in salted core samples.

Bre-X's Busang was served to its shareholders as a "resource" rather than a "reserve". Just the same, the practice of interpolation by kriging between widely spaced lines of boreholes (step-out drilling in exploration jargon) is a scientific fraud whenever sets of boreholes within lines do not display a significant degree of spatial dependence. Indeed, interpolation without justification will always be a scientific fraud no matter how many neologisms are used to dress it up!

 
 
©Matrix Consultants Limited All rights reserved.