Articles and letters

We found out in 1989 that David's Geostatistical Ore Reserve Estimation and Clark 's Practical Geostatistics do not show how to derive unbiased confidence limits for metal contents and grades of in situ ore. So, we submitted to CIM Bulletin a paper titled Precision Estimates for Ore Reserves. David wondered why we presented a method, "without a single reference to 20 years worth or work in geostatistical ore reserve estimation." UBC's Sinclair was troubled by, "grossly insufficient references".

CIM Bulletin's Editor advised us to revise our paper accordingly. Scrutiny of David's and Clark's works revealed that geostatistics violates the requirement of functional independence and ignores the concept of degrees of freedom. We made token revisions that failed to pacify our peers. A score of gurus and pundits managed to make a mockery of geostatistical peer review before our paper was finally praised by and published in Erzmetall.

Articles to non-reviewed periodicals and letters to editors underscore my tireless efforts to explain why geostatistics is a fundamentally flawed variant of mathematical statistics. The more so because we showed how to estimate metal contents and grades of ore reserves in an unbiased manner and with an acceptable degree of precision. Thanks to the Internet, the geostatistical fraternity can finally be taken to the task.

Matheron's Note Statistique No 1
Young Matheron thought he was a statistician in 1954. He derived length-weighted average lead and silver grades of core samples. What Matheron didn't derive were variances of lead and silver grade of core samples with variable length.

Matheron's Synopsis to Gy's 1967 Minerals Sampling
Gy and Matheron stated, "As numerous readers in other countries are not very familar with the French language we have thought it desirable to publish English abstracts of the principal original articles."

A study on kriging small blocks,
Armstrong and Champigny, CIM Bulletin, March 1989
Troubled by the rise of the kriging covariance and the fall of the kriging variance, the authors caution against oversmoothing. The requirement of functional independence may be violated a little but not a lot.

Freedom of Speech, De Geostatisticis, July 1992
Dr Margaret Armstrong, in her stirring speech on Freedom of speech, juxtaposes popular opinion and scientific fact with blatant disregard for the requirement of functional independence and the concept of degrees of freedom.

Geostatistics or voodoo statistics
Guest Column, The Northern Miner, May 1992
My crusade against Matheron's fundamentally flawed variant of applied statistics is launched.

Geostatistics: A tool that works
Guest Column, The Northern Miner, May 1992
Normand Champigny, who coauthored A study on kriging small blocks, proclaims geostatistics works well on behalf of five anonymous ore reserve practitioners five years before the Bre-X salting scam.

Geostatistics for the Next Century
Geostatistical gurus gathered at McGill Univerisity on June 3-5, 1993, to honor Professor Dr Michel David, the author of the first textbook on geostatistics. Professor Dr Roussous Dimitrakopoulos chaired the bash. So, he had no time to explain why The Properties of Variances had no role to play in the next century.

Challenging geostatistics
Editorial, The Northern Miner, September 1998
Vivian Danielson, coauthor of Gold Today, Gone Tomorrow, irks the enforcers of geostatistical dogma at CIM Bulletin.

Geostatistics credible and widely used
Letter to the Editor, The Northern Miner, November 2-8, 1998
Professor Dr Alistair J Sinclair, PEng, PGeo, CIM Bulletin's most dedicated geostatistical scholar West of the Rocky Mountains, makes the bizarre claim that "the concept of degrees of freedom breaks down in matters of spatial correlation."

Geostatistics a complex issue
Letter to the Editor, The Northern Miner, November 16-22, 1998
Marcel Vallee, Past President of the CIM Geological Society, is deeply troubled by "the glorification of analysis of variance" but vigorously defends the junk science of interpolation by kriging without justification despite the role of geostatistics in the Bre-X fraud.

Critic should give examples
Letter to the Editor of the National Post was published in The Northern Miner, November 4-10, 2002
Professor Dr Alistair J Sinclair, PEng, PGeo, accepts that the variance of a single distance-weighted average can be replaced with the kriging variance of a set of distance-weighted averages-cum-kriged estimates.

Applied stats examples are documented
Letter to the Editor, The Northern Miner, December 2-8, 2002
I remind Professor Dr Alistair J Sinclair, PEng, PGeo, that he reviewed my work for CIM Bulletin and attended my workshop at UBC.

Whistleblower raises doubts over ore bodies
Sandra Rubin, Senior Business Writer, The Financial Post, reports on geostatistical ore reserve estimation, which is based on assuming spatial dependence, kriging, smoothing and rigging the rules of mathematical statistics.

Agterberg's tribute
This tribute to Professor Dr Georges Matheron was writtten after his passing in 2000. Agterberg called him the Founder of Spatial Statistics but Matheron himself thought he was the Creator of Geostatistics.