|
Gayle,
I can't give you any solution but can inform you of my past
experience;
- Gold deposits normally display skewed distributions,
so that if the
assays are used for modelling (usually block
modelling) the result
will be that the grade will be
overestimated.
- Lognormal kriging was developed to overcome this. You
krige the
natural logarithm (not base 10) of the
grades then back transform
the results making an adjustment for the
variance.
- The problem with lognormal kriging is that it is
highly sensitive to
your interpreted variograms,
and infact the error in your grade
estimate is directly proportional
to the error in your variogram
e.g. sill estimation.
Hence lognormal kriging is best
reserved for deposits which show
very well formed lognormal
variograms.
- Gold data of course is not always strictly lognormal
and often
shows mixed distribution
characteristics (you may use
disjunctive kriging
for mixed distributions which you may like for a
comparison with other estimation
methods) which worsens the
prospects of
accurate modelling.
- In such cases as a non strictly lognormal population
and poor
variograms which may be the case the
old hand method is to cut
the grade population prior to modelling
then just use inverse
distance squared or cubed
(experience that cubed is better for
skewed gold distributions) or ordinary
kriging modelling.
- Determing the cut value is the problem. Some use
a cut value
based on experience or
reconciliation of deposits in the region.
I have used a method whereby I calculate
the sichel mean of
the dataset (an estimate of the
true mean of a lognormally
distributed population from a sample
dataset) then cut the dataset
until the arithmetic mean of the cut
dataset equals the previously
calculated sichel mean.
I am currently working on an improved
top cut method
calculation in association with
Frans Manns.
Software used is any generalised mining package for the
modelling
and for calculation of the topcut, if you don't have a
macro based
software or are intimately familiar with a database package,
spreadsheet software will suffice.
I have not checked or do not know if there is any software in
the
Stanford University GSLIB software library for treatment
or
processing of mixed populations.
If you would like further details on any of the above processes
please
feel free to contact me.
Regards Digby J. Millikan BEng.
|
* By using the ai-geostats mailing list you agree to follow its rules ( see http://www.ai-geostats.org/help_ai-geostats.htm )
* To unsubscribe to ai-geostats, send the following in the subject or in the body (plain text format) of an email message to [EMAIL PROTECTED] Signoff ai-geostats
