Geoscience Reference
In-Depth Information
b. The QA/QC program should be extended to include not
only sample preparation and assaying, but also of the
sample collection process, drill hole collar and sample
locations, including drill hole deviations, drill hole spa-
tial coverage, geologic mapping and logging, and a clear
definition of the variables to be used and their purpose.
c. Issues such as the combined use of soft and hard data
should be properly dealt with and justified.
d. Density determinations should be sufficient to adequately
provide spatial coverage, and also should be subject to
quality control procedures.
e. Detailed, written QA/QC reports should exist for every
step of the sampling and data handling process. These
procedures should include corrective measures as re-
quired.
f. Overall database errors rates, for all elements stored,
should be below 2 %.
Best practice additionally includes:
a. The use of all possible (and relevant) geologic, grade,
and other data to obtain a resource model. The process
of defining the variables and their characteristics should
be well documented. If metallurgical information is avail-
able, a complete description of the relationships between
geology, grade, and metallurgical performance should be
available, such as grade-recovery curves, hardness-plant
throughput curves, etc.
b. There should be detailed reports on the QA/QC proce-
dures implemented available, including its results and
relevant discussions, both for current and historic cam-
paigns. Also, records of internal checks and audits per-
formed for each step of the process should be compiled
and archived.
c. A summary of qualitative or quantitative data uncertainty
should be available for each data component. These can
be derived from the database error rates (for qualitative or
categorical variables), from sampling variance studies for
sample preparation and analysis protocols, or from other
statistical analysis.
d. Database error rates should be below 1 % for all variables
being used.
The primary metal in the hypothetical deposit of this ex-
ercise is copper. The copper mineral is chalcocite (Cu
2
S).
Chalcocite has a density,
λ
m
, of 5.5. The average copper
grade is 2.0 % (note that this is not the average chalcocite
grade). The host rock is a granite with a density,
λ
g
, of 2.3.
The Liberation size,
d
, of the chalcocite is 50 µm. The shape
factor,
f
, for chalcocite is 0.47. The granulometric factor,
g
,
is 0.25.
Samples are taken using diamond core drilling. The core
has a diameter of 52.0 mm. Half of the core will be sent for
assay and the other half will be retained. The nominal sample
length is 2.5 m.
The fundamental sample error is defined as:
†
‡
1
1
σ
2
=
IH
MM
−
ˆ
‰
FE
L
Š
‹
S
2
S
2
3
IH
=
clfgd
L
where
F
σ
is the sampling error introduced when splitting
the sample from
M
S1
to
M
S2
,
d
is the particle size when the
sample is being split,
c
is the mineralogical factor, and
l
is the
liberation factor. The mineralogical and liberation factors are
calculated in part one of this exercise.
5.10.1
Part One: Prerequisites for
the Sampling Nomograph
Question 1:
Calculate the amount of material from 1/2 of
the drill core for a 2.5 m sample length. This
is the starting mass for the sampling protocol.
Question 2:
Calculate the chalcocite content,
a
L
, in the
sample. The calculation needs to be done
using fractions, not a percentage or ppm. Use
the average copper grade for this calculation.
The molecular weight for chalcocite is
159.17, for copper it is 63.55, and for sulfur it
is 32.70.
Question 3:
Calculate the mineralogical factor,
c
, for chal-
cocite. The result from question 2 is required
for this step.
2
5.10
Exercises
−
= +−
(1
a
)
c
λ
L
λ
(1
a
)
m
g
L
a
L
The objective of this exercise is to review some sampling
theory and gain some experience with sampling nomographs.
Some specific (geo)statistical software may be required. The
functionality may be available in different public domain or
commercial software. Please acquire the required software
before beginning the exercise. The data files are available for
download from the author's website—a search engine will
reveal the location.
Question 4:
The next step is to calculate the location of
the size lines for the nomograph plot. This
requires some iteration. The irst step is to
choose some nominal particle sizes. The
second step is to assume a sample mass for
each particle size. This mass is not related to
the actual sample. The mass is used for plot-
