Geoscience Reference
In-Depth Information
data) or external reviews of the information should be
specified.
c. All original drill hole information should be properly
stored for possible future use: laboratory checking, re-
mapping, or for other types of reviews. The information
should be electronically stored in no less than two dif-
ferent locations, one of them away from the project or
operation site. In addition, all information relating to each
drill hole, including topography, drill hole deviation mea-
surements, mapped geologic attributes, and a copy of the
assays returned should be available in hardcopy with a
single folder for each drill hole.
d. Explicit definition and modeling of at least the most im-
portant mineralization controls is required. In the case
where no strong geologic controls are evident, the geom-
etry of the deposit and the data extents should be used
to limit the resource model. If the deposit has no natural
limits, then some form of grade outline may be required
to constrain grade interpolation.
e. The geologic interpretation should be completed on two
orthogonal planes (cross sections and plans, for example),
or, if modeling a vein or tabular deposit, a single cross
sectional set of views, orthogonal to the mineralization
strike, should be used. Plan or longitudinal views should
be used for proper control of the main features.
f. The geologic interpretation should be used to code the
block model. The interpreted geology can be converted
into solids using simple techniques such as extrusion ad-
jacent sections to or with areas of influence. At the earli-
est stages of exploration, a computerized geologic mod-
eling (using a nearest-neighbor technique, for example)
could be used to assign geologic codes to blocks, instead
of explicitly interpreting geology.
g. Simple visual checks should be made to ensure that the
blocks have been correctly flagged with the geologic at-
tributes. These checks may include plotting sections and
plans of drill hole information against assigned codes for
all geologic variables modeled in different geologic units.
h. The block model should consider the deposit geometry
and the characteristics of the (potential) mining opera-
tion. The geometric considerations should include vol-
ume, shape of the blocks, the block sizes adequate for the
drill hole density, the use or not of sub-blocks for better
geologic contact definition, and the mining selectivity en-
visioned.
In addition to the above, good practice should include the
following:
a. Periodic internal checks should be made on all aspects of
the database, including the database geology against the
original log sheets. This review should include at least
20 % of the incremental drill holes, and should be done
after each major drilling campaign. Most mining explora-
tion projects and operations work with a yearly budget
that defines the periodicity of the drilling campaigns.
b. Detailed definition of demonstrable geologic controls
should be available, with properly documented geologic
and statistical supporting evidence. This should include
reports or memoranda describing the different con-
trols, and which geologic variables should be explicitly
modeled. Scatterplots and Q-Q plots (Chap. 2) should
be used to describe the relationships between grades and
each geologic variable.
c. Whenever possible, two-dimensional surfaces should be
used to interpret planar geologic features, and three-di-
mensional solids should be created to represent volumes,
including vein and tabular deposits of significant width.
d. A clear audit trail and documentation should be left to
facilitate third party reviews of the work. This includes
working and final sections and plans in paper (typically at
a 1:500 scale), as well as procedures, reports and memo-
randa documenting the internal checks, and the step-by-
step process followed.
e. The geologic model should be checked for volumetric
biases. The declustered drill hole database represents
the proportions or relative volumes among the different
modeled geologic attributes. These relationships should
be maintained in the model. Statistics and proportions
of each geologic variable (code) in the database and in
the resulting block model should be compared. If the ob-
served differences are larger than 10 %, then a clear justi-
fication should be made for accepting such a difference.
f. The block model should characterize the contact between
the different geologic attributes to the level of detail war-
ranted by the data. This can be done through the use of
partial blocks or sub-blocks. The re-blocked model should
preserve the information obtained at the better resolution.
Additionally, best practice includes the following:
a. Multiple procedures and checks should be in place to ensure
constant quality of the geologic database. Automatic map-
ping, logging and input into the computer database in the
field is recommended through the use of computerized log-
ging. The database where the information is stored should be
relational and a sufficient number of automatic basic checks
should be in place to facilitate database maintenance.
b. Exhaustive and detailed work on the definition of geo-
logic controls, including a ranking of importance, if ap-
propriate, should be done and available as part of the
audit trail. The geologic descriptions and the statistical
methods used should be detailed, including analysis such
as Classification and Regression Trees (CART), possibly
varying from one sector of the deposit to another.
c. The geology should be interpreted on all three orthogo-
nal planes (cross sections, longitudinal section, and plan
views). The geological controls and limits should be mod-
