Geoscience Reference
In-Depth Information
they appear realistic and appear to have used the latest avail-
able methodology and software. Senior professionals and
gatekeepers in various organizations will guard against this
and promote better work practices.
Regarding the data for resource estimation, it is likely
that there will be improved use of secondary data coming
from geophysical measurements of less expensive drilling
and sampling techniques. Cokriging and other techniques
to simultaneously use high quality data together with lower
quality data will be developed and implemented in the wide-
ly used commercial software.
The use of multiple point statistics and advanced multivar-
iate methods will be increased to permit improved realism in
the spatial distribution of the models and capture the com-
plex mineralogical relationships present in most of our
deposits. Challenges posed by incomplete sampling of all
the data and local variations in data quality will be met with
improved techniques for data imputation and dealing with
non-stationarity.
The longstanding challenge of non-stationarity will not
go away, but improved techniques will become available
to help subset the data appropriately, model the chosen do-
mains, understand and quantify the nature of contacts and
consider trends in the important variables.
We also feel that professionals (and the available soft-
ware tools) will become better equipped to manage the high
degree of uncertainty associated with resource estimation.
Probabilistic models, scenarios, risk assessment and simula-
tion techniques will become more commonplace and less in
the hands of a few specialists.
References
Matheron G (1971) The theory of regionalized variables and its appli-
cations. In: Fasc 5, Paris School of Mines, p 212
