Geostatistical Simulations for the Mining Industry

5 days

September 8-12, 2008 « Register
Lecturer: Jacques Deraisme
Venue: Geovariances offices in Avon (Paris area)
For specific location or date « Register
For tailored courses, please contact your sales representative.

OBJECTIVE

When estimating recoverable resources after cut-off, you are faced with three important points:

Change of support: your deposit is sampled on small cores and will be exploited using blocks. You have to take this into account when applying a cut-off to get an estimate of recoverable resources.

Support Effect: in order to forecast the distribution of grades for
supports of different dimensions, geostatistics provides models of change
of support, which are based on the experimental histogram of sample
grades as well as their spatial correlation through the variogram.

Information Effect: geostatistics quantifies the amount and the effect of the misclassification.
Information effect: what you will recover are actual grades of blocks, while they are selected using their estimated grades. This information effect must be taken into account to obtain a more realistic recoverable reserves estimate.
Mining constraints: methods based on gaussian anamorphosis assume free selection of the blocks above the cut-off. However, you have to account for more technical constraints due to mining methods, or more generally, to complex rules for selectivity.

Only the simulation techniques allow you to take into account the real variability of the grades and provide you with unbiased estimates.

After reviewing the fundamental concepts of variances handled by geostatistics and the basis on non linear geostatistics, several techniques for simulating grades will be presented: turning bands and sequential gaussian simulations. Emphasis will be given on the practical issues for the exploitation of the simulations: distribution of errors, sensitivity to support, optimization of sampling for quality control...

WHO SHOULD ATTEND

Mine and exploration geologists, mining engineers and managers involved in grade control, ore reserves estimation, feasibility studies or medium to long term planning. Only a basic understanding of geostatistics theory (variograms and kriging) is recommended.

COURSE CONTENT

Reminders of linear and non linear geostatistics:
- Limitations of linear geostatistics,
- Properties of Gaussian random functions,
- Gaussian anamorphosis,
- Dispersion variance and Change of support.
Turning bands simulations: how many bands?
Sequential gaussian simulations: how to choose the neighbourhood?
Control of the simulated results: histograms, scatter diagrams, variograms.
Analysis of the statistics of the estimation error obtained from conditional simulations.
Post processing of simulations.
Tips and tricks for performing operational simulations using Isatis.

ADDITIONAL INFORMATION

Half of the course is dedicated to practical computer exercises, using Isatis, that reinforce the previously presented theoretical notions.