Clare Mawdesley, Geovariances consultant in Australia, is to give a paper presentation in partnership with BMA about "Local uncertainty benchmarking - a coal case study -" Be there!
Session 3: Case studies in resource modelling
Wednesday, September 20, 2017 – 2:35 pm
Kriged estimates are not typically calculated on blocks substantially smaller than the sample spacing (e.g. the Standard Mining Unit or SMU). Moreover, the sample spacing for quality variables is often much greater than the SMU size. For short to medium term planning decisions, the shorter term variability in the quality variables is a key economic consideration in defining, setting and meeting coal product specifications.
Global uncertainties can be calculated using Drill Hole Spacing Analyses (DHSA). However, knowing the precision of a variable for a one or two year production volume is usually not adequate for variables with low spatial continuity that may fluctuate significantly over daily, weekly, SMU, or shipment volumes.
Is there a feasible, reasonable and valid geostatistical method to model the uncertainty of coal quality variables on blocks substantially smaller than the drill spacing (say on the SMU)?
The validity of using co-kriging variances, calculated on SMUs to characterise local uncertainty, was tested using a suite of conditional simulations generated for several quality variables for BMA’s Blackwater operation. The conditional simulation results were used as a benchmark against which the results derived from the kriging variances could be compared.
The agreement between the kriging and simulation methods for thickness is excellent, with much closer correlations than for the quality variables. Typically, the kriging variances give very reasonable approximations to the simulation-based uncertainties, although there could be some systematic discrepancies which may become much more critical if a classification is to be based on a thresholding of the SMUs relative uncertainties.