Pirot, Guillaume1 and Lindsay, Mark1 and Jessell, Mark1
1Centre for Exploration Targeting, The University of Western Australia, Crawley, Australia
In a context of early stage exploration, geological modelling relies heavily on sparse surface and drillhole geological observations. These data allow us to develop interpretations of the stratigraphy, organizing identified lithological formations. Though the data may provide local estimations about the thickness and depth of some formations, geological uncertainty away from the observations may be huge. One way to reduce this geological uncertainty is to acquire more data by drilling. However, given the associated costs and risk to the project, the next drilling location is carefully selected. This is difficult as the next location for drilling is often decided under uncertainty due to a dearth of supporting information and knowledge.
Here, we explore different strategies to optimize the selection of successive drilling locations. The first strategy relies on the volume of influence around the borehole design, acting as a moving window filter, and on the value of new information, that is assumed proportional to the current state of geological uncertainty. The second exploration strategy aims at reducing the geological misfit with a Bayesian Global Optimizer.
We test the different strategies on a synthetic case based on a Precambrian basin setting. The initial geological knowledge is composed of surface data and five initial boreholes, whose locations are determined by Latin Hypercube Sampling. At each iteration, an ensemble of geological realizations is generated by stochastic perturbation of the current geological knowledge. Geological uncertainty is summarized from different indicators based on the cardinality, entropy, connectivity, topology and geostatistics of both lithological formations and their underlying scalar-fields.
Preliminary results show that the first strategy allows a decrease of both the mean and ninetieth percentile of the geological uncertainty. The Bayesian Global Optimisation approach reduces locally the geological misfit. A third strategy combining the first two approaches gives a good compromise to reduce both the geological uncertainty and the geological misfit and shows promise to support decision-making in practical regional geological exploration scenarios.
This work is supported by the ARC-funded Loop: Enabling Stochastic 3D Geological Modelling consortia (LP170100985) and DECRA (DE190100431) and by the Mineral Exploration Cooperative Research Centre whose activities are funded by the Australian Government’s Cooperative Research Centre Programme. This is MinEx CRC Document 2020/45.
Guillaume joined the Centre for Exploration Targeting in September 2019. He is involved in the ‘Automated 3D Modelling’ MinEx CRC project and as Work-Package 5 leader in the LOOP consortium (loop3d.org), where he develop tools to improve the characterization, the propagation and the reduction of prediction uncertainties.