Machine learning is becoming an appealing tool in various fields of earth sciences, especially in resources estimation. Six machine learning algorithms have been used to predict the presence of gold mineralization in drill core from geophysical logs acquired at the Lalor deposit, Manitoba, Canada. Results show that the integration of a set of rock physical properties — measured at closely spaced intervals along the drill core — with ensemble machine learning algorithms allows the detection of gold-bearing intervals with an adequate rate of success. Since the resulting prediction is continuous along the drill core, the use of this type of tool in the future will help geologists in selecting sound intervals for assay sampling and in modeling more continuous ore bodies during the entire life of a mine.

A nested workflow of multiple-point geostatistics (MPG) and sequential Gaussian simulation (SGS) was tested on a study area of 6 km(2) located about 20 km northwest of Quebec City, Canada. In order to assess its geological and hydrogeological parameter heterogeneity and to provide tools to evaluate uncertainties in aquifer management, direct and indirect field measurements are used as inputs in the geostatistical simulations to reproduce large and small-scale heterogeneities. To do so, the lithological information is first associated to equivalent hydrogeological facies (hydrofacies) according to hydraulic properties measured at several wells. Then, heterogeneous hydrofacies (HF) realizations are generated using a prior geological model as training image (TI) with the MPG algorithm. The hydraulic conductivity (K) heterogeneity modeling within each HF is finally computed using SGS algorithm. Different K models are integrated in a finite-element hydrogeological model to calculate multiple transport simulations. Different scenarios exhibit variations in mass transport path and dispersion associated with the large- and small-scale heterogeneity respectively. Three-dimensional maps showing the probability of overpassing different thresholds are presented as examples of management tools.

The accurate inference of reservoir properties such as porosity and permeability is crucial in reservoir characterization for oil and gas exploration and production as well as for other geologic applications. In most cases, direct measurements of those properties are done in wells that provide high vertical resolution but limited lateral coverage. To fill this gap, geophysical methods can often offer data with dense 3D coverage that can serve as proxy for the variable of interest. All the information available can then be integrated using multivariate geostatistical methods to provide stochastic or deterministic estimate of the reservoir properties. Our objective is to generate multiple scenarios of porosity at different scales, considering four formations of the Fort Worth Basin altogether and then restricting the process to the Marble Falls limestones. Under the hypothesis that a statistical relation between 3D seismic attributes and porosity can be inferred from well logs, a Bayesian sequential simulation (BSS) framework proved to be an efficient approach to infer reservoir porosity from an acoustic impedance cube. However, previous BBS approaches only took two variables upscaled at the resolution of the seismic data, which is not suitable for thin-bed reservoirs. We have developed three modified BSS algorithms that better adapt the BSS approach for unconventional reservoir petrophysical properties estimation from deterministic prestack seismic inversion. A methodology that includes a stochastic downscaling procedure is built and one that integrates two secondary downscaled constraints to the porosity estimation process. Results suggest that when working at resolution higher than surface seismic, it is better to execute the workflow for each geologic formation separately.

Accurate inference of the interfaces between geological units showing different hydraulic properties is a key step for a reliable hydrogeological characterization of regional aquifers. In this study, we developed a workflow that combines multiple geological and geophysical data sets having different intrinsic resolution to map a stratigraphic interface of the regional aquifer located in Montérégie, Quebec, Canada. One of the principal goals was to optimally assimilate all the data at all stages in the workflow. Firstly, the experimental variogram showed two structures of different ranges: one coming from highly sampled geophysical data and the other from conventional borehole geological markers. Secondly, a secondary variable is constructed with all secondary data (e.g., geological interpretations of low-resolution electromagnetic surveys), each having its own accuracy and resolution. To account for the variable secondary data, different reliability indexes were assigned as weights in a discrete smooth interpolation (DSI). Thirdly, a classic kriging with an external drift (KED) operator was used to interpolate the more reliable well data on the entire region. The approach was tested on the estimation of the bedrock interface elevation in a regional hydrogeological characterization study. The resulting map shows bedrock elevations coherent with geological structure of the region, representing main features such as outcrops and valleys. A cross section is presented to illustrate the philosophy behind the tools employed to achieve the estimation process. It also shows an example of visual quality control undertaken to validate the workflow.

No abstract