ASEG Extended Abstracts - 2nd Australasian Exploration Geoscience Conference: Data to Discovery, 2019

Australia is at last seeing common-sense with some of its key politicians at State level moving to remove barriers to the use of fraccing as an exploration and production technique. The most recent and politically important have been the decisions of Western Australia and the Northern Territory to allow fraccing under controlled conditions and apparently strict regulations. Based on publicly available material this extended abstract will explore exactly how far those two Governments are prepared to go in 2019 and the best indications of the areas of regulation and the conditions that will be applied.

This abstract will not address any political, social, health or cultural issues focusing only on technical scientific matters including some of the environmental aspects.

Understanding of gas flow through cleats in coal seams is important for many applications from gas content estimation, production rate assessment and well abandonment. Direct observations of the kinetic process of gas diffusion and adsorption in the cleat-matrix structure through experimental studies are difficult to perform. Therefore, robust numerical simulations are needed for investigations. We present a fundamental study of the microscopic mechanism of gas migration by a numerical model of coal seam gas diffusion and adsorption considering effects of surface roughness to study the effects of rough surfaces on the methane transport.

We found that surface roughness has a significant effect on gas transport by considering fractal dimension ranging from 1.5 to 1.8. Increases of surface roughness enlarge the gas-solid contact area causing higher adsorption capacities which decrease the concentration of gas breakthrough. Results of this work can provide an improved understanding of the effect of microscopic mechanisms of methane migration in coal on the overall reservoir scale.

In recent years, magnetotelluric (MT) processing has become computationally intensive as the scale and size of MT surveys being run increases. Consequently, High Performance Computing (HPC) is now becoming a valuable tool for timely processing and modelling of these large MT datasets. As part of the MT component of the 2017-2019 Australian Research Data Commons (ARDC) funded Geoscience Data Enhanced Virtual Laboratory (DeVL) continuity project, The National Computational Infrastructure (NCI) at the Australian National University will enable MT datasets from The University of Adelaide to be added to the NCI HPC platform with the goal of creating a more Findable, Accessible, Interoperable, Reusable (FAIR) and open public resource. A focus will be on making the time series datasets more suitable for use on HPC and more interoperable with other Earth science disciplines, where High Performance Data (HPD) formats will allow for better scalability and performance. Metadata attributes, as defined by the Australian MT research community, will be added directly to the time series data files. Additionally, time series processing and 3D inversion codes are being optimised for HPD/HPC, with the end goal of rapid time series processing and 3D inversion. Making FAIR MT time series available on HPC can lead to a transformative change in the way MT data analysis is routinely conducted and such a change has the capacity to create new ways of doing collaborative and transparent MT analysis.

The main objective of this study is to present the value of multi-hole borehole images interpretation in unlocking the structural configuration of the rock mass in a quick, yet effective way. This analysis revealed that the block is a small fold with moderately dipping limbs with a rock sequence of four main rock types or mechanical units. Fractures are developed only in one particular rock type showing the full control of mechanical stratigraphy on fracture/joint distribution. A fault gouge/broken zone was also identified and correlated in all three boreholes. These results were used to build a conceptual 3D model of the block accommodating both structural features and geotechnical rock types. This which was performed in only 2 days after the data acquisition can be used in any mining process particularly those requiring geological settings understanding of a specific location.

At least four petrographically, geochemically, temporally and spatially distinct dolomites have been identified in mid-late Devonian carbonate banks and reefal buildups of the Emanuel Range on the south eastern Lennard Shelf. Detailed mapping of the distinct dolomite bodies and their relationship to extensional faults and MVT-sulphide mineralisation has allowed the structural evolution of the southeastern Emanuel Range to be reconstructed providing constraints on the structural evolution of the range during oil migration and MVT mineralisation.

The earliest dolomite (DI) is a stratigraphicallycontrolled, finely crystalline, non-ferroan, fabric mimetic syndepositional dolomite that is associated with evaporitic mudflats and predates major fault movement. Dolomite II (DII) is also stratigraphically controlled and is restricted to the basal section of the Argutastrea Unit which overlies the Lower Dolomite Unit. DII comprises fine to medium crystalline weakly ferroan replacement dolomite and cements which formed during early burial (<500 m) when primary porosity was relatively high.

Following extension-related major movement on the Cadjebut Fault during the Mid to Late Devonian, hot 60-106 °C highly saline basinal brines entered the reefs precipitating a medium to coarsely crystalline, variably ferroan Dolomite III (DIII). DIII distribution indicates that brines utilised the Pinnacles and Cadjebut Splay Faults rather than the Cadjebut Fault. The onset of oil migration coincided with the later stages of DIII formation and the onset of MVT-sulphide mineralisation. Leaching of bioclasts associated with DII and DIII resulted in the development of variably abundant secondary mouldic and vuggy porosity which in some cases is maintained in the subsurface forming potential reservoir facies.

The final generation of dolomite is coarsely crystalline, variably ferroan Dolomite (DIV), which can exhibit pronounced lattice curvature and is associated with MVT sulphides. Both Dolomites DIII and DIV are hydrothermal dolomites (HTD) having formed at temperatures up to 30°C higher than maximum burial temperatures of the host sediments.

Kerogen slides were made from 92 core samples selected from the Upper Triassic Mungaroo Formation in four wells in the Gorgon area of the Northern Carnarvon Basin. The slides were examined to investigate relationships between kerogen assemblages and their depositional environments (“depofacies”). Although the assemblages naturally vary and overlap to an extent, each depofacies has a characteristic kerogen assemblage. Moreover, depofacies which are genetically similar tend to have similar assemblages even though they may have been deposited in different parts of the delta. For example, active channels tend to have similar kerogen assemblages (abundant black-opaque particles, few cuticles, sparse palynomorphs) irrespective of whether they are fluvial, crevasse or distributary channels; in this case, it is inferred that the overwhelming factor in the kerogen assemblages is the high energy level of the environment of deposition, and its consequent inhibition of local vegetation and promotion of mechanical degradation of organic particles.

Surfactant is routinely injected after waterflooding where substantial amount of oil trapped within the formation porous media. Surfactant, as one of the promising chemical enhanced oil recovery (CEOR), aims at lowering the interfacial tension (IFT) between the formation fluids. However, during the production, the technique was challenge by surfactant loss due to the adsorption of surfactant onto native rock surface. Surfactant adsorption has been proved to be reversible, provided that a foreign material is introduced in the extended water process, which increases thereby the production cost. In this study rather, we investigated the desorption and subsequently the rate of reversibility by alteration of process water, during initial and extended stages. Using sodium dodecyl benzene sulfonate (SDBS) as surfactant and Berea sandstone as adsorbent, we showed that the adsorption increased with salinity gradient, while desorption exhibited a reverse trend. Reducing water salinity from 3 (or 5 wt.%) to 1 wt.% NaCl, up to twofold of surfactant desorption was enhanced. This increment is owing to a better solubility in a less saline medium water and electrostatic repulsion. We extended the analysis to oil recovery, herein performed by spontaneous imbibition tests, using a light dead crude oil (API 31.06°). The results revealed an increase in oil recovery up to 4.6% of initial oil-in-place (IOIP) during extended water process by salinity alteration. The production was found subsequent to a decrease of about 82% in IFT .Furthermore, microscopic analysis of the sandstone surface after salinity alteration revealed that the increment in oil recovery was inherent to a reduction in pore blockage.

As major mineral discoveries have become rarer over the last two decades, the industry has begun to turn to new technologies to assist in the exploration process. One such advancement is the application of machine learning and artificial intelligence (AI) to geoscience data. Mineral prospectivity mapping has been around for decades but with the increase in computer power, recently it has gained traction again as a means for exploration teams to take full advantage of the numerous datasets at their disposal. Although having a team of human experts with a wealth of geoscience knowledge and experience is still fundamental to the exploration process, the ability to robustly integrate and analyse large geoscience datasets over vast spatial regions quickly becomes unwieldy if done manually.

In this study, we developed a new algorithm for mineral prospectivity mapping using a VNet deep convolutional neural network and applied it to finding gold at the Committee Bay greenstone belt in the Canadian Arctic. The machine learning network took all the geoscience data available from the area and generated a prospectivity map for targeting economic orogenic gold mineralization. The results were subsequently validated on a separate nearby region where the machine predictions were compared to gold assay values from drilling. The gold assays from this region were not included in the training process, and the method demonstrated good success in predicting where the highest gold mineralization occurred.

A subsequent gold prospectivity map was produced for the main area in question, and in addition to many new targets the VNet algorithm predicted many targets that the exploration team had previously generated. This suggests that this process assists the exploration team in vetting old targets while opening their eyes to new targets as well. In this way, the algorithm helps to vector in on prospective new and old areas while maximizing the value of all available geoscience data.

Direct detection of drillable targets is among the primary goals of airborne electromagnetic surveys. As larger government funded regional surveys move towards large (> 5 km) line spacings, it is natural to question the likelihood of such surveys of directly detecting orebodies. A natural subsequent question relates to the degree to which data from such surveys can characterise orebodies. Here we model a Tempest prospecting system and address the first question though a numerical modelling study of 1024 targets with all model parameters randomised. We show that the likelihood of target detection decreases with increasing line spacing. A generally-conductive regolith as it is common in Australia means that target detectability increases with increasing time. We show that while over 80% targets are detectable at 16.2 ms using a line spacing of 1 km, at most 24% of targets are detectable with a 10 km line spacing. We address target characterisation by inverting numerical modelling data. We show that is possible to recover correct parameters within a few percentage points for targets with strike lengths less than 6.5 km provided such targets are bisected by a flight line. Our results suggest that Tempest-based surveys with line spacing less than 2.5 km are required to detect discrete targets in a tenement and that surveys with a line spacing of 1 km are optimal when direct target detection is a survey goal.

This contribution uses several case studies to illustrate how regional aeromagnetic and gravity data is used to undertake tectonic analysis. Regional aeromagnetic and gravity data is a powerful tool for tectonic analysis because it can be interpreted and modelled at different scales, and it is very effective at imaging different crustal levels. The signal in the data can also be linked to geological features and processes, and importantly, it is amenable to structural analysis, which can be used to inform 3D geometry, kinematics, and overprinting relationships. When combined with geological context the data can constrain tectonic settings and evolutions, and importantly provide context for mineral system analysis. We use examples from the IOCG belts of Proterozoic Australia. We present data from the Mount Woods Inlier in the northern Gawler Craton, to this part of the craton is highly extended, resulting the development of a metamorphic core complex. We then illustrate the tectonic setting for IOCG mineralisation in the Curnamona Province, illustrating how structural analysis of the data provides key constraints on tectonic transport direction.

The Bight Basin is considered to be one of Australia’s most prospective petroleum regions. However, the sparsity of geologic data means that potential petroleum plays in the basin have yet to be firmly established. In late 2017, International Ocean Discovery Program (IODP) Expedition 369 extracted 700 m of core from the Upper Cretaceous Tiger Supersequence from Site U1512 in the western Ceduna Sub-basin. The core represents the most substantial lithological dataset obtained from the basin and the first opportunity to characterise the succession. Two dinocyst zones (the Turonian Paleohystrichophora infusorides and the Coniacian Conosphaeridum striatoconum zones) assigned at this locality are consistent with shipboard nannofossil zonations. Facies analysis supports deposition by hypopycnal and hyperpycnal flows in a prodelta setting with seismic data suggesting postdepositional deformation in the upper 350 m of the succession. Palynofacies assemblages are dominated by phytoclasts that indicate close proximity to a fluvio-deltaic source. Source rock analyses reveal samples have limited total organic carbon (TOC) values (<1.52%) with little response for S1 due to the immature nature of the samples. However, other datasets suggest that the base of the hole at Site U1512 was close to intersecting potential organicrich black shales associated with Ocean Anoxic Events 2. The new data provide insights into the Tiger Supersequence in this locality and further work will help refine its character, evolution and the petroleum prospectivity of the region.

In mining areas with limited previous seismic acquisition, 2D seismic lines are often acquired to determine viability of the method and optimal survey parameters in advance of a more comprehensive 3D survey. Consideration of the design and acquisition strategy can allow additional value to be leveraged from the nominally 2D data. At Escondida, four 2D lines were processed to yield a pseudo-3D volume over a portion of the survey area, allowing improved structural interpretation above what could be achieved from 2D imaging alone.

Vehicle towed transient TEM systems have opened up possibilities of comprehensive mapping of groundwater resources at farm scale. At this scale airborne TEM usually has too much setup cost and conventional TEM conducted on foot is usually too laborious and rarely economically viable. Interpretation of data requires understanding of diverse geology, from alluvial geomorphology, volcanogenic deposition, weathering and erosion processes and the often complex combination of these. Further, impacts on the data by fences, buried cables and modelling imperfections must be separated from geological impact. Typically, it is the detail of the survey that renders it interpretable rather than system quality alone.

Zircon is a refractory mineral that is able to crystallize within hydrothermal, igneous and metamorphic environments, resulting in extreme variability in its external morphology, internal textures and chemistry. The chemistry of zircon is sensitive to its source rock type and crystallisation environment. Zircon has a tendency to incorporate a range of minor and trace elements, largely determined by its crystal structure and changes in temperature, pressure and composition. Internal textures of zircon can be used to indicate the type of crystallisation environment, as well as the environment(s) to which it was subjected following crystallisation. Given the complexity of zircon textures and the uncertainty associated with interpreting isotopic ages, trace element analysis is becoming widely applied as another line of evidence for more confident petrogenic interpretation.

Here we present geochemical and morphological data from zircon from the Gawler Craton, South Australia. This area preserves a complex geological history dating back to the late Archean and preserves multiple igneous units that each have their own unique geochemical characteristics and are associated with iron oxide-copper-gold mineralisation. We show that zircon will preserve chemistry reflective of its host rock, and possibly of alteration associated with mineralisation.

The next generation of mineral and energy system discoveries in Australia will be made under deeper cover or water and require knowledge of the deep earth. Primarily, these discoveries will be driven by understanding the lithospheric properties and resource transport that underpin the systems. Nevertheless, our lack of good understanding of the physical-chemical properties of lithospheric scale mineral systems hinders our interpretation of available data and makes predictive models difficult to use. These properties are very difficult to measure in the laboratory but are accessible through molecular dynamic simulations.

We used molecular dynamics simulations to investigate the chemical and physical properties of the NaCl-bearing fluids over wide range of temperature (25-1000 °C), pressure (1-60 kbar) and salinity (0-10 m) using high-performance computers. The equation-of-state, ion association and diffusion constant of NaCl solutions were predicted and fitted into an electrical conductivity model. By integrating predictive geophysical properties with large-scale models, this study will develop essential guides that underpin interpretation of geophysical data for mineral exploration.

3D seismic is routinely used for structural characterisation for underground coal operations, however the implementation of this technology for shallow, open-cut resources has been limited. The perceived high cost of shallow 3D seismic has been one of the main factors holding it back. This paper will demonstrate that by adopting and adapting best practice technologies from the petroleum industry, 3D seismic is not only cost effective but also delivers higher confidence resource knowledge compared to traditional pattern grid structural and coal surface drilling. With higher resolution understanding of structure affecting the coal resource, the business can better predict and mitigate risks such as Geotechnical events, and minimise impacts to production with an improved mine plan.

A series of trial surveys were acquired over a range of geological settings, including variable thickness Tertiary basalt cover, multiple seams and very shallow target seams. Success of the 3D seismic technique in all of these trials validated the larger scale implementation at other open-cut mines. Next, through a process of iterative improvement of parameters and survey design, improvements in productivity and data resolution of these surveys were made to maximise value of information. The third step was to embed the subsurface knowledge obtained from seismic data into mine planning workflows and decisions. This work has already resulted in reductions in drilling and adjustments to geotechnical models.

Current work is focussing on further optimisation of the data acquisition, processing and interpretation workflows to ensure fit-for-purpose, long-lived information is delivered to the business. This includes revisiting and reprocessing legacy 3D datasets to harness advances in processing algorithms.

The central North West Shelf has been the focus of a regional mapping program by Geoscience Australia targeting the Triassic succession. Resulting updates to the regional structural understanding are presented, showcasing variations in structural style across the region. The Triassic section is affected by fault sets with two predominant orientations across the study area: northtrending and northeast-trending, with localised areas of northwest-trending faulting. There is typically vertical separation of faulting between the upper Triassic and lower Triassic successions, resulting in different fault fabrics mapped on the top and base Triassic surfaces. In some areas major faults penetrate through the Mesozoic section and into the Paleozoic basement, forming features with significant displacement such as the Thouin Graben, Whitetail Graben, Naranco High, and the Barcoo Subbasin half graben. Isochore maps reveal two Triassic depocentres separated by an area of thin Triassic extending from the inboard Bedout Sub-basin out to the western Rowley Sub-basin and northeast Exmouth Plateau. This new mapping shows that there is a poor correlation between Triassic depocentres and existing basin boundaries, making it difficult to clearly describe regions of interest and their evolution. Greater integration of new structural insights into a regional structural framework is necessary to improve our understanding of the tectonostratigraphic evolution of the margin, and the stratigraphic and structural aspects of exploration risk.

The massive availability of remote sensing data and advances in data analytics have improved the capacity for mapping land cover and subsurface.

Specifically, in case of multi-sensor remote sensing images, the independent analysis of each image ignores the valuable complementary information available in other images. To address this problem, a framework of multisensor image analysis using fuzzy collaborative clustering is proposed. The proposed framework avoids the independent analysis of each image but combines the information available in one image with the complementary information given by the all other images for improved image understanding and segmentation. Specifically, the clustering of pixels in each image is collaborated with the clustering results of other images to refine its results. The proposed framework can simultaneously process different heterogeneous images from various sensors. The proposed framework was evaluated and validated through an experiment in which two multi-sensor images, i.e. Landsat-5 TM and ENVISAT ASAR were used over the Beijing urban area and compared with the standard fuzzy c-means clustering. Experimental results show that the proposed framework outperforms the independent image segmentation analysis in detecting the urban growth of Beijing. This framework serves as a useful tool for various earth science applications.

We describe our latest developments on the inversion of subsurface electrical resistivity from deep directional borehole resistivity measurements. Two different methods are introduced for fast inversion of tri-axial induction measurements with multiple transmitter-receiver positions. The first one approximates transversely isotropic (TI) formations with a sequence of “stitched” 1D planarly layered TI sections, which efficiently approximates the solution via 1.5D simulations. The second method uses a pixel-based inversion approach and employs full 3D modelling of borehole EM induction with either an iterative or a direct solver. This enables the inversion of spatial distributions of electrical resistivity of any complexity without restrictions on the symmetry of the models. Numerical examples for several challenging synthetic conditions confirm the accuracy and efficiency of the developed algorithms.