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

Understanding geomechanics influence early in the field development phase facilitates reservoir management planning. To capture complex geology and associated field development, a 3D Mechanical Earth Model (3D MEM) with Finite Element Method (FEM) approach was selected to analyse the geomechanical related risks associated for two fields in North West Shelf, Australia.

The 3D MEMs were constructed using the geological static models, and seismic derived horizons and faults. The 3D properties were propagated based on core calibrated 1D properties and controlled by stratigraphy, 3D facies and seismic inversion volumes. FEM was used to calculate the equilibrium of stresses and strains within the 3D MEMs. The 3D properties and pre-production stresses were validated in blind test wells prior to forward modelling. The 3D MEMs were linked to the dynamic reservoir models to capture the pressure evolution throughout the field lifecycle.

The results were used to analyse the risks associated with compaction, subsidence, fault instability, completion integrity and drilling stability of infill wells through depleted reservoirs. The results provided insight in managing the risk early in field development stage.

The study’s largest challenge was integrating large volume of data to ensure that the structural complexity and rock heterogeneity are captured and consistent with the geological understanding of the field. A multi-disciplinary team of Earth scientists, reservoir, and geomechanics engineers worked together, and the value of data integration, good communication and teamwork were key success factors. Lessons learned and best practices were captured throughout the study and provided valuable feedback for future works.

In Western Australia, iron ore deposits are divided into three types, Channel Iron Deposits (CID), Detrital Iron Deposits (DID) and Bedded Iron Deposits (BID). The latter are hosted in the banded iron-formation (BIF) of the Marra Mamba and Brockman Iron Formations. Two types of BID are mined, the Martite-Microplaty and the Martite-Goethite (M-G), the latter representing the current bulk of the direct shipping ore. The Upper to Lower Miocene CID have been successfully dated using the (U-Th)/He method. Here we report the first dating of three samples from M-G mineralisation occurring in the Dales Gorge member from the Brockman Iron Formation. In unweathered MG ores, the original texture of the BIF, from the micro- to the macroscale, is fully preserved. The primary hematite is unchanged, but primary magnetite is oxidised to martite and the original gangue minerals such as chert, silicates, and iron-rich carbonates are pseudomorphed by goethite. Three samples were collected from a diamond drill hole at depths of 163m, 116m and 90m from the surface. The samples were characterised using an optical microscope and after microscopic identification of goethite and martite, samples were microdrilled. The average ages of iron oxides were 45 ± 6 Ma or Mid-Eocene for the sample at 163 m; 42 ± 5 Ma or Mid-Eocene for the sample at 116 m and 28 ± 3 Ma or Early Oligocene for the sample at 90 m.

Small conduit or chonolith style intrusions dominated by olivine- and pyroxene-rich cumulates are well known to be favourable hosts to magmatic Ni-Cu-(PGE) sulfide mineralization. In many cases, such mineralization is closely associated with partially assimilated country rock xenoliths, volatile-enriched vari-textured or taxitic rocks and other evidence for assimilation of country rocks. We used the technique of microbeam XRF element mapping along with LA-ICP-MS to reveal other distinctive features and chemistries of a number of diverse mineralized small intrusions: transient saturation in Cr-rich spinel; complex zoning patterns of Cr in cumulus and poikilitic pyroxenes, and occasional development of dendritic growth textures in olivine. We further suggest that these features are indicative of dynamic assimilation of conduit wall rocks accompanied by rapid, disequilibrium fluctuations in silica content and redox state, and that these features may be if not diagnostic then at least indicative of Ni-Cu sulfide mineral potential in magmatic conduit systems.

In situ overpressures in sedimentary basins are commonly attributed to disequilibrium compaction or fluid expansion mechanisms, though overpressures in shallow sedimentary sequences may also develop by vertical transfer of pressure from deeper basin levels, for example via faults. Mafic sill complexes are common features of sedimentary basins at rifted continental margins, often comprising networks of interconnected sills and dikes that facilitate the transfer of magma over considerable vertical distances to shallow basinal depths. Here we document evidence for deep sills (depths >5 km (16,000 ft)) hosting permeable, open fracture systems that may have allowed transmission of overpressure from ultra-deep basinal (>7 km (23,000 ft)) levels. Most notably, well 214/28-1 encountered overpressured, thin (<8 m (26 ft)) and fractured gas-charged intrusions, which resulted in temporary loss of well control. While the overpressure could reflect local gas generation related to thermal maturation of Cretaceous shales into which the sills were emplaced, this would require the overpressures to have been sustained for unfeasibly long timescales (>58 Myr). We instead suggest that transgressive, interconnected sill complexes, such as those penetrated by well 214/28-1, may represent a previously unrecognized mechanism of transferring overpressures (and indeed hydrocarbons) laterally and vertically from deep to shallow levels in sedimentary basins, and that they represent a potentially under-recognized hazard to both scientific and petroleum drilling in the vicinity of subsurface igneous complexes.

Electromagnetic logging while drilling is commonly used to infer information about the electrical properties around the wellbore and to aid in geosteering. Data from modern tools, which combine multiple transmitter and receiver orientations and offsets, can be difficult to manually interpret in all but the simplest of environments. Inversion is required to optimally extract and use the information from this data. Although low dimensional inversions can provide useful information in certain environments, full, 3D solutions are required to extract the maximum possible amount of information from the data.

In this work, we present the first fully 3D inversion of electromagnetic logging-while-drilling data. Moreover, we demonstrate that using semi-structured meshing and mesh decoupling, along with advanced data integration techniques, enables the inversions to be performed in real time.

A regional, high-quality seismic traverse from the coast to oceanic crust across the Bight Basin has been assembled and interpreted in detail, then balanced, restored, decompacted and replaced at paleo-water depths. The Late Cretaceous Ceduna Delta developed above a Late Jurassic-Early Cretaceous rift basin in three stages punctuated by significant pulses of uplift and erosion across areas >100 km wide and with up to 1 km of erosion. The Cenomanian White Pointer delta prograded into deepening water and hence underwent gravitational collapse. This was terminated in the Santonian when the Antarctic margin was pulled out from below, thus supplying heat to a remnant thicker outer margin crust causing doming and erosion. Importantly, this established the saucer-shaped geometry of the Ceduna Delta that persisted throughout its development, so that any hydrocarbons generated in the southern half of the basin would have migrated towards this outer margin high. The Tiger Formation was deposited in shallow water in a full rift basin prior to breakup which was followed by regional thermal subsidence. The Hammerhead delta developed on the newly formed passive margin, but was terminated by another pulse of uplift and erosion, perhaps associated with a Paleogene change in plate motion at the end of the Cretaceous. Finite element modelling of this proposed tectonic evolution will test its validity and predict hydrocarbon generation and migration through time.

Sand dune is known as sand grains and it is formed and accumulated by wind. Sand Dunes can be found in many regions such as Rubaii Alkhali in Saudi Arabia and sub-Sahara Africa. Distributed acoustic sensing (DAS) is a new technology that can be applied for detecting acoustic waves using a fiber-optic cable. The objectives of this study are to compare different methods of acquisitions in order to image the near surface covering with very loosy sand dunes. We used the conventional seismic acquisition method and DAS. In addition, we used in this test different kinds of DAS cables with different depths over the sand.

Direct current resistivity (DCR) method is one of the most commonly applied geophysical exploration methods. The development of data acquisition techniques enables the acquisition of multiple data sets of various electrode arrays with a little extra measurement time in comparison with the time that needs to install the system. Accordingly, the data processor is required to utilise as much as possible useful information to build a more reliable geoelectrical model. This study aims to test using the co-operative inversion process to the multiple data sets of various electrode configurations. We use a synthetic model with the most common electrode arrays: Wenner-Schlumberger (WS), Dipole-Dipole (DD), Pole-Dipole (PD) and Pole-Pole to investigate the possibility of the co-operative inversion schemes. The results show that the co-operative inversion of the combined data sets is better than the inversion of the individual ones. The order of inversion for each data set can produce different results. Fuzzy c-means constraint may assist the inversion to produce better results.

Geophysical data is one of the key decision support factors for successful discovery of hydrocarbons, and operators have multi-decadal inventories of survey data in key basins. Woodside has a history of exploration in Australia from 1954 and seismic data from as far back as 1964 for the Northwest Shelf and other global exploration areas. Since 2017, Woodside has been developing a cloud based Next Generation Seismic Data Management platform that will allow end users to extract maximum value from legacy and modern data.

The platform is open and vendor agnostic for storage platforms and applications and services consuming data. It uses industry standard and public domain standards for data structures, APIs, and transfer protocols, and streamlines access and delivery of seismic products across the data lifecycle, from acquisition to regulatory submission and divestiture.

In 2019, Woodside data management received executive approval to accelerate data re-baselining to have all Woodside seismic data transcribed from magnetic tape media into cloud storage, while enabling direct cloud delivery of acquired data. At completion, this project will have handled over 100,000 individual pieces of storage media, over 200,000 documents, 650,000 individually indexed data objects representing 54,000 2D lines and 400+ 3D surveys, over one trillion seismic traces, and an additional 200,000 spatial data objects representing available multi-client seismic data.

The total volume of data from tape exceeds 20 petabytes, a volume nearing that kept on tape by the Curtin Institute of Radio Astronomy as a precursor to an instrument that will at full capacity require data links for volumes equivalent to all internet traffic. The primary conclusion of the project work is that cloud storage methodologies have enabled all the goals of the project and led to additional value and benefits beyond the anticipated scope.

During the drilling process, numerous decisions must be made, often with limited information. In a known area prior knowledge often guides these decisions. In a recent drilling program in the Northern Perth Basin, as part of a research project to monitor the use of aquifers, petrophysical logs were used to help inform some critical decisions. Further analysis of the data collected from the borehole was going to be used to monitor quality, extent and connectivity of the numerous aquifers that make up the Northern Perth Basin. As an aquifer monitoring program, it is critical to ensure that the bore is accessing the aquifer in question. Therefore, accurate knowledge of depth and quality of formation is key. This paper shows how to use a simple set of petrophysical logs including natural gamma, resistivity and borehole magnetic resonance logs can be used to make informed decisions. For example, using a composite log, decisions were able to be made on screen placement prior to running, and also helped decide if extra bores on the same drilling pad were required. Finally, we were able to determine the salinity of ground water from wireline logs. The ability to make these decisions with accurate information not only ensures successful well completion but also maximises resource use.

A major constraint on development in the Anangu Pitjantjatjara Yankunytjatjara (APY) Lands is the lack of reliable water supplies. Sandstone strata found within palaeovalleys has long been recognised as potentially productive aquifers in outback South Australia. Although the APY Lands are known to contain such geologic features, their development as a viable water supply has been hindered by an acute lack of knowledge.

An extensive airborne electromagnetic (AEM) survey was acquired as part of the Goyder Institute for Water Research’s Facilitating Long-term Outback Water Solutions Stage 3 (G-FLOWS S3) project in order to improve the understanding of groundwater resource potential in this remote region of South Australia. The AEM data revealed an intricate drainage pattern below the modern-day land surface that represents palaeovalleys. This facilitated the spatial refinement of these palaeodrainage features and provided a better understanding of their overall thickness.

A targeted drilling campaign followed the AEM survey. Whereas the AEM data covers a large area of the APY Lands, drilling focused on two areas near the community of Kaltjiti (Fregon) in the eastern part of the APY Lands. Two hydrogeological test sites helped to understand and monitor the groundwater system associated with the Lindsay East Palaeovalley. Furthermore, drilling determined the true depth of the palaeovalley and provided a stratigraphic record. This enabled the validation of the AEM hydrogeophysical model, which included the identification of groundwater bearing zones within the palaeovalley. In total, 11 groundwater wells were constructed and continuous drill core was collected at the two sites.

Drilling near the centre of the Lindsay East Palaeovalley suggests there are at least three groundwater-bearing zones. The basal coarse-grained sandstone unit of the younger palaeovalley fill shows promise as a productive aquifer, with development yields varying between 5 and 20 L/sec and salinities of <1000 mg/L TDS.

The Gippsland basin geological history is modelled using the Badlands software constrained by a realistic 3D structural and stratigraphic model built in Petrel. The aim is to assess and calibrate the theoretical tectonic and sedimentary models using empirical data for a rift basin. The theoretical models are used to assess and measure the relative effect of significant variables for sedimentary basins, including climate, extension, subsidence, uplift, erosion and sedimentation.

The modelling results indicate several insights for the Gippsland Basin. The initial paleo-topography at ~145 Ma was an extensive highland area. The Early Cretaceous paleo-environment was intracratonic, with sediment transport from east to west, and at some stage included an inland sea. The Mid Cretaceous uplift caused emergence of the entire basin, substantial regional erosion and changed the basin architecture. Subsidence associated with Tasman Sea rifting formed the Central Deep and flipped the fluvial paleo-drainage system towards the east. Latrobe Group sediments filled the basin being progressively transgressed by rising sea level to flood most areas by the Oligocene. The models simulate the progradation of the carbonate shelf sediments, sub-marine channels and anticlines over the basin since then.

Probabilistic inversions of wide angle reflection and refraction data for crustal velocity models are regularly employed to understand the robustness of velocity models that can be inferred from these data. It is well understood that the uncertainties associated with the picks of individual arrivals contribute to overall model uncertainty. Typically only a modicum of effort is devoted to quantifying uncertainty in the traveltime picks; a constant noise estimate is commonly assigned to a given class of arrivals. Further, determining the class of arrivals is often left to the behest of the interpreter, contributing additional uncertainty to the data that is both difficult to quantify and may be altogether incorrect. Given the crucial role data uncertainty plays in characterising model robustness, there is a need to thoroughly and appropriately quantify uncertainty in the traveltime data which itself is inferred from the waveform. Here we propose a method that treats arrival or phase classification as part of the velocity model building (inversion) framework using the well-established expectation-maximization (EM) algorithm.

Managing seismic depth uncertainty is a key consideration in placing horizontal development wells into thin reservoir targets. Seismic depth uncertainty has the potential to erode project value through missed reservoir (late landing or reservoir exits) and/or the cost of corrective actions (e.g. sidetracks). The placement of Laverda Canyon development wells in 2018 utilised a 30 Hz full waveform inversion velocity model and deep directional resistivity data in combination with a near real-time depth update process to optimise well placement.

Utilising these methods resulted in a significant reduction in seismic depth uncertainty which culminated in the final Laverda Canyon development well (LAV04WI) being successfully geosteered for ~ 2,200 m within a 10-15 m thick reservoir with only one reservoir exit.

The widely used scheme of naming basins and their regional subdivisions on the North West Shelf emerged from relatively sparse data collected during the early stages of exploration. Such data allow the recognition of large-scale structures and depocentres with a broadly distinct tectono-stratigraphic framework. While that scheme has endured, the availability of extensive, high quality seismic data and stratigraphic information from numerous exploration wells means that we can define much more precisely the structural elements that comprise the margin and the stratigraphic signatures of the basin fill. This has highlighted some inconsistencies in the existing nomenclature, the presence of structural elements of different ages and the presence of boundaries between basins that sometimes can appear rather arbitrary.

We present a revised map for the North West Shelf that shows the structural elements with distinct tectono-stratigraphic signatures that comprise the margin, and applies a consistent nomenclature to them. The aim is to provide a framework that will allow for the better demarcation of distinct hydrocarbon provinces and improved targeting of exploration programmes. This is a work in progress and we invite you to annotate the map shown on our poster with your own comments or to provide feedback via our blog.

The autocorrelation of the seismic transmission response of a layered medium (autocorrelogram), in the presence of a free surface, corresponds to the reflection response.

Despite many studies on the imaging of local structures through retrieval and forward modelling of stacked autocorrelograms, there is limited work on the inversion of these data. In this study, we show that the probabilistic inversion of autocorrelograms is efficient, and can be used as an alternative imaging tool when other approaches are not applicable. Here, we calculate autocorrelograms of teleseismic P-wave coda recorded on more than 1200 permanent and temporary seismic stations across Australia, and utilise a Bayesian framework to invert these data for imaging the Moho structure. The results show patterns of structures that are consistent with those seen in the Australian Seismological Reference Model (AuSREM).

Nova-Bollinger is a structurally modified magmatic Ni-Cu sulphide deposit in which sulphide melt was mechanically extracted from chonolithic intrusions and concentrated into structurally controlled positions within the footwall country rocks. The fertile intrusions were emplaced late during granulite facies high-temperature low-pressure (HT LP) metamorphism and deformation during Stage 1 of the Albany-Fraser Orogeny. Sulphide mineralisation, occurred initially as a magmatic melt and subsequently as a metamorphic modified melt that was mobilised along foliations and into F2 fold hinges, post-foliation pegmatites and a shear zone in footwall granulite gneisses. Disseminated and net-textured sulphide was retained within the chonoliths. The Nova-Bollinger system therefore represents an important example of a structurally modified magmatic Ni-Cu deposit.

The rapid advancement of computer vision is changing the way many industries approach image analysis. Object detection, image segmentation, textual characterisation and other related processes are helping people extract more meaningful information and relationships from their new and existing imagery. Although more typically applied to photography, this paper shows how computer vision techniques can be meaningfully applied to geophysical and other geoscientific datasets.

An introduction to neural networks is followed by an application of computer vision via the ResNet-50 convolutional neural network to the GGMplus regional gravity model of Australia. This results in the quantitative characterisation of geophysical texture and the ability to spatially query areas for morphological similarity.

Realising and protecting the value of near-field exploration and appraisal is a constant challenge, especially during the recent low oil price cycles. This paper discusses the 2015 Pyxis gas discovery, and the means by which a cost effective and value driven approach to data acquisition and optimisation drove the opportunity.

The Pyxis field is situated approximately 10 km from the drilling manifold and centre of the Pluto Field. We will demonstrate how targeted studies, a carefully selected well location and data gathering programme when combined with a seismic survey synergies with the nearby Pluto Field, delivered a cost-effective, single-well discovery and appraisal.

Pre-drill economic viability of Pyxis was challenged by the stratigraphic nature of the trap, significant seismic data issues, sub-tuning reservoir thickness and limited offset-well information. Opportunistic data gathering and cost-effective, detailed Quantitative interpretation (QI )work allowed these challenges to be overcome, and the prospect to be drilled.

In the success case, the Pyxis-1 exploration well was planned to acquire sufficient data to obviate the need for further appraisal. Post-discovery evaluation continued the low-cost approach, using new Pluto Field seismic to assist reservoir characterisation.

We conclude that this modest-sized, near-field opportunity has been optimized in terms of potential economic viability by using appropriate technology, targeted appraisal, and integration with nearby field activities.