ASEG Extended Abstracts - 1st Australasian Exploration Geoscience Conference – Exploration Innovation Integration, 2018

The presence of chargeable materials can significantly impact the data in electromagnetic (EM) surveys. This affected data has traditionally been treated as noise that must be removed prior to interpretation or inversion. The ability to extract induced polarization (IP) information from an airborne platform would be a valuable tool in the mineral exploration industry, and thus the pursuit of this ability has recently led to significant interest in the interpretation of IP effects in airborne data. A variety of interpretation methodologies have been proposed to aid in the identification and extraction of information from time domain EM data containing IP effects. Any interpretation scheme needs to be thoroughly tested on realistic synthetic examples so that the strengths and weaknesses of the method are well understood.

In this work, we present a methodology for accurately and efficiently simulating the response of a time domain EM experiment by modelling the convolution that occurs in Ohm’s Law in the presence of a frequency dependent conductivity. This method is free of any assumptions about the dimensionality or frequency dependence of the chargeable material and can be used to simulate the response of any time domain system.

Cooperative inversion has the potential to significantly improve subsurface imaging. However, success or failure can be highly dependent on knowledge of underlying site specific geological and petrophysical relationships. Combinations of structural or textural seismic attributes can be integrated into geostatistical clustering to provide a framework able to carry inversion of lower resolution EM or potential field data to an outcome with improved detail and accuracy. Cross-gradient type methods link direction of change of different physical parameters within inversion. Outcomes will be dependent on the presumption that the direction of change of petrophysical parameters like velocity, density and electrical conductivity are indeed linked. Cooperative and joint inversion need to be validated by information harvested at drill sites. Here new low cost multi-scale, multi-parameter logging while drilling technologies could be designed to feed real time imaging based on cooperative inversion. We will; (i) examine theoretical possibilities, (ii) give examples of practical successes and failures, and (iii) consider the future of cooperative inversion.

Cambrian to Silurian rocks of the Dundas Trough, Tasmania, including the economically important Mount Read Volcanics, follow a curving trend around the northern margin of the Tyennan Block. This arcuate trend has recently been interpreted to be part of an orocline, the eastern limb of which is obscured below Jurassic dolerite. Other workers consider the arcuate trend of the Dundas Trough to be the result of Cambrian rifting, with the underlying basement continuing north below Bass Strait. We tested the orocline hypothesis by palaeomagnetic sampling of rocks from the Owen, Gordon and Eldon groups. Previous palaeomagnetic studies concluded that these rocks were overprinted. By use of modern demagnetisation protocols and principal component analysis of demagnetisation vectors we were able to isolate a pre-deformation characteristic remanence carried by magnetite and hematite from a normal polarity overprint carried only by hematite. The overprint resembles Cretaceous overprints recognised from the eastern Australian seaboard, and thought to be related to Tasman Sea rifting. The characteristic remanence passes the statistical orocline test, confirming the orocline hypothesis for the Dundas Trough. Rocks equivalent to the Mount Read Volcanics therefore continue under Jurassic dolerite cover below eastern Tasmania.

Bone beds containing Pleistocene megafauna fossils dated to between 35ka and 60ka occur near Lancefield (Victoria). The bones lie within clays above gravels and extensive Quaternary basalt flows. Evidence that the bones have been subject to alluvial transportation suggests that profiling the basalt basement for paleo-channels will assist with locating further bone beds. Passive seismic (microtremor) methods, as developed variously for earthquake hazard studies and regolith studies, have been applied to this problem, using both HVSR spectral ratio methods and two-station SPAC (spatially averaged coherency) methods. Clay layers have shear-wave velocities (Vs) in the range 120-180m/s and thicknesses 1.5 to 2.5m. Microtremor data in the frequency band 5-60Hz provides excellent resolution of the Vs and thickness of the clay layers, allowing the bedrock profile to be established to an accuracy of 0.5m or better. The results indicate existence of soft clays some 40m west of an existing excavation site and thus identify a future site for excavation.

Exploration methods, technology and equipment may have changed greatly over the last few decades but still does not replace sound geological principles in making new coal resource discoveries. Starting with a case history of the early big discoveries in the Bowen Basin, Central Queensland, author and co presenter Mr Lex Hansen will provide an overview of the methods, techniques, equipment and successes of exploration during the early 1960’s in the Bowen Basin.

Mr Hansen was a member of the team of 3 geologists working for Utah development Company (purchased by BHP Ltd in the 1980’s) that made the first big coking coal discoveries in the Bowen Basin extending from Blackwater in the south along the western side of the Basin to Goonyella in the north. The majority of these discoveries are still in production today, some 50 years later.

These are compared and contrasted with a more recent case study involving greenfields discovery’s at the Meteor Downs South Project located also in the Bowen Basin Central Queensland. This will provide a unique opportunity to review coal exploration methodology, technology and successes. Co author and co presenter Mr Darren Walker lead the teams responsible for this discovery.

The authors then contrast and compare exploration from the ‘then’ and ‘now’ and discuss the importance of ‘grass roots’ geological field work, data analysis and principles in successful exploration regardless of the perceived exploration or resource maturity of a mineral province. Much has changed in the 60 odd years between the discovery of the world class coking coal deposits along the western edge of the Bowen Basin with the most significant being technology and significant increases in regulation, community expectations and approvals now governing coal exploration in Queensland.

This begs the question as to whether this change has achieved the benefits desired and resulted in further major discoveries, or if its has further hindered the exploration industry and the discovery of the future of the coal resources in Queensland.

The Canning Basin of Western Australia has recently become a site of interest for unconventional hydrocarbons, with several formations within deeper basin depocentres being investigated for resources. Modern petroleum resource evaluation is generally dependent on an understanding of both local and regional stresses, as well as the geomechanical properties of reservoir units. Presently, there are significant gaps in our understanding of these factors within the Canning Basin. This study is part of a greater effort from Geoscience Australia to understand the present-day stress field of northern Australia.

A generally NE-SW oriented maximum horizontal stress azimuth is identified from interpretation of borehole failure in five petroleum wells, and a broadly strike-slip faulting stress regime is interpreted from wireline data and wellbore testing. Variations in stress regime at different crustal levels within the Canning Basin are highlighted by one-dimensional mechanical earth models that show changes in the stress regime with depth as well as by lithology, with a general move towards a normal faulting stress regime at depths greater than ~2.5 km.

A wireline log model predicts carbonate cemented zones within Late Cretaceous to Paleocene reservoir sandstones of the Latrobe Group, Gippsland Basin. Predictions match published evidence. These sandstones were once heavily cemented prior to development of secondary porosity that produced the world-class petroleum reservoirs we see today. Cemented zones that remain must act as obstructions to reservoir fluid migration. They may also react with the mild carbonic acid that will be introduced by CO2 storage operations of the future. Model predictions show that cemented zones are sparse, spatially sporadic and fall well below seismic resolution at modern-day reservoir depths. Their significance and irregular spatial occurrence mean there is a need to map their distribution.

Synthetic seismograms generated for a number of Gippsland Basin wells predict high amplitude seismic reflectors away from major lithostratigraphic boundaries. Many occur where cemented zones are predicted. An investigation of the complex seismic trace demonstrates seismic sensitivity to these zones in the frequency range 100-125 Hz. An elevated moving average of instantaneous frequency correlates with some of them as does a modified instantaneous Q-factor. Others are indicated by a change in the difference of normalised instantaneous amplitude between the original frequency-filtered complex trace and a frequency-filtered complex trace composed of sinusoids with the same magnitude and phase (arithmetic averages of components of the original complex trace passed after frequency filtering). These subtle phase disturbances at high seismic frequencies are hypothesized to be caused by the presence of thin cemented zones. This idea is tested using instantaneous attributes calculated from 3D seismic survey data available across the Gippsland Basin.

A sampling program was undertaken to assess pegmatite geochemistry with the view to using the data to vector towards economic mineralisation. The work has shown that pegmatites hosted in high grade metamorphic rocks (granulite and upper amphibolite facies) are either leucogranites, quartz-feldspar-mica pegmatites or feldspar-rich pegmatites. In contrast those that occur within lower grade metamorphic rocks (lower amphibolite to greenschist facies) are feldspar poor and quartz-muscovite rich, locally highly sheared, with tourmaline on their margins and as replacements of surrounding host sediments.

Element classification diagrams suggest a geochemical continuum from poorly evolved pegmatites hosted in the highest grade metamorphic rocks to highly evolved pegmatites in the lowest grade metamorphic rocks. Spatially, this trend is broadly from the south and southwest toward the north and northeast and from west to east.

Economic element content is similarly divided into two broad groups;

1. Pegmatites within higher grade metamorphic rocks host elevated lead-zinc-silver-manganese.
2. Pegmatites within lower grade metamorphic rocks contain elevated tin, tungsten, lithium, niobium and tantalum. The presence of abundant tourmaline in association with these indicates highly elevated boron.

Elevated base metals and silver in the granulite and amphibolite facies rock is attributed to the elevated nature of these elements in the metasedimentary protolith being largely the Broken Hill Group. Our interpretation suggests these are probably not allochthonous but are nearly in-situ bodies resulting from partial melting of the Broken Hill Group. Their chemistry reflects that of the surrounding rock. Analysis of the spatial distribution of pegmatites with elevated lead-zinc-silver-manganese indicates many of these are located close to known base metal and/or silver occurrences of the Broken Hill types (BHTs).

In contrast the concentration of tin, tungsten, lithium and other incompatible elements in pegmatites which are allochthonous and intrude rocks of lower metamorphic grade, suggests these elements have accumulated as fractionation has taken place.

This study has highlighted three areas of elevated base metal-silver within pegmatites where no known mineralisation is documented either as an occurrence or in historic exploration data. Systematic pegmatite sampling at Broken Hill has the potential to enable vectoring toward buried silver-lead-zinc ores of the Broken Hill type.

IP effects in AEM data are subject of current research around the world, due to the recent recognition of their significance for exploration and general (hydro)geological mapping. There have been success stories and it is now practical to model AIP on thousands of line kms of AEM data. However there still is a need to study more accurately the boundaries of the effect and of its relevance, beyond common past acceptance. In this paper we present a systematic, extended analysis of AIP effect in different AEM (TEM) systems’ data, , based on synthetic modelling of different pseudo geologies. Its goal is to provide a clear overview of possible AIP effects in the data space, without imposing simplistic assumptions (e.g., fixing some parameters to arbitrary values or limited boundaries). We produce 1D forward responses with dispersive resistivity for hundreds of thousands of combinations of Cole-Cole model parameters and AEM system transfer functions. The results are analyzed using various metrics (e.g., sum of negative voltages, exponential fitting) that capture different AIP signatures in the transients. Experiments include half spaces, 2 and 3 layer models, combined with different waveforms, Rx types (dB/dt and B), Tx-Rx geometries, flying heights, transients’ binning, base frequencies. The results, presented as 4D hyperspaces, each with 10⁴transients obtained from the combinations of 4 x 10 different Cole-Cole parameters, allow a clear assessment of the AIP effects over a wide range of geophysical situations. Some of the main observations are: AIP effects are increased most often by the presence of a resistive bedrock, often using slow turn-off of the waveform, are generally better observed recording the B field instead of its derivative, and in any most cases lowering base frequencies to 12.5 Hz. In general, they are more pervasive than previously thought and should be carefully considered in virtually any AEM survey. If present, they can often be sensitive to model parameters down to depth.

The integration of geological modelling, petrophysics and geophysics in a single inversion scheme is a complex and powerful strategy for solving challenges faced in geoscientific resource exploration. Probabilistic geological modelling and geophysical inversion are non-linear processes that show various degrees of sensitivity to uncertainty in input measurements. Using field geological measurements from the Mansfield area (Victoria, Australia) and synthetic geophysical data, we present a synthetic case study investigating the impact of geological uncertainty on constrained joint geophysical inversion. We investigate the influence of uncertain geological measurements on geologically constrained inversion through a sensitivity analysis to uncertainty in orientation data. Probabilistic geological models used to define constraints for geophysical joint inversion are obtained through a Monte-Carlo based uncertainty estimation (MCUE) method. We simulate a broad range of possible cases through a parameter sweep on uncertainty levels in geological measurements to provide a reference for practitioners. The analysis and comparison of the results at varying uncertainty levels show that results can be grouped into two main categories. For the highest uncertainty levels, significant portions of the models retain the characteristic features of geologically unconstrained inversions. Meanwhile, below a threshold in uncertainty level, inversion benefits from the interaction of geophysical data and geologically conditioned constraints. In such cases, inverted models are improved compared to both the geological modelling alone and geologically unconstrained inversion. The conclusion of this work is that knowledge of this threshold is critical for the interpretation of results and decision making because it indicates whether the datasets provide enough information to take advantage of the complementarities between geological modelling and geophysical inversion. Knowledge of this threshold can also support decision making pertaining to inversion strategies and geological field data collection.

There is growing evidence that the distribution of significant giant magmatic and hydrothermal ore deposits are linked to the presence or absence of metasomatised lithospheric mantle. It follows that mapping the distribution of this mantle should be an important component of exploration programs for world class deposits, yet to date there has not been a robust means of spatially constraining the distribution of metasomatised mantle. Classically, metasomatism has been identified through petrological and geochemical analysis of mantle xenoliths and mantle derived melts which provide information on the vertical distribution of metasomatism beneath magmatic centres. Here, we show this classical information integrated with constraints on lithospheric thickness and conductivity, derived from passive seismic and magnetotelluric imaging of the lithosphere provide an effective means of mapping both the lateral and vertical distribution of mantle metasomatism. As a case study we show the integration of the aforementioned datasets over south-eastern Australia where the Australian Lithospheric Architecture Magnetotelluric Project (AusLAMP) was started.

In May 2014, a severe storm caused substantial damage in the Balkan area by floods and landslides. As a contribution of geophysicists and geotechnical engineers to the effort of prevention of further damage, a Geoscientists without Borders (GwB) project was organised by Association of Geoscientists and Environmentalists of Serbia to assess the potential of further occurrence of landslide in the region supported by SEG and many other organisations, governments and individuals of many countries. Local and international experts conducted field data acquisition with students from four countries. The project benefited the students to get practical experience in geophysical fieldwork, local governments received information of landslide risk in their area and the residents of the area were made aware of landslide potential of around their home land.

Geophysical surveys with seismic and electric methods were carried out in three phases, June and September 2015 and June 2016, in six locations in Serbia and Bosnia and Herzegovina. About 7000m of seismic data are acquired in the sites where landslide potential is considered high. Lesser amount of electric survey was conducted in the same locations.

This paper presents some of the result of the geophysical surveys of some of the project areas comparing seismic reflection, MASW and electric resistivity methods, and subsequent assessment of risk of landslide. This information is used by the engineers of local government in their plan of mitigation of disasters.

Seismic surveys are routinely used for building precise structural images of the coal bearing formations in Australia but coal production related hazards such as weak strata and zones with an increased gas content remain to be fully resolved by seismic measurements. One way of investigating these issues is through the application of Full Waveform Inversion (FWI) methods. To utilise the full power of these methods a high quality seismic dataset is needed. Such conditions are often met by 2D and 3D reflection seismic data acquired over coal seams in Bowen and Sydney basins. FWI can be used for a high resolution estimate of P- and S-wave velocities and the density that can also be translated into geotechnical parameters of interest to mining operations. In this study, we evaluate the applicability of FWI methods for estimating elastic parameters (P and S wave velocities and density) from the inversion of a synthetic seismic dataset that was recorded over the surface of a 2D earth model that represents subsurface geology in Goonyella coal mine in Queensland, Australia. We generated elastic synthetic shot records using finite difference algorithm and inverted these data back for model parameters to assess the potential of the FWI algorithm. Using only a short array of surface receivers (cheaper option), we show that the application of FWI method can still improve the original earth models towards the true solutions. We were also able to reconstruct the elastic boundaries for a major part of the subsurface models within the seismic bandwidth. Interpretation of the estimated parameters for coal mining objectives is then straightforward.

The Browse Basin hosts considerable gas and condensate resources, including the Ichthys and Prelude fields that are being developed for liquefied natural gas (LNG) production. Oil discoveries are sub-economic. This multi-disciplinary study integrating sequence stratigraphy, palaeogeography and geochemical data has mapped the spatial and temporal distribution of Jurassic to earliest Cretaceous source rocks. This study allows a better understanding of the source rocks contribution to the known hydrocarbon accumulations and charge history in the basin, including in underexplored areas. The Jurassic to earliest Cretaceous source rocks have been identified as being the primary sources of the gases and condensates recovered from accumulations in the Browse Basin as follows:

1. The Lower-Middle Jurassic J10-J20 (Plover Formation) organic-rich source rocks have been deposited along the northeast-southwest trending fluvial-deltaic system associated with a phase of pre-breakup extension. They have charged gas reservoired within J10-J20 accumulations on the Scott Reef Trend and in the central Caswell Sub-basin at Ichthys/Prelude, and in the Lower Cretaceous K40 supersequence on the Yampi Shelf.
2. Late Jurassic-earliest Cretaceous J30-K10 source rocks are interpreted to have been deposited in a rift, north of the Scott Reef Trend and along the Heywood Fault System (e.g. Callovian-Tithonian J30-J50 supersequences, lower Vulcan Formation). The J30-K10 shales are believed to have sourced wet gas reservoired in the K10 sandstone (Brewster Member) in the Ichthys/Prelude and Burnside accumulations, and potentially similar plays in the southern Caswell Sub-basin.
3. The organic-rich source rocks observed in the Heywood Graben may be associated with deeper water marine shales with higher plant input into the isolated inboard rift. They are the potential source of fluids reservoired within the Crux accumulation, which has a geochemical composition more closely resembling a petroleum system in the southern Bonaparte Basin.

CO2 storage in the subsurface is a key aspect of climate mitigation. The UQ is investigating whether the Precipice Sandstone and Evergreen Formation in the Surat Basin, Queensland, are an appropriate reservoir-seal pair for the long-term storage of greenhouse gases. However, the Precipice-Evergreen succession remains poorly constrained from a paleo-depositional and stratigraphic standpoint. Studies have mostly applied lithostratigraphy for local correlation, and the understanding of time-stratigraphic relationships across the basin needs development. This has greatly hindered the capacity to construct robust reservoir models and is an active area of research.

We utilized core, wireline logs, seismic reflection surveys, and pressure data to compare the dynamic response to various CO2-injection scenarios with contrasting stratigraphic architectures. A lithostratigraphic prediction of reservoir and seal intervals consisted of a layer-cake model of fluvio-deltaic deposits. The models suggest that reservoir layers are laterally well-connected with the gas plume primarily migrating parallel to bedding. In contrast, a sequence stratigraphic arrangement of facies resulted in a more complex architecture, where reservoir and non-reservoir strata cross-cut and intersect one another. The resulting models showed greater reservoir heterogeneity and potentially more complex fluid transmission pathways in both the lateral and vertical directions that could result in slower plume migration and more residual trapping. This is due to the fact that discontinuous mudstone intervals potentially baffle the CO2 plume and may allow for more CO2 trapping within the lower parts of Evergreen succession. The contrasting models show different geological realizations arising from the same dataset, interpreted in different ways than identify where there is uncertainty. They may highlight certain areas of the basin are more conducive to carbon storage than others.

Fluid flow is highly sensitive to the stratigraphic arrangement of reservoir and non-reservoir intervals. Refining static and dynamic models using sequence stratigraphy may result in a significant improvement in history matching. Modellers should carefully consider the implications of stratigraphic correlations during static model construction.

The Hera Au-Pb-Zn-Ag deposit is located within the Cobar Basin of the Lachlan Orogen. The deposit was classified as a Cobar-type deposit, however the presence of skarn gangue mineral assemblages associated with the main mineralisation and host-rocks suggest a much higher temperature of ore formation. These gangue assemblages also are the first occurrence of these for deposits within the Cobar Basin. The gold at Hera shows poor correlation with the other metals and minerals suggesting multiple generations of crystallisation. The primary silver minerals in the deposits are Ag-sulfosalts (tetrahedrite), acanthite and dyscrasite?. The presence of Ag-sulfosalts often closely associated with gudmundite within the main mineral assemblages. Scheelite is quite common and associated with the main sulphides and also within the host siltstone.

This paper addresses the question of how to include structural information, for example from a magnetic image, into an airborne electromagnetic (AEM) inversion. The kind of information we are interested in is the trend directions seen in the magnetic image, such as strike directions of dipping bodies, or the shape of palaeochannels.

A commonly-used technique for including prior information is to use a model covariance matrix, describing the spatial covariance between different model points. However, these covariances are usually constructed from a stationary covariance function which is dependent on the vector distance between two points, but is the same for the entire model. However, if a palaeochannel is visible in the magnetics, then we know that the AEM model is more likely to be similar along the channel than away from the channel. We therefore wish to construct a covariance matrix that can take curved and branching structure into account.

We construct an inhomogeneous covariance matrix from an image by breaking the image up into multiple windows, and then computing an elliptical distance metric in each window, such that distances in the direction of the features in that window are shorter than distances across those features. This collection of distance metrics then allows us to compute, between any two points in the image, a shortest path that curves to follow the directions of trends in the image. Using this curved-path distance allows us to generate a covariance matrix that encourages the inverted model to follow the trends in the image.

Geoscience Australia (GA), as a part of the Exploring for the Future (EFTF) Programme, is aiming to create a high-resolution three-dimensional (3D) seismic model of Australia to infer physical properties of the lithosphere from depths of few meters to hundreds of kilometres. This work is based on new data collected from National Seismological Network and a new movable seismic array complimented by legacy seismological data obtained by universities. GA has deployed a movable array of 115 broadband seismic stations for one year between Mount Isa and Tennant Creek arranged in a grid pattern with interstation distance of approximately 55 kilometres in order to attain horizontal resolution of at least 20 kilometres. This dense network is reinforced by fifteen semipermanent higher sensitivity broadband seismic stations located predominantly in the Northern Territory and Western Australia in order to increase imaging resolution within the array and within areas where the National Seismological Network has gaps. Multiple seismological methods are being combined together to obtain robust constraints on 3D lithospheric architecture. For the first time, particular attention is focused on shallow structures located at depths of less than 1 kilometre.

The construction of geological and geotechnical models in typical Pilbara iron ore environments is vital to enable an optimized mine design for the life of the asset, while maintaining pit wall integrity and overall mine safety. Geotechnical assessments require the measurement of geomechanical properties, such as the triaxial shear, direct shear and unconfined compressive strength tests and pressure and shear wave velocities on diamond core samples. Ideally, these velocities would be measured in Reverse Circulation (RC) boreholes as their spatial density is far higher than diamond drilled holes. Unfortunately, despite its value, such data is seldom collected as a large proportion of the holes are above the water table, limiting the use of sonic-logging tools. Even if measurements are possible, damage to the borehole caused by drilling biases the resulting velocity measurements.

This paper details the results of a trial using the vertical seismic profile method to directly measure in-situ seismic velocities in RC boreholes. The method was successful in determining the velocities of the formations through the entire length of the holes. The data in several boreholes was of sufficient quality for the application of more advanced processing methods, important for geological mapping and the processing and interpretation of surface seismic data.

The success of this first trial has implications for future iron-ore developments in the Pilbara. The widespread acquisition of accurate seismic velocity data is likely to enable the creation of more accurate geotechnical models and could improve future development decisions.