b'AEGC 2023Short abstractsproperties, not always readily identifiable in geochemical orNext generation land processing gives confidence in petrophysical measurements. Being able to quantitativelyold basinsdescribe texture to integrate with other continuous data is difficult and has been attempted to be done as far back as theGraeme Eastwood 1 Michael Dunne2, Brendon Mitchell2 and 1980s with varying levels of success. Modern computer visionRob Jefferson1techniques can help bridge this gap and allow the integration and subsequent interpretation of all these datasets together-1 Land Seismic Noise Specialists allowing for a quantitative linking between geology and the2 Origin Energydata collected from it. Noise, particularly distortion of signal in the immediate By numerically quantifying texture we can incorporate thissubsurface, is often the dominant problem with land seismic information into modern machine learning workflows. Thisdata and very little has changed with regards to typical land includes both unsupervised analysis, such as clustering, andprocessing workflows in the last 5 years or more. Acquisition supervised analysis for classification. This is particularly usefuladvancements have been made with deployments of more to quantify similarity between images and provide the abilityreceivers and compressed sensing technology, but data to identify a texture that is of interest and quickly query a largequality is still limited by shallow subsurface noise, particularly database of images for other examples of this texture. Thisscattering. New algorithms, which can utilise sparsity to handle technique has been used successfully in industry using otheraliased noise, are able to better remove or account for this numeric datasets and has allowed for a deeper understandingshallow scattering noise and we show here how this has been of mineral systems as well as providing another valuable tool inused to build confidence in vintage data from the Beetaloo Sub-the mineral explorers toolkit. basin. Some 2000 km of vintage 2D data (1989-2012) covering much of the Beetaloo Sub-basin was reprocessed through a Here we demonstrate this approach, using precompetitivemodern workflow aimed at reducing the impact of the noise drilling data sets from the Geological Survey of South Australia.and creating a dataset fit for delineating structural geometries Using the computer vision platform developed by Datarock, weand planning future well and seismic work in the area.quantify features from depth registered drill core image tiles,Key elements of this new workflow were:and then integrate this with downhole chemistry, hyperspectral data and petrophysics. These results demonstrate the powerWave Equation Refraction Staticsa technique which utilises of integrating imagery with downhole data to aid in theenergy after the first breaks to compute statics and phase interpretation and logging of drill core and elucidating thecorrections within frequency bands.relationships between units and holes. Full Waveform Correctionsa surface consistent tool utilising FWI technology to partially undo scattering distortion in the The complex petroleum system of the Rankin Trendnear surface using the reflection data.and Brigadier Trend, Northern Carnarvon Basin,Denoise in various domainsRadon, FK, Cadzow, etc.Australia. New insights from the Triassic MungarooApplicability of processing across all vintages to provide a robust multi-line solution.Formation in Ironbark-1. Carnarvon Basin, Australia.The combination of these technologies has resulted in uplifted Vincent Dykmans and Dan Bishop data quality and a unified dataset, enabling planning for BP Developments Australia Pty Ltd future activities to be made with much more confidence than was possible using previous disparate datasets and hopefully The first significant gas discoveries on the Rankin Trend wereleading to future development of an area discarded by previous made in the early 1970s. The area would prove to be one ofoperators.the most prolific hydrocarbon trends along the North West Margin of Australia. Over 24 Tcf of gas (GIIP) was discoveredSeismic velocity (Vp) vs alteration mineralogy in the in the fluvio-deltaic sands of the Upper Triassic Mungaroo Formation. Yet exploration wells targeting the same stratigraphyOlympic Dam IOCG deposit.45 km to the north on the Brigadier Trend were unsuccessful.Kathy Ehrig 1 David Haddow1, Max Shawcross1, Jared Townsend2 Failure to discover hydrocarbons here has over the years beenand Heather Schijns2attributed to a lack of source rock, lack of cross fault seal, lack of a hydrocarbon kitchen, and unfavourable migration paths. 1 BHP Olympic Dam 2 BHP Group LimitedIronbark-1 was drilled at the end of 2020 to test the deeper gas potential of the Mungaroo Fm in the lower TR20 play interval,Olympic Dam (OD) is a breccia hosted iron-oxide Cu-Au-Ag-U which remained undrilled on the Brigadier Trend. The welldeposit. The OD Breccia Complex hosts disseminated copper reached a TD of 5618.0 mMDRT in the S.speciosus spore-pollensulfide mineralisation and occurs within the Roxby Downs zone and was the deepest penetration of the Mungaroo Fm at theGranite (RDG). The deposit is concealed beneath ~350 m of time. No significant hydrocarbons were encountered. However,undeformed/unaltered flat lying, sedimentary rocks which are the well data, together with new observations and interpretationspost-brecciation and post -mineralisation. The breccia complex of the fluvial system from 3D seismic data, and new geochemicalhas an areal extent of ~50 km2, while sulfide mineralisation data from the wider basin area, provided valuable insights into(~ 0.3% Cu) occurs within a volume of approximately 6x3x0.8 the complex nature of the petroleum system in the area. Thesekm. However, in some areas of the deposit, mineralisation findings assist in further explaining the success of the Rankinextends to 2 km depth. The transition from unaltered RDG Trend and the failure of the Brigadier Trend and will ultimately(K-feldspar, quartz, plagioclase, 10% biotite, amphibole, guide gas exploration in the underexplored parts of the Northernmagnetite and accessory minerals) to the edge of the breccia Carnarvon Basin and Bedout Sub-basin further to the east. complex is gradational and typified by weak potassic and FEBRUARY 2023 PREVIEW 94'