Half-page abstract

3D inversion of potential-field data is a powerful technique for investigating subsurface geology. A common problem for any geophysicist doing 3D potential-field modelling is acquiring sufficient computational power to produce models that cover the area of interest with an appropriate resolution. Being limited by computing power often means that models have generally degraded in their resolution or scale to ensure they are computed within available resources, which in turn limits the geologic interpretation. A collaborative arrangement between Geoscience Australia and the National Computational Infrastructure (NCI) hosted by the Australian National University has increased GA's capabilities to ensure that potential-field inversions are calculated at a resolution that truly honours the available national-scale data.

Detailed field mapping between Cloncurry and Selwyn has established the existence of a common stratigraphic/tectonic history of almost all the geology east of the Overhang Shear Zone, a major suture separating the Cloncurry-Selwyn Zone from the Quamby-Malbon Belt and Mitakoodi Block. The major exception is a discrete tectonic belt in the far south of the region, the Gin Creek Block, which forms an anomalous zone of older stratigraphy and high grade metamorphism enveloped by tectonic boundaries with the surrounding units. The Cloncurry-Selwyn Zone itself could be subdivided into several sub-regions with similar internal characteristics, but for simplicity the key findings reveal that there are two principal supra-crustal packages folded and interleaved together along major faults and intruded by 1550-1510Ma granitic rocks.

Deep-water Otway and Sorell basins developed during Gondwana break-up when Australia rifted away from Antarctica. The 2D and 3D gravity modelling in conjunction with seismic and geological interpretation has led us to an improved understanding of basement architecture of the study area. 2D gravity modelling particularly along selected seismic lines reveals a N-S crustal-scale lineament extending down to the Moho. A distinct density contrast of 0.16 t/m3 (3.05 t/m3 and 2.89 t/m3) across the structure points to a significant lithological difference at middle to lower crustal depths, interpreted here to reflect a change from dominantly basaltic to felsic lower crust. This structure is assumed to be inherited from a pre-existing basement structure and supports the hypothesis that the evolution of the Sorell Basin was probably basement controlled. The 2D models also help us to conclude the basaltic underplating in the lower-crustal region resulting from the breakup history, all long the margin. The computed 3D gravitational response of the basin-wide seismic interpretation correlates moderately well to the observed gravity trend, which implies (a) consistency between the seismic and gravity data of the inferred model. (b) Throws some light on basement topography, hence gives an idea of possible depo-centres. The depth to magnetic basement map derived independently from magnetic data has given a close proximity with that obtained from the 3D forward modelling, which essentially enhance reliability on the derived model to a good extent.

Integrated analysis of the landscape of the Kununurra Region in Western Australia, using airborne electromagnetics, sonic drilling, airborne LiDAR and geomorphic mapping has elucidated the way the fluvial landscape has progressively infilled the bedrock topography. These new baseline datasets have been used to inform land management decisions associated with the Ord Phase 2 expansion.

SGA 2009 abstract

In 2008-2009 Geoscience Australia, contracted Fugro Airborne Surveys and Geotech Airborne, to respectively acquire TEMPEST and VTEM airborne electromagnetic (AEM) data with broad line spacings covering more than 71 000 km² in the Pine Creek region, Northern Territory. The Pine Creek survey (Figure 1) is the second regional AEM survey funded by the Onshore Energy Security Program (OESP) at Geoscience Australia. Geoscience Australia funded the flying of 19 500 line km, subscriber companies funded 10 400 line km. The 5 000 m line spacing provide regional information with 1 666 m, 555 m and closer line spacing providing detail for mineral systems analysis and deposit scale mapping. One of the main survey objectives was to reduce exploration risk and encourage exploration in the region by mapping, under cover, in areas where gravity and magnetics are quiet. Geological targets included detecting: conductive unites within the Pine Creek Orogen (PCO) sequence; Kombolgie Sandstone / PCO unconformity; Tolmer Group/ Finniss River Group unconformity. Geoscience Australia undertook conductivity logging (Figure 2) in the Pine Creek region. Conductivity logs were processed and as input into forward models, ground truth AEM results and for geological interpretations. To facilitate interpretation, subsurface electrical conductivity predictions using a layered earth inversion (sample by sample) algorithm developed by Geoscience Australia (GA-LEI) were derived from the AEM survey data. Conductivity characterisation of large regional units using the AEM data show: the Rum Jungle Complex is a consistently resistive area with an average conductivity value of less than 2 m/S; the Mt Partridge Group has a conductivity value up to 100 m/S; the Kombolgie Sandstone has a conductivity range of less than 2 m/S in more areas. Detecting conductivity contrasts in areas with known uranium prospectivity aids in a mineral systems analysis and geological interpretation of uranium deposits.

The Peel 2008 LiDAR data was captured over the Peel region during February, 2008. The data was acquired by AAMHatch (now AAMGroup) and Fugro Spatial Solutions through a number of separate missions as part of the larger Swan Coast LiDAR Survey that covers the regions of Perth, Peel, Harvey, Bunbury and Busselton. The project was funded by Department of Water, WA for the purposes of coastal inundation modelling and a range of local and regional planning. The data are made available under licence for use by Commonwealth, State and Local Government. The data was captured with point density of 1 point per square metre and overall vertical accuracy has been confirmed at <15cm (68% confidence). The data are available as a number of products including mass point files (ASCII, LAS) and ESRI GRID files with 1m grid spacing. A 2m posting hydrologically enforced digital elevation model (HDEM) and inundation contours has also been derived for low lying coastal areas.

The northern Perth Basin is an under-explored part of the southwest continental margin of Australia. Parts of this basin have proven hydrocarbon potential. The basin is extensively covered by mostly 2D seismic reflection data and marine gravity and magnetic data. The seismic data helps to resolve the structural framework of the basin, but in deepwater regions, the basement-cover contact and deeper basement structure are generally not well imaged. To help overcome this limitation, integrated 3D gravity modelling was used to investigate crustal structure in onshore and offshore parts of the basin. Such modelling also relies on knowledge of crustal thickness variations, but these variations too are poorly constrained in this area. Multiple models were constructed in which the seismic data were used to fix the geometry of sedimentary layers and the fit to observed gravity was examined for various different scenarios of Moho geometry. These scenarios included: 1) a Moho defined by Airy isostatic balance, 2) a Moho based on independently-published Australia-wide gravity inversion, and 3) attempts to remove the Moho gravity effect by subtracting a long-wavelength regional trend defined by GRACE/GOCE satellite data. The modelling results suggest that the best fit to observed gravity is achieved for a model in which the thickness of the crystalline crust remains roughly constant (i.e. deeper Moho under sediment depocentres) for all but the outermost parts of the basin. This finding has implications for understanding the evolution of the Perth Basin, but remains susceptible to uncertainties in sediment thickness.

Receiver function studies of Northern Sumatra T. Volti and A. Gorbatov Geoscience Australia, GPO Box 378 Canberra ACT 2601 Australia The Northern Sumatra subduction zone is distinguished by the occurrence of the 2004 Sumatra-Andaman megathrust earthquake and has a peculiar subduction of two major bathymetric structures; the Investigator fracture zone and the Wharton fossil ridge. Four stations in Northern Sumatra (BSI, PSI, PPI, GSI) and two stations in Malaysia (KUM and KOM) have been selected to construct migrated images based on receiver functions (RF) in order to study Earth structure and subduction processes in the region. Waveforms from 304 teleseismic earthquakes with Mb >5.5 and a distance range of 30º to 95º recorded from April 2006 to December 2008 were used for the analysis. The number of RF for each station varies from 20 to 192 depending on the signal/noise ratio. The computed RF clearly show pS conversions at major seismic velocity discontinuities associated with the subduction process where the Moho is visible at 5.5, 4, 3.5, and 2 sec for BSI, PSI, PPI, and GSI, respectively. RF for KUM and KOM have only conversions at the Moho near ~4 sec. The subducted slab is visible below Sumatra as a positive amplitude conversion preceded by a negative one, which we interpret as a low-velocity structure above the subducted slab. RF for PSI located at Toba supervolcano reveal pockets of low-velocity zones extending from a ~50 km depth down to the subducted slab. Forward modellings of RF suggest that seismic velocity contrasts can reach ~18% that is in accordance with previous local tomographic studies.