Accurate seismic velocity model is essential for depth conversion and rock property determination in the context of fluid flow modelling to support site selection for secure storage of carbon dioxide. The Bonaparte CO2 Storage project funded by the Australian Government will assess the carbon dioxide geological storage potential of two blocks in the Petrel Sub-basin on the Australian NW Margin. These blocks were offered as part of the 2009 release of offshore areas for greenhouse gas (GHG) storage assessment. The Petrel Sub-basin is a northwest-trending Paleozoic rift within the southern Bonaparte Basin. The geological reservoirs of interest include the Jurassic Plover Formation and the Early Cretaceous Sandpiper Sandstone. Primary and secondary seals of interest include the Late Jurassic Frigate Formation and the Cretaceous Bathurst Island Group (regional seal). Trapping mechanisms for injected CO2 may include faulted anticlines, stratigraphic traps, salt diapirs and/or migration dissolution and residual trapping. Water depths are generally less than 100m and depths to reservoir/seal pairs range between 800-2500m below the sea surface. All three main types of seismic velocity measurements are available within the area of our study: velocities derived from stacking of multi-channel reflection seismic data; velocities determined in the process of ray tracing modelling of large offset refraction data acquired by the ocean bottom seismographs (OBS) along the coincident reflection/refraction transect, and velocities from well log (sonic, vertical seismic profiling and check shot) measurements.

A method for calibrating seismic stacking velocities against velocities from well measurements has been developed to quantitatively assess the validity of stacking velocities in the vicinity of boreholes and to improve quality of stacking velocities for use in regional depth conversion of interpreted seismic horizons. Accurate depth conversion of seismic interpretation is vital for use as constraints in gravity modelling and in other basin modelling tasks. Examples of this methodology are given for the northern Perth Basin, Australia. The suggested workflow for calibrating seismic stacking velocities against well velocities in a simplified form is as follows: 1. Check each velocity dataset for errors 2. Modify the datum of each dataset to the sea floor 3. Convert all datasets to two-way time and depth domain 4. Resample all velocity datasets to the same two-way time intervals 5. Cross plot stacking velocity depths near a well site with corresponding well depths for equal two-way times 6. Fit a linear polynomial to this cross-plot (higher order polynomials were tried also), and determine calibration coefficient from the gradient of the polynomial. 7. Grid calibration coefficients 8. Multiply depths derived from stacking velocities by calibration coefficient grid An assessment of depth conversion errors relative to wells shows that this methodology improves depth conversion results to within ±50 m down to the maximum well depth analysed (3.5 km below sea floor); this depth uncertainty translates into a modelled gravity anomaly error of about ±20 gu, which is acceptable for regional scale gravity modelling.

TBA

Stacking velocities for surveys 1001 (Shell Petrel) and 1053 (Esso R74A) over the Bremer and Denmark Sub-basins were analysed for depth and time.

The northern Perth Basin is an elongate sedimentary basin, located off the southwestern margin of Australia. The basin is prospective for petroleum resources, but is relatively under-explored, and the nature of the sediment-basement contact is relatively unknown due to a high degree of structuring and deep basement depth inhibiting seismic imagining. Accurate depth conversion of seismic interpretation is vital for use as constraints in gravity modelling and in other basin modelling tasks, but depth conversion requires good quality seismic velocity information. The number and distribution of wells with velocity information in the northern Perth Basin is poor, but there exists a large amount of seismic stacking velocities. Seismic stacking velocities are an outcome of seismic processing and are thus not a direct measurement of the speed of sound in rocks. To improve the quality of stacking velocities we propose a methodology to calibrate stacking velocities against well velocities, which is as follows: 1. Check each velocity dataset for errors 2. Modify the datum of each dataset to the sea floor 3. Convert all datasets to TWT and depth domain 4. Resample all velocity datasets to the same depth intervals 5. Cross plot stacking velocity depths near a well site with corresponding well depths 6. Fit a linear polynomial to this cross-plot (higher order polynomials were tried also), and determine calibration coefficient from the gradient of the polynomial. 7. Grid calibration coefficients 8. Multiply depths derived from stacking velocities by calibration coefficient grid An assessment of depth conversion errors relative to wells shows that this methodology improves depth conversion results to within ±50m; this depth uncertainty translates into a gravity anomaly error of about ±20 gu, which is acceptable for regional scale gravity modelling.

Despite long history of studies the Wallaby Plateau offshore Western Australia remains a controversial feature. Analysis of interval seismic velocities from Geoscience Australia's 2008/09 seismic survey 310 in conjunction with seismic reflection interpretation provides new insights into the geology of the Plateau. Seismically distinctive divergent dipping reflector (DDR) packages have been identified. The seismic character of the DDR packages is similar to seaward dipping reflector (SDR) packages of inferred volcanic composition. Initial analysis of seismic velocity profiles indicated affinities between the DDR packages and known sedimentary strata in the Houtman Sub-basin. Effect of water loading on seismic velocities is commonly ignored in offshore studies. However, direct comparative analysis of interval velocity patterns between areas of significantly different water depth requires various water pressure related changes in velocity to be accounted for. There are controversies in methodology and application of water depth adjustment to seismic velocities, and presentation of velocity models as function of pressure rather than two-way time, or depth emerges as a possible solution. Water depth adjustment of seismic velocities analysed in our study reduces distinction between SDRs, DDRs and sedimentary strata such that discrimination between volcanic and sedimentary strata in DDR or SDR packages is equivocal. A major uncertainty of this interpretation is due to a lack of the reference velocity model of SDRs and DDRs investigated globally.

Geoscience Australia acquired the Canning Coastal Deep Crustal Seismic Survey in 2014. The survey involved the acquisition of seismic reflection and gravity data along two traverses, 14GA-CC1 (562km) and 14GA-CC2 (143km) between Port Hedland and Derby, WA. The purpose of the survey was to image the crustal architecture of the geology underlying the Canning Basin and its relationship to the boundaries between the crystalline hard rock areas of the North (Kimberley) and West Australian (Pilbara) cratons. As well as establishing the subsurface extent of the Canning Basin and the extent and nature of its sub-basins and troughs. The project was collaboration between the Geological Survey of Western Australia and Geoscience Australia with funding from the Western Australian Royalty for Regions Scheme. Raw data for this survey are available on request from clientservices@ga.gov.au

Geoscience Australia in collaboration with the Geological Survey of NSW acquired the Yathong Trough Deep Crustal Seismic Survey in 2013. The survey involved the acquisition of seismic reflection and gravity data along two traverses, 13GA-YT1 (98km) and 13GA-YT2 (132km) near Hillston, NSW. The purpose of the survey was to acquire new data to better understand the regional geology and major structured of the Yathong Trough within the Darling Basin, NSW. Funding was from the Geological Survey NSW through the New Frontiers Initative. Raw data for this survey are available on request from clientservices@ga.gov.au

Reflection and refraction seismic work was done in 1960 to complete a reconnaissance survey which was commenced in 1959 across the northern part of the Surat Basin. A reconnaissance line now extends in an easterly direction from 30 miles west of Surat to Jondaryan, and this line is also tied to the geologically well-known Roma area. Two good marker horizons have been established in the seismic work - one a strong reflector and the other a refractor in which the velocity averages 19,000 ft/sec and which may represent basement. A deep trough of sediments, possibly 20,000 ft thick at Meandarra, exists between Surat and Tara, and there is a large uplift west of Tara. The eastern margin of this large trough is 12 miles east of Tara, but sediments about 4000 ft thick probably continue to the east, at least as far as Jondaryan.

The preliminary investigation was made when the Bureaut s seismic party was held up by flooded rivers, while on its way t o Christmas Creek in May, 1954. Results show that the seismic aethod is applicable to the Broome area, and that a sedimentary section of the order of 12,500 feet exists. They further show that a syncline and anticline not known from the surface geology may possibly exist at depth.