<div><strong>Output type: </strong>Exploring for the Future Extended Abstract</div><div><br></div><div><strong>Short Abstract: </strong>Seismic tomography has been used for more than 50 years to map the seismic velocity structure of Earth’s interior. Here, we use data from the Exploring for the Future program, AusArray 2o deployment, to perform ambient noise tomography of the Australian continent. In this approach, stacks of cross-correlations of background seismic noise recorded by pairs of seismometers are employed to extract an approximation to the surface wave trains travelling between the seismometers. We have developed a semi-automatic approach to estimate dispersion properties of surface waves as a function of frequency at 0.01 – 1 Hz and deployed the largest ever network of broadband seismometers across the country to image the continental crust of Australia. In this study, we present an ambient noise tomography map of the Australian continent at 0.4 Hz (2.5 seconds), which is sensitive to the top 3 km of the Earth’s crust. Our model shows improved resolution across the country, for example, we observed a large low-velocity anomaly (~2.5 km/s) which delineates the shape of the entire Caning basin in Western Australia. This basin has never been imaged at this detail before, as previous tomographic studies do not measure surface wave velocity up to 0.4 Hz and do not have stations deployed in this area. The outcome demonstrates the utility of the ambient noise tomography method of imaging first-order features, that could be built upon for resource potential assessments.</div><div><br></div><div><strong>Citation: </strong>Hejrani B., Hassan R., Gorbatov A. & Zhao J., 2024. Towards continental-scale ambient noise tomography of Australia: a preliminary result from AusArray data. In: Czarnota, K. (ed.) Exploring for the Future: Extended Abstracts, Geoscience Australia, Canberra, https://doi.org/10.26186/149637</div>

<div>Elastic seismic waves propagate in the solid Earth and contain valuable information for inferring subsurface physical properties. Traditionally, the data acquisition requires active artificial sources for generating seismic waves, which can be air-gun arrays for offshore seismic surveys, and explosive or vibroseis sources for onshore seismic surveys. Active-source seismic data have a wide frequency spectrum often from 4 Hz to 50 Hz and contain reflected and refracted P and S waves and surface waves, which have been widely used for obtaining high-resolution images of subsurface structures across industry and academia in the past decades. In this report, we study the feasibility of combining seismic data from borehole drilling (drilling noise) and the state-of-the-art seismic imaging technique, reverse time migration, for a direct migration (imaging) of drilling-generated seismic noise data without intermediate steps (such as seismic interferometry for virtual common-source seismic data), in the aim of obtaining subsurface structural images. The developed imaging method enables informed decision-making during the drilling process by revealing the three-dimensional structure beneath the drill rig. With precise knowledge of the depth to bedrock and the surrounding geological formations, better planning can be achieved, significantly reducing both costs and time.</div>

<div>As part of the first phase (2016-2020) of the Exploring for the Future (EFTF) program, Geoscience Australia deployed 119 broad band seismic stations in northern Australia. This deployment was part of the Australian Passive Seismic Array (AusArray) Project. Data from these stations were used to image the seismic structure using various techniques, including ambient noise tomography (ANT). The first ANT model (Hejrani et al, 2020) was focused on a narrow range of frequencies and used the Hawkins and Sambridge (2019) approach to estimate dispersion curves. This new approach starts from the original work by Aki (1957) to estimate phase velocity in the frequency domain, and then takes a step further to ensure a smooth curve is achieved. In Hejrani et al., (2022), using minimum Signal-to-Noise-Ratio (SNR) threshold of 2, about 4,000 data points (out of 7,000+) were used to generate surface wave velocity maps at a resolution of 1 degree at four frequencies (sensitive to different depths). This model was subsequently updated in September 2021 by using all 7,000+ data points (no SNR threshold) of phase velocity measurements across AusArray year one to provide a 0.25 degree resolution model. All dispersion curves regardless of their quality were utilized. A number of artefacts were identified in that model, which motivated further investigations. During 2022, I developed a new automated and scalable approach to estimate dispersion curves, which was completed in December 2022. This new method starts from the original idea by Aki (1957), but takes a different approach to stabilize the dispersion curves and to avoid cycle skipping. </div><div>This record represents the preferred 2D velocity models for AusArray year one data based on the newly estimated dispersion curves and a comparison with previous models and interpretations; is an update from Hejrani et al. (2020) and should be read in conjunction. Work is currently under way to invert these 2D surface wave models to obtain 3D velocity models for the crust and mantle. Such 3D velocity models would be suitable for joint interpretations with other data such active seismic, gravity and magnetic. The code will be made publicly available at the conclusion of EFTF.</div>

For more than half a century, seismic tomography has been used to map the volumetric structure of Earth’s interior, but only recent advances in computation have enabled the application of this technique at scale. Estimates of surface waves that travel between two seismic stations can be reconstructed from a stack of cross-correlations of continuous data recorded by seismometers. Here, we use data from the Exploring for the Future program AusArray deployment to extract this ambient noise signal of Rayleigh waves and use it to image mid- to upper-crustal structure between Tennant Creek and Mount Isa. Our aim was to establish a repeatable, semi-automatic workflow that can be extended to the entire Australian continent and beyond. Shear wave velocity models at 4, 6, 8 and 10 s periods are presented. A strong low-velocity anomaly (2.5 km/s) at a period of 4 s (~2–4 km depth) delineates the outline of the newly discovered, and prospective for hydrocarbons, Carrara Sub-basin. A near-vertical high-velocity anomaly (3.5 km/s) north of Mount Isa extends from the near surface down to ~12 km and merges with northeast-trending anomalies. These elongate features are likely to reflect compositional variations within the mid-crust associated with major structures inferred to be associated with base metal deposits. These outcomes demonstrate the utility of the ambient noise tomography method of imaging first-order features, which feed into resource potential assessments. <b>Citation: </b>Hejrani, B., Hassan, R., Gorbatov, A., Sambridge, M. Hawkins, R., Valentine, A., Czarnota, K. and Zhao, J., 2020. Ambient noise tomography of Australia: application to AusArray deployment. In: Czarnota, K., Roach, I., Abbott, S., Haynes, M., Kositcin, N., Ray, A. and Slatter, E. (eds.) Exploring for the Future: Extended Abstracts, Geoscience Australia, Canberra, 1–4. <b>See eCat record <a href="https://dx.doi.org/10.26186/148676">#148676</a> for the updated version of the model package.</b>