Spatially heterogeneous ground motions generated by earthquakes are partially attributed to near surface conditions (i.e., site response), quantification of which enables the selection and calibration of region-specific ground-motion models (GMMs) used in seismic hazard assessments. The 21 September 2021 MW 5.9 earthquake, the largest magnitude earthquake instrumentally recorded in Victoria, was well recorded across the University of Melbourne’s seismic network (UMSN) in eastern Victoria, providing a rare opportunity to assess the utility of published GMMs for large-magnitude earthquakes, and thereby improve seismic hazard assessment in the region. To benchmark these ground-motion recordings, estimating seismic site response is required, in support of which we present results of site characterisation for 12 UMSN. The analysis of surface wave components of ambient noise using microtremor surveys can provide well constrained shear-wave velocity models for the uppermost few hundred metres of the subsurface. By combining spatially averaged coherency (SPAC) and frequency-wavenumber (FK) array techniques, we retrieve multimode Rayleigh and Love phase velocity dispersion curves and Rayleigh-wave ellipticity, which we jointly invert for site-specific, near-surface shear wave velocity models. Previous studies demonstrate that incorporating dispersion of Love waves in the inversions provide stronger constraints of the uppermost shear-wave velocity structure. These inversions produced a set of representative 1-D shear wave velocity profiles for each site, which will contribute to the Australian Ground-Motion Database - a national repository of earthquake ground-motion data and site metadata for engineering applications. Presented at the 2023 Australian Earthquake Engineering Society (AEES) National Conference

The Surface Geology web map service provides two seamless national coverages of Australian bedrock and surficial geology, compiled at 1:1 million scale (displays only at scales less than 1:1500000), and 1:2.5 million scale (displays only at scales greater than 1:1500000). It also contains 1:5 million scale geological regions and metamorphic geology. The service represents outcropping or near-outcropping bedrock units, and unconsolidated or poorly consolidated regolith material covering bedrock.

The Surface Geology web map service provides two seamless national coverages of Australian bedrock and surficial geology, compiled at 1:1 million scale (displays only at scales less than 1:1500000), and 1:2.5 million scale (displays only at scales greater than 1:1500000). It also contains 1:5 million scale geological regions and metamorphic geology. The service represents outcropping or near-outcropping bedrock units, and unconsolidated or poorly consolidated regolith material covering bedrock.

The Surface Geology web map service provides two seamless national coverages of Australian bedrock and surficial geology, compiled at 1:1 million scale (displays only at scales less than 1:1500000), and 1:2.5 million scale (displays only at scales greater than 1:1500000). It also contains 1:5 million scale geological regions and metamorphic geology. The service represents outcropping or near-outcropping bedrock units, and unconsolidated or poorly consolidated regolith material covering bedrock.

<div>&nbsp;Seismic site classification is essential for seismic hazard analysis as it helps constrain the impact of local geological conditions on the near-surface seismic-wave propagation and observed ground motion. The Southwest Australia Seismic Network (SWAN) temporary array was established to record local earthquakes for seismic hazard applications and to improve rendering of the 3D seismic structure of the crust and mantle lithosphere in southwestern Australia. Notably, the SWAN project has recorded significant seismic events, including the 2022 Arthur River earthquake sequence and the 2023 MW 5.0 Gnowangerup earthquake. These earthquakes, together with other well-recorded events across the SWAN network, offer a rare opportunity to assess the utility of published ground-motion models (GMMs) for large-magnitude earthquakes, thereby significantly improving seismic hazard assessment in the region. Moreover, the importance of site classification is underscored as it is a critical component of GMMs, and can substantially enhance the accuracy and reliability of these models. This study uses microtremor survey methods to estimate the shallow shear-wave velocity profiles and VS30 values, which are the primary factors for site classification at seismic stations. Microtremor array measurements, such as high-resolution frequency-wavenumber and modified spatial autocorrelation methods, were utilized to analyse ambient vibrations, producing detailed dispersion curves for each station. To enhance the depth accuracy of velocity profiles, ellipticity curves were extracted using the RayDec method and jointly inverted with the dispersion curves. Additionally, OpenHVSR software was employed for the inversion of single-station ellipticity curves.</div><div><br></div><b>Citation</b>: Ebrahimi, R and Allen, TI 2024, Site classification and VS30 determination for seismic hazard evaluation in the SWAN seismic network, Western Australia, in South West Australia Network (SWAN): passive seismic imaging and hazard analysis compiled by RE Murdie and MS Miller: Geological Survey of Western Australia, Report 255, p. 58–67