Surface-wave studies of the Australian region

The subduction zones to the north and east of Australia provide frequent events over a broad range of depths which can be exploited in surface-wave tomography of the Australian region. The mid-ocean ridge to the south of the continent provides a less frequent, but very useful, set of shallow events. There are only a limited number of permanent seismic stations with high-fidelity broadband recording in the region and these have been supplemented by extensive deployments of portable broadband instruments. The SKIPPY experiment (van der Hilst et al. 1994) occupied 60 sites across the continent in a series of deployments from 1993 to 1996 and subsequent experiments have improved coverage in southeastern Australia and different parts of Western Australia (Kennett 2003). Fundamental and higher mode Rayleigh waves recorded on vertical component seismograms have been exploited in a number of tomographic studies of the shear-wave speed distribution beneath the Australian region. Zielhuis and van der Hilst (1996) analysed the data from the SKIPPY stations in eastern Australia using the Partitioned Waveform Inversion technique of Nolet (1990). They produced the first clear evidence of a substantial difference in the structure in the mantle beneath the Precambrian regions of the centre and west of the continent and beneath the Phanerozoic belt in the east. The path coverage was extended to nearly the whole continent by van der Hilst et al. (1998). This work revealed a consistent pattern of a mantle lithosphere extending to around 250 km depth beneath the shield regions, contrasted with a thinner lithosphere (about 100 km thick) below the eastern part of Australia, underlain by a significant low velocity zone for S waves. Subsequently the number of available data have been augmented in various ways. Simons et al. (1999) have included data from the Australian Geological Survey Organisation stations and undertaken a study of regionalisation based on the geologic provinces in the continent. Debayle and Kennett (2000a) have exploited almost 2200 paths using an automated waveform analysis procedure (Debayle 1999) based on the work of Cara and Lévêque Geol. Soc. Australia Spec. Publ. 22, and Geol. Soc. America Spec. Pap. 372 (2003), 25–40