Three-dimensional seismic structure beneath Australia: the SKIPPY Project 1993-1998

B.L.N. Kennett
Seismology Group, Research School of Earth Sciences

Over the five year period from May 1993 to April 1998 the Seismology Group at the Research School of Earth Sciences was engaged on a major project in observational seismology with the object of mapping the three-dimensional structure beneath the Australian continent.

The project exploits the information on seismograms recorded at sites in Australia from earthquakes lying in the major earthquake belt to the north and east of the continent. Suitable events are quite frequent in the zone of seismicity which extends from Indonesia, through New Guinea to Fiji and then down through Tonga and New Zealand. Each of these earthquakes generate a number of different types of waves which travel through the Earth to the recording stations and acquire during their passage information about the seismic velocities along their path from source to receiver.

The techniques used to extract information from the seismograms on threedimensional structure are similar in principle to medical tomography. By using information on the characteristics of waves which have travelled in many different directions through the region beneath Australia we can build up images of the seismic structure.

Although there is a good supply of seismic sources, there are only a few high quality seismic observatories on the Australian continent and in order to be achieve detailed coverage we have built up a continent wide array using portable recorders. A total of 65 recording sites have been occupied across the whole country, at each one we have installed a seismometer with a broad frequency response and a data-logger with 24 bit digital recording onto digital audio tape, as a result we have a high-fidelity recording of the ground motion at the site.
These portable units record for 9 weeks continuously onto a 90 m tape with a total of 2 Gbytes of data per tape. From this full record we extract in Canberra the portions corresponding to the seismic events of interest, and these segments are then used for further analysis.

It was not possible to cover the whole continent at one time with available hardware, and so we have proceeded in stages. The first array of 8 instruments was deployed in Queensland in May 1993, and in subsequent stages up to 12 units have been used at a time. The movement of the array of instruments from region to region has led to the name Skippy for the whole project. Each deployment normally lasts 5 months which is sufficient to achieve coverage of regional seismicity, and the active teleseismic belt extending to the north through the Philippines to Japan and the Aleutians. After three and a half years of intensive field work, we have achieved the goal of covering the entire Australian continent (see figure 1) and we have been able to get quite close to the original design goal of a 400 km spacing between instruments. The arrays of instrument have covered many of the desert areas of Australia and logistics have at times been complex.

Figure 1: The portable seismic recorder deployments from 1993 -1996 in the SKIPPY project.

A major focus of the data analysis has been on the construction of detailed images of the three-dimensional variation in seismic wave speed beneath the Australasian region. We have so far incorporated the data from all the recording stations in central and eastern Australia into our tomographic study of the seismic structure. This has enabled us to produce images of structure below 60 km depth with a resolution of about 200 km in the horizontal direction, which is a factor of 5 better than previous studies using permanent stations. As a result we see many features that are unresolvable with limited coverage. We are continuing to develop the models with the inclusion of data from the remaining stations.

Figure 2: The paths used so far in the construction of 3-D models for the structure beneath Australia. The waves penetrate deep into the earth so that we are able to build images to 400 km deep.

The main geological structure of the Australian continent is quite well known and from studies of rocks at the surface the basement structure can be divided by age into three divisions which major stages in the evolution of the crust: the Archaean (older than 2500 Ma), the Proterozoic (2500-550 Ma) and the Phanerozoic (younger than 550 Ma). The major Archaean outcrop lies in the west and the Proterozoic rocks are found in the east of the continent (figure 3). However, very little is known about the mantle underlying the crust beneath Australia.
Some of the questions that the Skippy project was set up to address were: what is the thickness of the zone that is associated with the Australian continent in its plate motion to the north at 6 cm/yr? and, whether we can detect the boundaries of between the major provinces? if so, what is the character and depth extent of such features?.

Figure 3: Geological key map of the Australian Continent (also showing the location of the sections in figure 5.

With the results already obtained we have already been able to add a new dimension to our knowledge of the structures associated with Australia down to 400 km depth. We have also significantly enhanced the available information on structure in the crust across the continent by the analysis of distant earthquakes at each of the portable recording sites. In this work we use the information on reverberations in the crust contained in the seismograms to develop models for the seismic velocities and thichness of the Earth's crust.

The major feature that is revealed from the seismic tomography studies is a strong contrast in the deep structure between central Australia and the eastern seaboard (figure 4). The three-dimensional images of seismic wavespeed show that beneath the region of ancient rocks in the centre and west a zone of high seismic velocities extends to below 200 km in depth, whilst in the east the seismic velocities are much lower and there is a pronounced zone of reduced velocities centered at 140 km depth. The pattern of lowered velocities fits in well with the presence of recent volcanism and regions of high heat flow, and so it is likely that the cause of the reduction of the seismic wave speeds is hotter material.

Figure 4: A slice through the three-dimensional model of shear wave speed at 140 km showing the strong contrasts between central and eastern Australia.

The nature and depth extent of the contrasts in the structure can also be seen clearly in a the set of vertical cross-sections through the new threedimensional model at different latitudes (figure 5).

Figure 5: Vertical slices through the three-dimensional model of shear wave speed showing the contrasts in seismic velocity at depth.

These new images of the structure at depth beneath Australia should improve understanding of the way in which the continent has evolved. The strong contrast in seismic structure in the mantle does not have a simple relationship to the edge of the ancient Precambrian rocks as seen at the surface even though there is a general association with lower seismic velocities at depth beneath the younger rocks of eastern Australia. Within the zone of higher velocities we see clear indications of substructure which appears to be related to the way in which the different parts of the ancient core were assembled in the distant past.

The tomographic images supplement more detailed studies of particular regions at the surface and will help to unravel the way in which the processes in the Earth's interior interact with the surface rocks in the formation of mineral deposits and other natural resources. For example, the high seismic velocities extending to 200 km and below in central and western Australia represent favourable conditions for the formation of diamonds but in order for economic deposits to be formed the diamonds still need to be brounght to the surface in an eruption.

Interpretation of the data from the SKIPPY project continues, and is being supplemented by the inclusion of additional data from permanent stations and further portable deployments covering portions of the Australian Continent. For eCxample, field work in 1997-1998 was concentrated in the Kimberley region in north-western Australia.


Zielhuis, A. & van der Hilst, R.D. (1996)
Upper-mantle shear velocity beneath eastern Australia from inversion of waveforms from SKIPPY portable arrays.
Geophys. J. Int, 127, 1-16.

van der Hilst, R.D., Kennett, B.L.N. & Shibutani, T. (1998)
Upper mantle structure beneath Australia from portable array deployments, 39-58.
in Structure and Evolution of the Australian Continent, ed. J. Braun et al, AGU Geodynamics Series, 26

Clitheroe G. & van der Hilst, R.D. (1998).
Complex anistotropy in the Australian lithosphere from shear-wave splitting in broad-band SKS records, 73-78.
in Structure and Evolution of the Australian Continent, ed. J. Braun et al, AGU Geodynamics Series, 26

Kennett, B.L.N. (1997)
The mantle beneath Australia. AGSO Journal of Australian Geology & Geophysics, 17(1), 49-54.

Skippy Project: RSES Annual Report, 1996

Skippy Project: RSES Annual Report, 1997

For further information please contact: Professor Brian Kennett
Information on the Skippy stations response characteristics could be found here.

The distribution of the earthquakes used in Skippy project could be seen here.

The location of the Skippy stations during each stage (SK1-6) could be seen here.

A simpler version of the Skippy stations distribution (SK1-6) could be seen here.

You are

visitor of this page since 9 April 1997!
Return to Seismology Group

Last modified: 1998 April 9 by