An earthquake of magnitude 6.0-6.5 in the Sydney region of Australia is viewed by the global insurance community as one of the top 40 risks it faces worldwide from natural disasters . The high ranking of this perceived risk is due to the high population density, standards of construction and the level of insurance exposure in Sydney. Consequently, earthquake hazard and risk in Sydney is an important issue, and one that requires a focused and detailed study in order for the implications of such an earthquake to be fully understood.

The presence of regolith (soils, sediments and weathered rock) can dramatically affect the level of ground shaking experienced during an earthquake. The relatively soft materials that constitute regolith tend to have low seismic velocities that amplify ground shaking during an earthquake, increasing the potential for damage to buildings and other infrastructure in the affected area. Therefore, models of the response of regolith to an earthquake (referred to as site response) form an integral part of any earthquake risk assessment.

This report documents a preliminary study of potential ground motion amplification due to the regolith in the Botany area of Sydney, Australia. Botany was chosen due to the presence of a significant thickness of regolith and a high value and concentration of critical infrastructure. This report is intended to highlight the potential for significant levels of amplification within the study area, and draw attention to the need for more work on assessing the actual earthquake risk faced by the Sydney region.

In order to determine the amount of ground motion amplification that could be seen in the Botany area, the regolith was classified into a series of four site classes. These regolith site classes are differentiated in terms of geotechnical properties that control ground shaking potential. This classification was based upon published and unpublished geotechnical data as well as seismic velocities obtained by Geoscience Australia. Once geotechnical models were defined for each regolith site class, amplification factors were calculated using a vertically propagating shear wave model. This model accounts for the softening and critical damping of the regolith column during large earthquakes. The results demonstrate that there is significant potential for amplification of ground shaking within the study area. For example, the site class that covers the vast majority of the study area has a maximum amplification factor greater than 3.0 at a fundamental site period of approximately 0.5 s. This period of motion would be expected to strongly affect the structures in the study area.

The modelled amplification factors suggest that, should an earthquake impact the area, the potential for high levels of ground shaking would be dramatically increased due to the properties of the local regolith. An earthquake similar to the event experienced in Newcastle in 1989 was simulated, in order to demonstrate the potential amplification effect of the regolith during an earthquake. Whilst this simulation is in no way a full probabilistic risk analysis of the area, it does demonstrate that the amplification of ground shaking could cause response spectral accelerations in excess of 1.0 g, at periods of vibration that would be expected to cause damage to structures in the area.

It is important to emphasise that this work is intended to provide a point of focus to initiate discussion rather than be a definitive seismic hazard assessment product. The results have been derived with limited geotechnical data, and without a detailed analysis of the uncertainties present within either the data or the modelling process. Nevertheless, this work does provide a starting point for recognising and addressing the potential risk that earthquakes pose to the study area.