Abstract
Earthquake hazards are significant in New Zealand. Many active faults are known to be close to urban areas, and damaging earthquakes have been known to occur on previously unknown faults (such as the Mw~7.1 Darfield earthquake of September 2010). The potential for a large earthquake on the Akatore Fault represents a significant seismic hazard to Dunedin City. The Akatore Fault may be within 15~km of the city, and paleoseismological studies suggest the fault may be in a period of heightened activity. Determining the geometry of the Akatore Fault is key to seismic hazard analysis in South Dunedin, as is ground motion prediction. As Dunedin has not experienced a large earthquake since it was settled in 1840, the best way to quantify ground motion is through earthquake ground motion simulation. This study aims to investigate the geophysical properties of rocks and sediments pertinent to 3D ground motion simulations for an Akatore Fault earthquake event in Dunedin City and contribute to determining the geometry of the Akatore Fault.
New data acquired as a part of this study include three strategically placed seismic lines. A seismic line was collected across the Kaikorai Stream estuary, intersecting a proposed onshore location of the Akatore Fault. Two seismic lines were collected in South Dunedin. The data were processed to produce near-surface velocity models, stacked reflection profiles and stacking velocity models. The stacked profiles were interpreted in terms of mapped stratigraphic and basement geological units. The stacked reflection profile at Kaikorai Estuary contained no evidence of the Akatore Fault extending onshore. This sets an upper limit on the fault's length for future hazard analysis.
Active source and passive surface wave data were collected at four sites across South Dunedin and analysed to estimate the near-surface S-wave velocity profile. Surface wave data were extracted from existing seismic lines, analysed, and compared to the dedicated active source seismic surveys to assess the potential for inversion. The data were found to be too low quality to invert alone but could be used in conjunction with MASW data collected at the same site.
Gravity data collected across South Dunedin by Lutter (2018) was reprocessed to produce a fence diagram of 2D profiles following significant roads. These were used to create a surface between the low-density late Quaternary sediments and the higher density Dunedin sequence sediments.
Ambient seismic vibrations recorded across South Dunedin were used to produce Horizontal to Vertical Spectral Ratio (HVSR) curves which were used to identify site period. The HVSR peaks were analysed in conjunction with the gravity data and the S-wave information to estimate the thickness of the soft sediments overlying the rocks of the Dunedin sequence below.
The data collected in this study, existing mapped contacts, and borehole data were used to build a 3D geological model of the path between the Akatore Fault and Dunedin City. The velocities of the units were informed, where possible, by the P-wave stacking velocities used in seismic lines collected. S-wave velocities were constrained in the near surface with surface wave methods. For deeper units s-wave velocities were calculated using empirical equations.
This study has collected and prepared the data and models required to characterise the path and site used in 3D ground motion simulations. This is a significant step forward concerning earthquake hazard analysis in Dunedin.