Earthquake hazard in Dunedin: paleoseismology of the Titri Fault

Abstract

The Titri Fault is a major southeast-dipping reverse fault in coastal Otago, southwest of Dunedin. The fault is up to 90 km long and has uplifted the coastal hills between Dunedin and Balclutha. Activity rates on the Titri Fault were previously poorly constrained, and the fault has not previously been included in the New Zealand National Seismic Hazard Model (NSHM). Two suspected fault scarps were identified on alluvial fans along the Titri Fault and were investigated via an Earthquake Commission Biennial Contestable Grants Programme project. The aim of the project was to trench the scarps and obtain quantitative data on the nature and history of the most recent rupture(s) of the Titri Fault, in order to improve the understanding of Dunedin’s seismic hazard. A secondary aim was to investigate whether ruptures of the Titri Fault have been clustered in time.
Three trenches were sited across the suspected surface fault scarps, and all exposed reverse faulted and folded Quaternary-age alluvial fan deposits and colluvial wedges. Collection of samples for chronological analysis comprised 21 samples for optically stimulated luminescence (OSL) dating and five samples for radiocarbon dating; the budget allowed the dating of seven OSL samples and one radiocarbon sample.

Stratigraphic evidence for past surface ruptures of the Titri Fault was found in all three trenches. Dating results show that the two most recent ruptures could have been the same in all three trenches. The most recent event was between 18.3 and 31.3 ka (1 ka = 1000 years) and the penultimate event was between 27.5 and 37.9 ka. Fault parameters quantified as a result of this study include a single-event net displacement in the range of 2.4 to 2.7 m, average fault slip rate in the range of 0.1 to 0.2 mm/yr, and a poorly constrained recurrence interval range of ~7000 to 19,000 years, with a preferred value towards the longer end of that range. The results provide hints, but not definitive evidence, of episodic activity (clustered ruptures) on the Titri Fault.

The results demonstrate that the Titri Fault is an active fault with relatively low activity and should be included in future versions of the NSHM. In seismic hazard modelling, the Titri Fault is represented as three ~30 km long segments with rupture of individual segments modelled as able to produce earthquakes of Mw ~7.0 and ruptures of two or all three segments generating earthquakes up to Mw ~7.7.