Paleoseismology and Seismic Hazard of the Hikurangi Megathrust, New Zealand

Abstract eng:
A major earthquake on the Hikurangi megathrust at the boundary of the Pacific and Australian plates is one of the more significant hazards facing New Zealand, but it is also one of the least well constrained. There have been no great (M >8) earthquakes on the subduction interface in the short period (~175 years) for which we have an historical record of seismicity. New Zealand’s National Seismic Hazard Model (NSHM) combines the plate motion rate budget, historical seismicity, geodetic observations of contemporary interseismic coupling, and slow slip occurrence to define seven subduction interface sources for the Hikurangi megathrust. These include wide and narrow subduction interface sources for each of the southern, central, and northern parts of the megathrust rupturing separately in M w 8.1-8.3 earthquakes every 550-1400 years and a single source spanning the length of the three previous parts rupturing in an M w 9.0 earthquake approximately every 7050 years. To calibrate models such as the NSHM we aim to determine the timing and size of pre-historic large ruptures of the Hikurangi megathrust by radiocarbon dating sedimentary evidence for coseismic vertical deformation, tsunami inundation and offshore turbidite deposition. A major challenge of this work is isolating evidence of megathrust earthquakes from earthquakes on the numerous active upper plate faults. We rely on comparison with timing of known paleoearthquakes on upper plate faults, expected patterns of subsidence and uplift from dislocation modelling, and the likely distribution of tsunami impact from tsunami modelling of different earthquake sources. Here we present several examples of paleoseismological studies along the Hikurangi Margin to illustrate our current state of knowledge about past earthquake occurrence with respect to the NSHM. Comparison of the paleoseismic record with the NSHM shows that, in terms of recurrence interval estimates for great (magnitude 8 or greater) and giant (magnitude 9 or greater) earthquakes, the two datasets are in reasonable agreement. At the southern Hikurangi margin there is only a short paleoseismic record (last thousand years) but the recurrence intervals are not inconsistent with the 340-year approximation given in the NSHM. In the central Hikurangi margin there is a longer paleoseismic record (7500 years) and, although the NSHM estimate of 590 years is shorter than the geologically derived estimate of 810 years, there are indications that earthquakes are missing from the geological record there. The northern Hikurangi margin shows the greatest discrepancy with 470 years used in the NSHM and 800 years derived from the geological record. Whole margin rupture in a magnitude 9 earthquake is not inconsistent with the geological data but has yet to be demonstrated unequivocally.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket