Hikurangi Subduction modelling

Natural Hazard Research Platform funded project

Research Overview

Large mega-thrust earthquakes, such as the 2011 Tohoku-oki earthquake in Japan, occur at subduction zones. In the lower North Island of New Zealand, the absence of recorded past large mega-thrust earthquakes and a unique tectonic situation preclude the use of purely empirical approaches to estimate the resulting ground-motion levels. Understanding earthquake source parameters controlling the levels of ground motions from destructive mega-thrust earthquakes is critical for seismic hazard assessments.

We derive key parameters for the engineering specification of seismic ground motions for the lower North Island of New Zealand using advanced wavefield simulations of plausible mega-thrust earthquakes through:

  • developing an extensive set of potential rupture scenarios for magnitude 8+ earthquakes
  • computing ground motions using a hybrid method that incorporates both deterministic and stochastic approaches at low and high frequencies, respectively.
  • capturing a greater variability of strong ground motions across the lower North Island at both high and low frequencies. 
Figure 7: Fault (source) parameters for 2 specific Hikurangi scenarios

Fault (source) parameters for 2 specific Hikurangi scenarios

Research work as part of It’s Our Fault project:

It’s Our Fault funded project

The goal of the It’s Our Fault programme is to see Wellington positioned to become a more resilient city through a comprehensive study of the likelihood of large Wellington earthquakes, the effects of these earthquakes, and their impacts on humans and the built environment.

The It’s Our Fault programme comprises three main phases: Likelihood, Effects and Impacts.
Ground motion modelling of Hikurangi earthquake source seats under the effect phase.

Research outcomes from the project are:

  • Ground motion modelling of a large subduction interface earthquake in Wellington, New Zealand (Holden and Zhao, 2011, 2013)
  • A modified ground-motion prediction equation to accommodate simulated Hikurangi subduction interface motions for Wellington (McVerry and Holden, 2014)
  • Ground motion modelling of local site effects in the Wellington region (Kaiser et al., 2012; 2014)

1947 Tsunami earthquakes

Hikurangi margin tsunami earthquake generated by slow seismic rupture over a subducted seamount (Bell et al., 2014)

Figure 8: A. Yellow stars are the epicentres of the March and May 1947 earthquakes, with red and blue bars at the coastline showing eye-witness observations of tsunami wave heights for the events respectively. Locations of a subducted seamount and underthrust sediments in the region of the March 1947 event are shown as brown and red patches (Bell et al., 2009). Relief from ETOPO2 (Smith and Sandwell, 1997) and swath bathymetry where available. Magnetic anomalies above +150 nT are plotted as contours (Sutherland, 1996). B Interseismic coupling coefficients offshore Gisborne derived from interpretation of GPS velocities (Wallace et al., 2004, 2009) and estimates of interseismic coupling coefficient (i.e. slip fraction) assumed for the March 1947 earthquake source model. C Seismic reflection profile 05CM-04 acquired using a 12 km long streamer (Bell et al., 2009)

A. Yellow stars are the epicentres of the March and May 1947 earthquakes, with red and blue bars at the coastline showing eye-witness observations of tsunami wave heights for the events respectively. Locations of a subducted seamount and underthrust sediments in the region of the March 1947 event are shown as brown and red patches (Bell et al., 2009). Relief from ETOPO2 (Smith and Sandwell, 1997) and swath bathymetry where available. Magnetic anomalies above +150 nT are plotted as contours (Sutherland, 1996). B Interseismic coupling coefficients offshore Gisborne derived from interpretation of GPS velocities (Wallace et al., 2004, 2009) and estimates of interseismic coupling coefficient (i.e. slip fraction) assumed for the March 1947 earthquake source model. C Seismic reflection profile 05CM-04 acquired using a 12 km long streamer (Bell et al., 2009)