Computer simulations using GeoNet seismological data allow us to watch the progress of the rupture as it punched its way up the east coast of the South Island over an approximately two minute period early in the morning of November 14th.

Because Geonet’s seismic network had numerous seismic stations situated very close to the rupture (just like its Canterbury Seismic Network during the 2010-2011 earthquake sequence), the data from these sensors allow us to essentially travel back in time and watch the event unfold. Computer simulations of the seismological data (using two very different methods—see figures below) suggest that the rupture started in the south, on a fault at approximately 15 km depth near Culverden. The initial rupture near Culverden subsequently triggered a domino effect, as the earthquake rupture jumped from fault to fault, essentially “unzipping” along a 150 km length of the northeast coast of the South Island. The data suggests that many faults were involved in the rupture.

The earthquake rupture picked up a head of steam as it passed Kaikoura, with the largest fault movements occurring between Kekerengu and Cape Campbell. This pattern is confirmed by GPS and satellite radar observations. Near the end of the earthquake’s runaway journey, it jumped onto a series of faults offshore. NIWA has found clear evidence of large earthquake displacement on one of these--the Needles fault (NIWA story). The earthquake rupture terminated abruptly just before reaching Cook Strait. There is no evidence at this time that the Hikurangi subduction zone was significantly involved in the rupture.

The Kaikoura M7.8 earthquake is arguably the most complex earthquake rupture ever to be observed in this level of detail with modern instrumentation (seismic sensors, GPS instruments, and satellite radar data). Multiple datasets (seismology, geodesy, and surface observations of faulting, uplift, and tsunami) are rapidly converging to give us a single, consistent view of exactly what happened in the earthquake. Preliminary models, will of course be refined, as new information continues to come in.

Science contacts: Caroline Holden and Bill Fry, GNS Science