Nov 16 2004 - Scientists Detect Slow Quake at Gisborne

GPS station location map.

The continuous GPS time series plots on the GeoNet web site show that a new episode of rapid deformation is taking place near Gisborne, on the Raukumara Peninsula in north-eastern New Zealand.

Starting on about October 31, the new period of rapid motion of the GISB continuous GPS station had reached a magnitude of nearly 20 mm by November 12, in a direction slightly south of east.

This motion of about 2 mm per day is called "rapid" because it is more than ten times faster than the normal geological motions of plate tectonics, which are about 50 mm per year (or 0.1 – 0.2 mm per day).

The ongoing motion is similar to an earlier event in October 2002 that GNS scientists have attributed to an episode of slow slip on the tectonic plate interface about 20 km below Poverty Bay, where the Pacific Plate subducts beneath New Zealand. The slip event is called "slow" because it describes slip between the rocks on two sides of a geological fault that takes place very slowly (up to tens of millimetres per day), compared to the slip rate during earthquakes (several kilometres per second).

Slow slip events are therefore sometimes called "slow earthquakes" because they relieve some of the accumulated stress between the rocks on either side of a geological fault, but they do it slowly (over a period of days or weeks) rather than rapidly (over seconds) as in a normal earthquake. The strong shaking and damage that occurs in a normal earthquake is due to the fact that movement between the rock on the two sides of the fault occurs at the huge rate of several kilometres per second.

Normally the interface between the subducting Pacific Plate and the overlying New Zealand crust is locked, which causes the southern Raukumara Peninsula to be pushed slowly westwards (at a few millimetres per year) due to the westward motion of the Pacific Plate. But during slow slip events the locking is temporarily released, allowing the overlying crust to move back eastwards at rates of a few millimetres per day.

A joint research project in progress between Victoria University of Wellington (VUW) and GNS has shown that the 20 - 25 mm eastward surface motion produced by the October 2002 event required a patch of the plate interface of about 25 km (east-west) by 60 km (north-south) extent to have slipped by about 180 mm over a period of about 10 days. The new November 2004 event looks to have a very similar character to the October 2002 event.

Slow slip events have been observed in recent years in other subduction zones, particularly in the Cascadia subduction zone (western Canada and northwestern U.S.) and in several subduction zones in Japan.

The Gisborne slow slip events appear to be occurring at significantly shallower depth than those in the Canadian and Japanese examples. However, in all these cases the slow slip seems to happening below that part of the plate interface where major subduction zone earthquakes occur. It’s just that the depth of this main seismogenic zone is much shallower offshore of Raukumara than it is in Japan or Cascadia.

A curious signature on seismic traces has been observed in both Japan and Cascadia in association with slow slip events. This signature is called subduction zone tremor, because it has some visual similarities to the volcanic tremor signal that is often detected by seismometers near volcanoes when magma (molten rock) is in motion within the volcano. The origin of subduction zone tremor is very unclear, but it may be related to the motion of fluids (perhaps water) in association with the slow slip episodes. The joint VUW/GNS research project mentioned above is presently examining New Zealand continuous seismic records for evidence of similar signals here.

In October 2002 only the GISB continuous GPS station plus a nearby privately-operated station in Gisborne city were in operation on the Raukumara Peninsula. Since then, three other stations have been installed, one of which (HIKB) is available on the GeoNet web site. The current event is not visible on any of the new stations. This shows that the present slip event is occurring within a fairly confined region, probably similar to the October 2002 event.

The absence of signal at the other sites also shows that a significantly closer spacing of continuous GPS stations is required for this type of slow slip event to be detected on several stations. Only by observing slow slip events on several stations can their locations and other properties be properly assessed, which will lead to a better understanding of the events and their implications. Similarly, the study of subduction zone tremor, if it is detected in New Zealand, should assist our understanding of slow slip events. Several more continuous GPS stations are planned to be installed in the Raukumara region over the next few years as part of the GeoNet project.

The presence of slow slip events will be important in future in reassessing the seismic hazard in particular regions, since slow slip events may relieve some of the stress that would otherwise be released in earthquakes.

It has also been suggested by a number of scientists that slow slip events on the deeper subduction interface may trigger the occurrence of large earthquakes within the main seismogenic zone immediately above. There is no real evidence in favour of this suggestion at present, but it is something that scientists are watching carefully.