A geodetic investigation of slow slip in the Hikurangi Subduction Zone beneath Raukumara Peninsula, New Zealand

Abstract

Fault slip episodes, occurring over much longer time periods (days-months) than earthquakes, have been recorded using Global Positioning System (GPS) instruments at subduction margins around the Pacific Rim. These so-called slow slip events may make a significant contribution to moment release in subduction zones, so estimating their size and frequency is a key task in characterizing seismic hazard at subduction zones.

In October 2002, a slow slip event of 20-30 mm surface displacement was observed on two continuous GPS instruments near Gisborne on the Raukumara Peninsula, North Island, New Zealand. This event was the first of its kind to have been documented with continuous GPS in New Zealand. Scientists from the Institute of Geological and Nuclear Sciences (GNS) interpreted the motion to have been caused by slip along the boundary between the Pacific and Australian tectonic plates under the seafloor beneath Poverty Bay.

As continuous GPS coverage of the Gisborne region was sparse in 2002, it has not been possible to quantify the extent of the October 2002 slow slip event using continuous GPS data alone. In this project, we have used regional campaign GPS data (collected during day-long deployments at 2-3 yr intervals over the past ten years) in combination with the available continuous GPS data to study the October 2002 slow slip event and to find out whether other such events may have occurred in the past ten years.

Due to infrequent sampling in the campaign GPS data we cannot constrain the spatial extent and amount of slip during the October 2002 event any further than with the continuous GPS data. However, we have successfully used campaign and continuous GPS time series from neighbouring .stations to make recurrence interval estimates for events of similar surface displacement to the 2002 event in the Gisborne region. Our calculations show that such events recur at 2-4 year intervals, which is in agreement with the occurrence of the November 2004 slow slip event in the same region as the 2002 one. Our studies using regional data sets also indicate that the effects of the Gisborne 2002 slow slip event were localised.

We have made a preliminary study into the possible association of seismic tremor with the slow slip events recorded on the Raukumara Peninsula, such as has been found in North America and Japan. We find evidence for tremor recorded on multiple stations across the Raukumara Peninsula broadband seismic network during both the 2004 Gisborne slow slip event, but we have not rigorously examined the spatial and temporal relationship of this tremor to the slow slip event.

---

Technical Abstract

In October 2002, a slow slip event of 20-30 mm surface displacement was observed on two continuous Global Positioning (GPS) instruments near Gisborne on the Raukumara Peninsula, North Island, New Zealand. This event was the first of its kind to be documented with continuous GPS in New Zealand.

We have processed and analysed nearly ten years of regional campaign GPS records (1995-2004), which we use, in conjunction with recent continuous GPS and broadband seismic data, to study the spatial extent of the Gisborne 2002 slow slip event and its effect on regional deformation. We find that infrequent sampling in the campaign GPS data set has aliased the Gisborne 2002 slow slip event and  we cannot constrain the spatial extent and moment release of this event any further than with the continuous GPS data. We successfully use campaign and continuous GPS time series from neighbouring stations to make estimates of the recurrence interval for events of similar surface displacement to the 2002 slow slip occurring in the Gisborne region. Our calculations reveal that such events recur at 2-4 year  intervals, which is in agreement with the recent November 2004 slow slip event in  the Gisborne region. Campaign GPS time series and regional deformation fields  show evidence for more than one slow slip event occurring in the Gisborne area prior to 2001. Forward modelling shows that the surface displacements recorded during the Gisborne 2002 slow slip event are the result of 18 cm of slip on a plane, ca. 60 km x 25 km, on the subduction interface offshore of Gisborne, though the fault length is poorly constrained. The Gisborne 2002 slow slip event is shown to have had no significant effect on inter-seismic coupling and slip deficit on the Hikurangi subduction margin, or on regional deformation patterns over the Raukumara Peninsula. Our model of slip suggests that slow slip beneath the Raukumara Peninsula currently occurs in episodic events involving local slip at the base of the seismogenic zone on the subduction interface offshore of Gisborne.

Our studies using regional data sets also indicate that effects of the Gisborne 2002 slow slip event were localised. We have made a preliminary study into the possible association of seismic tremor with the slow slip events recorded on the Raukumara Peninsula. We find
evidence for tremor recorded on multiple stations across the Raukumara Peninsula broadband seismic network during both the 2004 Gisborne slow slip event, but we have not rigorously examined the spatial and temporal relationship of this tremor to the slow slip event.