Kaiapoi Fault inactive for 10,000 years

Margie Waters has a series of questions on the faultline recently discovered by surveys in the ocean off the Kaiapoi shoreline:

A. How close to the sea floor is the displacement?

A new fault zone has recently been identified off Kaiapoi and has been referred to as the "Kaiapoi Fault Zone". The zone includes at least three discontinuous faults, each about 10 kilometres in length. Two have interpreted displacements estimated from seismic reflection profiles to 80 to 85 metres below the seabed.

B. Do you have an age for the most recent displacement?

The youngest measurable offset is Late Pleistocene in age and is tentatively correlated to lie in sediments above the Linwood Gravels in the Bexley borehole sequence, indicating activity since 120,000 years ago. There is no evidence of offset on a prominent post-last glacial erosion surface, indicating no fault activity on these structures over the past 10,000 years.

C. How long has it been active?

Most of the Late Pleistocene faults in Pegasus Bay are reactivated faults that are associated with much older Cretaceous-Paleogene faulting (35 to 90 million years old). These older extensional structures have been overprinted by recent compressional deformation associated with the development of the Pacific-Australian plate boundary over the past 20 million years. The most recent phase of activity started at least in the early Pleistocene (about 1 million years ago), based on deformation commencing near the top of the Kowai Formation (dated at between 1.63-0.66 million years old). The active faults in the Kaiapoi Fault Zone appear to be active during the last million years or so.

D. What is the rate of offset of the faultline in the last 10,000 to 20,000 years?

For all of the structures in Pegasus Bay, including the most active Pegasus Bay Fault, there is no evidence of faulting younger than 10,000 years due to the lack of deformation of the post-last glacial transgressive erosion surface. Calculated slip rates are based on vertical offsets of older Pleistocene units, suggesting that in the north of Pegasus Bay, slip rates are in the order of 0.05-0.28 millimetres a year, while in the south they are lower at 0.01-0.07mm a year. These estimated slip rates should be regarded as minimum estimates because the contribution of strike-slip (ie, lateral offset) cannot be accounted for from the available seismic reflection profiles. The rate of vertical displacement on sections of the newly recognised Kaiapoi Fault Zone is extremely low (about 0.01-0.03mm a year). - Dr Philip Barnes, principal scientist, Niwa (answers A toD).

E. Given the sequence of events since September 4, how does that affect the likelihood of more activity occurring on this faultline?

We cannot say what the change in likelihood of an earthquake on any particular fault is from the ongoing earthquake sequence. What we do know is that the probabilities for future earthquakes in the Canterbury region, including this fault, have increased compared with probabilities that existed before September 4. However, these probabilities generally decrease as the distance increases from where most of the aftershocks are occurring. - Dr Matt Gerstenberger, seismologist and hazard modeller, GNS Science.

After the September 4 earthquake, the steel casing of the well on my property (150mm diameter x 29m deep) appeared to rise about 200mm out of the ground, evidenced by the fact that the concrete collar was above ground level and the supply pipe sloped down from the wellhead to the pump house, whereas previously it had risen from the wellhead. Other well pipes in the vicinity also appeared to rise by different amounts above ground level. What would be the reason this has happened?

Is it more likely that the ground level over the whole area has subsided by differing amounts relative to the different location of these wells, although appearing to remain constant? Quite significant amounts of liquefaction occurred in the area. - Richard.

Your observation is very interesting - something that we have seen at a few localities - but haven't really thought hard enough about. My immediate reaction, and the general consensus among my colleagues, is that it is indeed the ground that has sunk.

There are large areas where the ground has changed its elevation, which are clearly identified in the brightly coloured satellite-based interferogram maps and by GPS surveying. Subsidence of 200mm or more was quite widespread to the north of the Greendale Fault after the magnitude-7.1 quake of September 4.

Accepting the "sinking" hypothesis implies the settling of the ground must be shallow, occurring in the top 29m between the borehole collar and the base of the hole (where it must be fixed).

The other "lifting" mechanism could only occur if the bore was capped and water pressures in the ground were sufficient to build up, overcome any cementation and friction to lift the entire weight of the casing out of the ground. Borehole pressures did indeed vary considerably, both up and down, locally rising by about 5m of head (=50 kPa) and greater right beside the Greendale Fault.

I attempted to do some quick calculations to see whether such a pressure rise could lift 29m of steel casing, but rapidly realised that they would be too oversimplified without considering, for example, the nature of the casing and screen, performance of the well and the aquifer, as well as the locality of the bore. It's not a simple problem.

So in summary: while the general consensus is that the near-surface gravels have settled and the ground has sunk around the bore, I would be interested to hear of any further observations or ideas that might suggest the contrary process has occurred. - Dr Simon Cox, geologist, GNS Science (s.cox@gns.cri.nz).

Why are the aftershocks on occasion so far away from the fault, and do you know what type of movement is taking place? A lot seem to be in clusters. On the surface, a fault appears narrow. Do you know how wide the faults are? - Colin.

The Greendale Fault that moved in the magnitude-7.1 main shock on September 4 is a rather narrow structure, as evidenced by the exposure of the fault at the surface. However, when such a fault moves, it changes the stress field in the surrounding rocks, increasing stress in some regions and decreasing it in others. As a result, some smaller faults in the surrounding region are brought closer to failure and produce aftershocks, while other faults are moved away from failure (ie, stabilised) and those regions don't produce aftershocks. These changes in stress can produce aftershocks up to a couple of fault lengths from the main-shock fault. The movement that takes place in these "off-fault" aftershocks is generally the same as that of the main shock, as the main driving force for these aftershocks is still the background tectonic stress caused by the collision of the Pacific and Australian plates in the South Island. Clustering of aftershocks in space repeats the stress-triggering process at a smaller scale. That is, the stress changes produced by fault movement in an aftershock can cause some other smaller nearby faults to fail, producing more aftershocks. - Dr Martin Reyners, seismologist, GNS Science.

* Email your questions to Vicki.Anderson@press.co.nz

The Press