Structure of the Alpine Fault, South Westland, New Zealand

Eales, B. C. (Bradley Charles)

Seismic events were recorded on the Alpine Fault by three seismometers located near the area of Haast in the period August 1999 to December 1999. One seismometer was located directly over the Alpine Fault trace at Red Hut, while the second was located to the west of the Alpine Fault at Zillion Hill and the third located to the east near the Boil Pool on the Okuru River. Hypocenter and frequency analysis of the data identify ten events which possibly contain fault-trapped waves. Fault-trapped waves occur as a result of coherent multiple reflections occurring within a low velocity zone, bounded by two higher velocity zones. Fault-trapped wave character is a function of the physical properties of the fault zone including fault zone width, fault zone velocity and velocity contrast. From the identified ten events, an event on the 25th August 1999, located 5 ±4km deep and 38km southwest along the Alpine Fault from Red Hut, was identified as having fault-trapped waves. The event on the 25th of August 1999 is modelled using two separate models: Ben-Zion and Aki (1990), Ben-Zion (1998) and Michael (2001), which identify a fault zone (from the event 38km southeast along the Alpine Fault to Red Hut) approximately 0.12 - 0.16km wide, with an S-wave velocity of 1.8 - 1.9kms-1 . S-wave velocity outside the fault zone ranges from 3.0 - 3.1kms-1 on the west, to 2.6 - 2.8kms-1 on the east. Q values in the fault zone range from 35 - 75, while outside the fault, Q ranges from 400 - 900. Velocity reduction therefore, outside to inside the fault zone is :::::32%, with an error of -±16%. Fracture density analysis northeast of the study area at Gaunt and Hare Mare Creeks identifies fracturing east of the fault decreasing away from the cataclasite zone and dropping off substantially at -40m away from the identified trace. Discrepancy between model and measurement suggests one of five possibilities: 1. Fracturing is not the cause of velocity reduction in the study area. 2. Fracturing is the cause of velocity reduction and the damage zone must be of greater width in the study area. 3. Fracturing is the cause of the velocity reduction, but what is seen at the surface is not representative of the fracturing at depth. 4. Fracturing is the cause of velocity reduction, but occurs to a greater extent on the western (unmeasured) side of the Alpine Fault. 5. Fracture density measurements at both Gaunt and Hare Mare Creeks are not accurate and/or representative of the fracturing occurring there. Calculation of the effects of clay content, effective pressure and fracture density in the fault zone leads to a preferred model of fluid filled fracturing, with clay content possibly boosting localised areas of fluid pressure to decrease effective pressure and hence decrease seismic velocity. This would indicate that a higher fracture density is the main cause of-velocity reduction within the fault zone, but whether fracture width increases with depth, is of greater width in the study area or is of greater extent on the western side of the fault zone still remains unknown.

xxii, 249 p. : ill. (some col.), maps ; 30 cm.

2001Eales

POLYGON ((167.728043567952113 -44.400251810132666,169.019249513108747 -43.666617701691308,169.356135896369949 -43.927985998508639,168.061889929603211 -44.724848135107209,167.728043567952113 -44.400251810132666))