A rare great earthquake on an oceanic fossil fracture zone

Broad-band body and mantle wave data are used to study the 2004 December 23, Tasman Sea earthquake. In common with other strike-slip earthquakes studied in the same fashion, the mantle wave data indicates that there are two pure-double couple constrained solutions, along with a range of mechanisms between them, that fit the data almost equally well. Aftershocks relocated for this study indicate that the rupture occurred on a fracture zone which bends sharply in the epicentral region. Horizontally polarised S body waves and P body waves are used to determine the rupture parameters. A model with two faults best fits the data. The northern fault plane, with strike 160°, dip 86° and rake 5°, is compatible with the first motion solution found in this study and has a strike consistent with the fracture zone north of the bend. The southern fault plane, with strike 178°, dip 54° and rake 65°, has a strike consistent with the portion of the fracture zone to the south of the bend and has a dip which can be explained by the style of deformation that the region is undergoing. The centroid moment tensor solution of the broad-band model is calculated and found to be consistent with the region of low misfit in mantle wave solution space. The broad-band solution has a moment of 1.53 × 1021 N m (Mw 8.1), again, consistent with the mantle wave data. Slip propagated bilaterally with an approximate rupture velocity of ∼3 km s−1∼80 per cent of local shear wave speed. The rupture front is less well resolved to the south of the epicentre than to the north. The majority (∼75 per cent) of moment originated from slip on the northern fault. The broad-band data requires significant slip below the oceanic Moho with as much as 70 per cent of moment due to slip in the brittle uppermost mantle in the preferred model.