A complex, young subduction zone imaged by three-dimensional seismic velocity, Fiordland, New Zealand

Abstract

Summary The Fiordland subduction zone, where subduction developed in the late Miocene, has been imaged with P and S–P arrival‐time data from 311 earthquakes in a simultaneous inversion for hypocentres and 3‐D VP and VP/VS models. The three‐month microearthquake survey, recorded with 24 portable seismographs, provides excellent coverage, and, since earthquakes to depths of 130 km are included, parts of the model are well‐resolved to depths of 100 km. The crustal features are generally consistent with geology. The low velocity in the upper 10 km is associated with the Te Anau and Waiau basins. The Western Fiordland Orthogneiss is associated with a prominent feature from near‐surface to over 40 km depth, which includes the residue from the basaltic source rocks. It is defined by high VP (7.4 km s−1 at 15 km depth) and slightly low VP/VS, and has distinct boundaries on its southern and eastern margins. Adjacent to the deepest earthquakes, there is high‐velocity Pacific mantle below 80 km depth, inferred to be the mantle expression of ongoing shortening since the early Miocene. As the subducting slab moves down and northeast, it is hindered by the high‐velocity body and bends to near‐vertical. Bending is accommodated by distributed fracturing evidenced by high VP/VS and persistent deep earthquake activity. Buckling of the subducted plate pushes up the Western Fiordland Orthogneiss. In the transition to the Alpine fault in northern Fiordland, a prominent low‐velocity crustal root is consistent with ductile thickening in combination with downwarp of the subducted plate.

Intermediate-Depth Earthquakes in a Region of Continental Convergence: South Island, New Zealand

“…The inversion data consisted of teleseismic travel times measured from the NZNSN waveforms and from waveforms recorded during the 1995-1996 Southern Alps Passive Seismic Experiment (Anderson et al, 1997). Crustal heterogeneity was accounted for by ray tracing through an independently obtained threedimensional crustal velocity and Moho depth model (Eberhart-Phillips and Reyners, 1997Reyners, , 2001Eberhart-Phillips and Bannister, 2002). The images show a near-vertical, highvelocity (2%-4%) structure in the uppermost mantle that underlies thickened crust along the north-northeast-southsouthwest axis of the Southern Alps.…”

“…Intermediate-depth seismicity in central South Island is clustered near the southern end of the Southern Alps (Fig. 2), where significantly more crustal thickening has occurred than in the northern end (Eberhart-Phillips and Reyners, 1997Reyners, , 2001Eberhart-Phillips and Bannister, 2002). …”

Uplift in the Fiordland Region, New Zealand: Implications for Incipient Subduction

“…Along strike, the positive gravity anomaly is centered to the south of the deepest seismicity (42,43). However, any association between seismicity and gravity must be interpreted with caution as there is some debate as to the origin of the anomaly:…”

“…The slab beneath Fiordland has been described as folded downward (42,45) or torn (32,46,47), which may influence patterns of uplift. For example, increased resistance to subduction of the Australian plate due to impingement into a thick, strong continental lithosphere below Otago, east of Fiordland may play a role in driving uplift (42). A seismic P-wave tomographic inversion beneath Fiordland indicates the Fiordland slab has encountered increased resistance to subduction and is being forced to descend vertically by a strong and thick lithospheric root to the east.…”

Geophysical structure of the Southern Alps Orogen, South Island, New Zealand

“…They document the change towards the southwest from lithospheric thickening under the Southern Alps to near-vertical subduction of oceanic Australian plate under Fiordland, and the transition from this near-vertical subduction to inferred lower-angle subduction of the Australian plate at the Puysegur trench south of South Island (Eberhart-Phillips and Reyners, 2001;Reyners et al, 2002). The data were used to image a high-velocity body in the Pacific mantle which they infer has caused the subducting Australian oceanic slab to bend progressively steeper to the northeast (Fig.…”