Kinematic modelling utilising the method of Haines & Holt extended to the case of cubic Bessel interpolation on curvilinear grids, allows analysis of presentday horizontal motions occurring in the Hikurangi margin, North Island, New Zealand. The velocity field solutions are derived from first order geological data; that is, rates and orientation of extension in the Taupo Volcanic Zone and rates and orientation of motion on the North Island Dextral Fault Belt, against the background of the pattern of uplift and subsidence in the margin.A basic (preferred) velocity field is presented, with four other solutions with different input data, to explain what controls features in the main solution. Every one of the five solutions is the best fitting solution for the input data in each case. All velocity fields are shown relative to the western boundary of the model, which is considered fixed, as part of the assumed non-deforming Australian plate.The velocity field in the main solution includes a strong clockwise rotation of the Hikurangi margin east of the Taupo Volcanic Zone in the north. Farther south, shear across the North Island Dextral Fault Belt facilitates the southwestward motion of the eastern part of the margin. An important boundary condition for the deformation in the North Island appears to be the higher rate of dextral shear in the Marlborough region, which accommodates the relative motion of the Australian and Pacific plates immediately south of the Hikurangi margin.The extension in the Taupo Volcanic Zone, with rates onshore of 5-10 mm/yr north of where the Taupo Volcanic Zone terminates in the centre of the North Island, and the strike-slip component of shear on the North Island Dextral Fault Belt, of c. 20 mm/yr in the south and <5 mm/yr in the north, account for most of the margin-parallel plate motion in the Hikurangi margin. No other major geological strains are required to be occurring in the North Island out of compatibility with these strains.

The Edgecumbe earthquake, 1987 March 2 (ML 6.3), was associated with renewed tectonic rupture on the Edgecumbe Fault, renewed movement on the Onepu and Rotoitipakau Faults, and several new surface breaks-the Awaiti, Otakiri, Te Teko, and OmeheuFaults. These northeast trending tectonic ruptures are widely distributed across the Rangitaiki Plains, range in length from 0.5 to 7 km, and are mostly downthrown to the northwest. They are associated with warping and prominent fissuring. Maximum displacement, 2.5 m vertical and 1.8 m extensional, occurred near the middle of the Edgecumbe Fault trace. Trenching investigations revealed that perhaps two faulting events have occurred in the past 1850 years in addition to the 1987 event; the earlier one is unsubstantiated, but may have occurred at about 1850 years B.P., and another occurred at about 800 years B.P. (the time of deposition of the Kaharoa Ash). The 800 year B.P. event was associated with warping, but little fissuring. An average slip vector, derived mainly from laterally offset cultural features, trends 330°, plunges 55°, and represents up to 3.1 m of normal fault slip on a plane of average strike about 055°. Our proposed fault model has a 55° dipping plane curving upwards to become almost vertical within 10 m of the ground surface. This 55° dip is probably representative to at least 100 m depth. The tectonic effects of the earthquake were influenced by the soft, wet sediments forming the Rangitaiki Plains and the relatively shallow hypocentre, but were otherwise typical of normal faulting events expected in the Taupo Volcanic Zone.

Seismic reflection and outcrop data from the onshore Hikurangi forearc reveal the styles and history of deformation for a c. 3000 km 2 region between Dannevirke and Hawke' s Bay. The data cover the forearc basin, including its western and eastern boundaries, and delineate folds and faults in a late Miocene-Recent sedimentary sequence. Five seismic horizons, including the basement/Neogene cover unconformity (variable age), base Waipipian (3.7 Ma), base Mangapanian (3.2 Ma), base Nukumaruan (2.6 Ma), and base Castlecliffian (1.6 Ma) were identified using outcrop and well ties. These horizons were traced across the study region to provide information on the geometry, spatial distribution, and timing of structures. To the west, the rangefront fault is predominantly reverse and separates uplifted Torlesse basement of the axial ranges from Neogene forearc basin sediments. Structures within the forearc basin are dominated by north-northeast-striking reverse faults and associated asymmetric folds which parallel the subduction margin and often have sinuous traces. Faults are planar to depths of at least 1-2 km and typically dip at 30-80°NW (most often at 40-70°). Many faults in the basin terminate northwards and fault-normal spacings decrease from 2-8 km in the Dannevirke area to c. 20 km near Hastings. Angular unconformities and syntectonic strata constrain the timing of deformation on fault/fold pairs. Faults within the forearc basin were active over two main periods, at c. ?3.1-2.5 Maandc. 1.6 Ma-Recent. Active fault traces are confined to the edges of the forearc basin. In the west, the Mohaka and Ruahine Faults have mainly recent right-lateral offsets, but interpretation of a seismic reflection line which crosses the Mohaka Fault indicates minimal (<300 m) right-lateral displacement over the last 3 m.y., and lateral slip may not significantly predate the late Quaternary. In the east, a zone of reverse faults (including the Longlands, Poukawa, Tukituki, and Oruawharo Faults) began forming ate. 1 Ma and remains active. Margin-normal shortening across the forearc basin and basin-bounding structures ranges up to 5 mm/yr (120- *

The magnitude 8 Wairarapa, New Zealand, earthquake of 1855 was associated with surface rapture along the Wairarapa fault and regional uplift of the southwest of the North Island. Forward elastic dislocation modelling shows that movement on a steeply dipping Wairarapa fault alone cannot account for the recorded deformation data. Modelling of movement on the subduction interface that underlies the Wellington region as well as the Wairarapa fault also fails to produce a satisfactory fit to the data. Although a complex Wairarapa fault model may be able to explain the deformation pattem if its location, subsurface geometry, and slip distribution could be indcpea•denfly constrained, the best effort supix)ned by available data, a flexed model incorporating a left side step of 8 km at the surface, incorrectly locates the deformation. The best fit to the data is obtained from a listtic Wairarapa fault model involving rapture on 0 to 50 km width of the deeper part of the subducfion interface. The shallower part of the subducfion interface, east of the Wairarapa fault, apparently did not rupture in 1855, and the uplift mechmfism for the overlying Aorangi Range remains unexplained. Partitioning of strike-slip and dip-slip components of the relative plate motions may involve separate earthquakes. Seismological verification of listtic fault rapture mechanisms is required to determine the plausibility of the listtic model presented here, because its implications are tlmt the 1855 earthquake did not completely account for fi•e relative plate motion in the region. 12.375confirming the 12-m value. 12,378DARBY AND BEANLAND: • 1855 WAIRARAPA EARTHQUAKE, NEW ZEALAND

No abstract