The 1904 Ms6.8 Mw7.0-7.2 Cape Turnagain, New Zealand, earthquake

The 1904 August 09 NZT (August 08 UT) MS6.8 earthquake caused widespread structural and chimney damage from Napier to Wellington and was felt over a large part of New Zealand. Other than a brief paper in 1905, and determinations of its surface wave magnitude in the last 20 years, little has been done to better locate the earthquake or detail its effects. Comprehensive data have now been obtained from searches of historical documents, including newspapers, private and government papers, as well as instrumental records. Interpretation of the intensity data shows that the earthquake was probably centred near Cape Turnagain at relatively shallow depth. The paucity of aftershocks suggests that the earthquake occurred either on the subduction interface, or in the lower seismicity band or upper mantle of the subducting Pacific Plate. The area encompassed by the higher intensity isoseismals suggests the earthquake had a magnitude greater than the calculated surface wave magnitude MS6.75 ± 0.14 — possibly as high as MW7.2. At this magnitude, the earthquake becomes a more significant event in New Zealand’s historical record, and certainly the largest earthquake suspected of rupturing the plate interface along the Hikurangi Margin. A notable feature of the earthquake is the chimney and parapet damage caused in parts of Wellington Central Business District, approximately 170 km from the epicentre. Much of the city and inner suburbs experienced MM5-6, while MM6-7 occurred in several areas, mostly in those areas that are recognised as possibly susceptible to shaking enhancement, but also in several locations outside these areas. The 1904 Cape Turnagain earthquake has several implications for seismic hazard dependent on whether it was intra-slab or on the plate interface. Of particular importance, are the questions whether the damage in Wellington is exceptional and could represent microzone, focussing or directivity effects; the goodness of fit of the intensity distribution to modelled isoseismals using published attenuation relations; the compatibility of the magnitude with the maximum magnitude/magnitude cut-offs used in this area in the New Zealand Probabilistic Seismic Hazard model; and finally, the possibility that the 1904 earthquake might characterise plate interface earthquakes in southern Hawke’s Bay.