Session 1 Session 2 Session 3 Session 4 Session 5 Session 6 Posters 

The 2003, MW 7.2 Fiordland Earthquake, and its near-source aftershock strong motion data

P. McGinty

The 2003 Fiordland earthquake was not only the best ever recorded interface earthquake to occur in New Zealand, it also provided the opportunity to collect near-source strong-motion data produced by its aftershocks covering a wide magnitude range. Near-source strong-motion data had been lacking in the New Zealand data set, on which current attenuation models are based. Here I present some preliminary results relating recorded peak ground accelerations in the near-source field to current attenuation models. The near-source data from the 2003 Fiordland earthquake sequence has shown that the observed data has a greater magnitude-dependence than that predicted by the current attenuation models. This new data will help to improve current models and lead to a better understanding of the attenuation process associated with New Zealand interface earthquakes.

Paper P19: [Read]

Empirical models for estimating liquefaction-induced lateral spread displacement

J.J. Zhang and J.X. Zhao

Existing empirical models for estimating liquefaction-induced lateral spread displacement (D_LL) were derived from a dataset poorly-distributed with respect to earthquake magnitude and source distance, and also, the dataset are from earthquakes with different tectonic source types and faulting mechanisms. These drawbacks may lead to unacceptably large uncertainties. To overcome these problems, we used modification factors, accounting for effects of topographic features, material properties and the nonlinear nature of the soil response, to scale the pseudo-displacement obtained from a spectral acceleration attenuation model for soft soil class. We selected an attenuation model derived from a very large and reasonably well-balanced Japanese dataset, supplemented by world-wide near-source records and this model accounts for effects of earthquake tectonic source types and faulting mechanisms. We determined the model coefficients of the modification factors by selecting the pseudo-displacements calculated from a number of spectral periods to achieve unbiased residuals distribution with respect to earthquake magnitude and source distance. Comparison with a limited number of data from the 1997 Kocaeli, Turkey earthquake suggests adequate estimates of the present model which can be considered as more robust than the existing models.

Paper P20: [Read]

Mapping active faults and mitigating surface rupture hazard in the Kapiti Coast District, New Zealand

R.J. Van Dissen, D. Heron, S. Hinton and A. Guerin

All known active faults in the Kapiti Coast District have been mapped according to the methodology presented in the Ministry for the Environment’s Interim Guidelines on planning for development of land on, or close to active faults. Fault Avoidance Zones are defined along all the faults based on the rupture complexity of the fault, and the precision to which its location can be constrained. These zones range in width from about 40 m to greater than 300 m, and are attributed as well defined, distributed, uncertain - constrained, or uncertain - poorly constrained. Based on existing data, a Recurrence Interval Class (RIC) is defined for each fault: Ohariu & Northern Ohariu faults, RIC II (>2000 to ≤3500 yrs); Gibbs & Otaki Forks faults, RIC III (>3500 to ≤5000 yrs); SE Reikorangi fault, RIC IV (>5000 to ≤10,000 yrs). Building Importance Category is used to characterise building type, and a hierarchical relationship is established between Building Importance Category and RIC, such that the greater the life-safety importance of a structure, the longer the fault avoidance RIC. By linking Building Importance Category and development status of a site (i.e. previously developed site, or undeveloped site) with RIC and Fault Avoidance Zones, a matrix of risk-based Resource Consent Categories (e.g. permitted, discretionary, non-complying) is defined to facilitate the mitigation of surface rupture hazard and assist in the responsible development of land on, or close to the active faults in the Kapiti Coast District.

Paper P21: [Read]

Microzoning effects on damage in two large New Zealand earthquakes

D.J. Dowrick, D.A. Rhoades and P.N. Davenport

This paper discusses microzoning effects on damage to houses and household contents in the 1931 M_w 7.8 Hawke’s Bay and the 1968 M_w 7.2 Inangahua earthquakes, covering a wide range of intensities, MM6.4 – MM10.5. The present study extends the range of intensities considered in a previous study by including Nelson at MM6.4. Different foundation and superstructure types are also considered. Some of the important complexities and surprises in our previous study are eliminated or explained. Over the range of intensities MM6.4 and MM10, D_rm for weatherboard houses on unbraced pile foundations is generally higher than for houses on concrete foundations. For houses on unbraced piles, D_rm is mostly lower for houses and contents for Ground Class D than for Ground Class C, for intensities MM6.4 to MM10.5 (except when ground damage occurs), the difference being statistically significant in about half of the comparisons made. This may result from differences in frequency content of the ground shaking. Microzoning maps need to be based on more information than that on surface geology maps, and the required extra criteria (such as engineering properties of the soil and structures) need to be better understood. The importance of soil-resonance cannot be over-emphasised.

Paper P22: [Read]

Development of tests for probabilistic seismic hazard models from historical and prehistorical data

M.W. Stirling

I compare predicted rates of seismic hazard (felt intensities or peak ground accelerations) from the national seismic hazard models for New Zealand and the USA against the historical rate of exceedance for specific felt intensity levels at towns and cities in the two countries. The comparisons reveal a tendency for the PSH model to slightly overestimate the historical hazard at about 30 towns and cities in New Zealand, and significantly underestimate the historical hazard at a similar number of centres in southern California, and across the continental USA. The discrepancies in the USA are most marked in the areas of lowest seismicity and seismic hazard, and where strong site response is likely to be observed during large earthquakes. The results of these comparisons are preliminary, and future work is expected to be geared towards determining the reasons for and statistical significance of the discrepancies observed in southern California/continental USA, the difference in result between southern California/continental USA and New Zealand, standardising the procedure for the three areas, and introducing constraints on ground motions provided by ancient precariously-balanced rocks to the overall analysis.

Paper P23: [Read]

Earthquake risk assessment for insurance purposes

W.D. Smith, A.B. King and W.J. Cousins

A prime requirement of the insurance industry is information on the level of risk to which insured assets are exposed. Insurers would like to know the probability that any given loss level will be sustained, because this enables appropriate premiums to be set. Until recently providers of earthquake insurance have had to use best guesses about the performance of buildings, together with regional aggregations of properties into geographical zones, to arrive at a hazard rating and thus set premium levels.

New procedures are now available to use (a) the best available model of earthquake occurrence, in terms of active faults and background seismicity, (b) appropriate attenuation functions to estimate the severity of strong ground motion, and (c) available knowledge of the performance of various types of buildings under earthquake loading, to provide the probabilistic loss estimates that are needed. Innovative insurance conditions can then be applied, in order to provide asset owners with an insurance package that represents both good governance and cost-effective cover.

Paper P24: [Read]

Session 1 Session 2 Session 3 Session 4 Session 5 Session 6 Posters