David Dowrick and David Rhoades
In empirical models of attenuation of strong motion, peak ground acceleration and spectral acceleration (PGA and SA), source-to-site distances are usually defined as the shortest distance from the site to a planar region, representing the fault rupture. Such a definition constrains the attenuation model to have racetrack-shaped lines of equal PGA or SA. It can result in errors (i.e., deviations of the fitted model from the data) which are substantial in the near-source region. This is demonstrated by comparison with the attenuation models that arise when no such constraints are built into the source-to-site distances, as is the case in the Dowrick and Rhoades attenuation model for Modified Mercalli intensity (MMI). The nature and magnitude of the errors in the strong-motion attenuation models are described for a magnitude 7.5 earthquake.
Paper P08: [Read]
Peter Davenport and Bill Stephenson
The 1968 Mw7.2 Inangahua earthquake caused damage to many houses on the West Coast of the South Island. During a study of the damage in Westport and Greymouth, a microtremor survey was carried out to assess the local site conditions in these towns. Employing the horizontal to vertical spectral ratio method to analyse ambient ground vibration is a practical application of a non-invasive technique to obtain site information relevant to seismic response. This paper reports the methodology used to select recording sites, gather the microtremor data, process the data and interpret the results. Comparison to results obtained at other locations in New Zealand indicates that some caution is needed in utilising this method.
Paper P09: [Read]
Anna Kaiser and Euan Smith
The first part of this study discusses the ReMi™ (Optim 2003) surface wave dispersion technique for determining shallow shear-wave velocities. Particular advantages and disadvantages of the ReMi technique are identified. Different sources of surface waves are compared for their effectiveness. Accurate dispersion results were obtained from both controlled sources (sledgehammer, dynamite and a hydraulic thumper) and noise sources (a vehicle running along the length of the array). Secondly, the shear-wave velocity results from the ReMi method at three sites in the Wairarapa are presented and used to make an assessment of the liquefaction risk, adding to information from previously used methods, such as cone penetrometer testing. Results indicate low shear-wave velocities at each site, in particular at the Lake Wairarapa Barrage site where S-wave velocity remains below 200m/s to a depth of 30m. Velocities are also low at the Kahautara Bridge site where liquefaction has occurred in the past, causing damage to the bridge in the 1942 Wairarapa earthquake. Shear-wave results indicate a potential liquefaction risk at these two sites, whereas the Ruamahanga Bridge site may have a lower liquefaction potential. This is in agreement with assessment based on CPT testing (Wick 2000).
Paper P10: [Read]
Effects of earthquake magnitude and source distance on the response spectral amplification ratios of soft-soil sites
John Zhao, Jian Zhang and Kojiro Irikura
We used a suite of rock site records from subduction slab earthquakes to evaluate the response spectral amplification ratio for soil sites using elastic and nonlinear modelling. The records are well distributed with respect to magnitude and source distance. Because a response spectrum is the peak of structural response to the excitation over a wide frequency range, response spectral amplification ratios of a soil site subjected two rock site records with an identical response spectral value at a given spectral period are not usually equal. Scatter of the amplification ratios is found to be considerable. Analyses of the scatter reveal that response spectral amplification ratios for an elastic soil site depend on both the earthquake magnitude and the source distance. It is also found that, at periods much shorter than or much longer than the site natural periods of the soil sites modelled as elastic, the amplification ratios decrease with increasing excitation, as does the scatter. The response spectral amplification ratios of a simple 2-dimensional nonlinear soil basin have similar characteristics to those of elastic sites. These findings have potential impact in establishing design spectra for soft soil sites using strong motion attenuation models or dynamic modelling.
Paper P11: [Read]