Probabilistic assessment of liquefaction potential for Christchurch in the next 50 years

(Inc. GST)
(Ex. GST)
Write a Review

Gerstenberger, M.C.; Cubrinovski, M.; McVerry, G.H.; Stirling, M.W.; Rhoades, D.A.; Bradley, B.; Langridge, R.M.; Webb, T.H.; Peng, B.; Pettinga, J.; Berryman, K.R.; Brackley, H.L. 2011 Probabilistic assessment of liquefaction potential for Christchurch in the next 50 years. Lower Hutt, N.Z.: GNS Science. GNS Science report 2011/15 25 p.

Abstract: At the request of Tonkin and Taylor, we provide preliminary estimates of the 50 year probability for damaging liquefaction in selected suburbs of Christchurch city that were most severely impacted by the M7.1 4 September 2010 Darfield and M6.3 22 February 2011 earthquakes. Seismicity in the region of the 4 September and 22 February earthquakes is presently very high relative to previous activity, requiring development of earthquake hazard estimates that model time-varying seismicity rates. We construct a composite earthquake hazard model to give the probability of exceedance for magnitude-weighted peak ground acceleration (PGA) levels that are expected to produce liquefaction. The composite model combines several earthquake source models that are based on different concepts and cover a wide range of time, space and magnitude scales. The models are: (1) the fault source model of the national seismic hazard model (NSHM; Stirling et al, in prep), with enhancement of the earthquake probabilities for the most major active faults near to the Canterbury region (Hope and Alpine Faults) to take account of the time elapsed since last earthquake; (2) the Proximity to Past Earthquakes (PPE) smoothed seismicity model; (3) the Short Term Earthquake Probability (STEP) model (Gerstenberger et al, 2005), which targets aftershocks and; (4) the Every Earthquake a Precursor According to Scale (EEPAS) model (Rhoades and Evison, 2004) which looks at longer-term earthquake clustering. Earthquake hazard levels produced by the composite model are found to be high in the region of the city, and equivalent to the hazard levels close to the major active faults to the northwest of Christchurch. PGA calculations are provided by the McVerry et al (2006) ground motion prediction equation (GMPE). The calculated PGA estimates incorporate a magnitude-dependent scaling as required for the liquefaction analysis. A critical issue is that the 22 February earthquake produced peak ground accelerations over much of the Christchurch region that were about twice the values modelled by the McVerry et al GMPE. Stress-drop scaling is included to partially account for this, however it is also likely to be due to specific characteristics of the causative fault, for example the rupture directivity of the earthquake, and is presently the focus of further investigation. Some of these characteristics are likely to occur in future earthquakes in the region, increasing the ground-motions from those given by the GMPE used in this study. Liquefaction evaluation is provided for the five suburbs of Avonside, Dallington, Avondale, Burwood and Bexley. These suburbs were chosen as a representative sub-set of urban areas most severely affected by liquefaction and lateral spreading in the 4 September and 22 February earthquakes, and for the availability of soil geotechnical data such as from cone penetrometer tests (CPT). We use a conventional semi-empirical approach for liquefaction evaluation that is based on CPT data (Youd and Idriss, 2001). Using an empirical CPT-based liquefaction curve, PGAs likely to produce damaging liquefaction are calculated for each of the suburbs for two soil depth levels. These PGAs are then compared to the PGAs produced from the composite hazard model for the next 50 years, and other time periods. Results show 50-year probabilities for damaging liquefaction to be between about 20% and 97% for the various deposits evaluated. In other words, liquefaction probabilities for the next 50 years are high for the most severely affected suburbs of the city, and are well in excess of the probabilities associated with the ground-shaking design levels defined in the New Zealand structural design standard NZS1170 (Standards New Zealand 2004). (auth)