Liquefaction features and the geologic environment in which they formed were carefully studied at two sites near Lincoln in southwest Christchurch. We undertook geomorphic mapping, excavated trenches, and obtained hand cores in areas with surficial evidence for liquefaction and areas where no surficial evidence for liquefaction was present at two sites (Hardwick and Marchand). The liquefaction features identified include (1) sand blows (singular and aligned along linear fissures), (2) blisters or injections of subhorizontal dikes into the topsoil, (3) dikes related to the blows and blisters, and (4) a collapse structure. The spatial distribution of these surface liquefaction features correlates strongly with the ridges of scroll bars in meander settings. In addition, we discovered paleoliquefaction features, including several dikes and a sand blow, in excavations at the sites of modern liquefaction. The paleoliquefaction event at the Hardwick site is dated at A.D. 908-1336, and the one at the Marchand site is dated at A.D. 1017-1840 (95% confidence intervals of probability density functions obtained by Bayesian analysis). If both events are the same, given proximity of the sites, the time of the event is A.D. 1019-1337. If they are not, the one at the Marchand site could have been much younger. Taking into account a preliminary liquefaction-triggering threshold of equivalent peak ground acceleration for an Mw 7.5 event (PGA7:5) of 0:07g, existing magnitude-bounded relations for paleoliquefaction, and the timing of the paleoearthquakes and the potential PGA7:5 estimated for regional faults, we propose that the Porters Pass fault, Alpine fault, or the subduction zone faults are the most likely sources that could have triggered liquefaction at the study sites. There are other nearby regional faults that may have been the source, but there is no paleoseismic data with which to make the temporal link.
In this paper, we perform hybrid broadband (0-10 Hz) ground motion simulations for the ten most significant events (Mw 4.7-7.1) in the 2010-2011 Canterbury earthquake sequence. Taking advantage of having repeated recordings at same stations, we validate our simulations using both recordings and an empirically-developed ground motion prediction equation (GMPE). The simulation clearly captures the sedimentary basin amplification and the rupture directivity effects. Quantitative comparisons of the simulations with both recordings and the GMPE, as well as analyses of the total residuals (indicating model bias) show that simulations perform better than the empirical GMPE, especially for long period. To scrutinize the ground motion variability, we partitioned the total residuals into different components. The total residual appears to be unbiased, and the use of a 3D velocity structure reduces the long period systematic bias particularly for stations located close to the Banks Peninsula volcanic area.
We present ground motion simulations of the Porters Pass (PP) fault in the Canterbury region of New Zealand; a major active source near Christchurch city. The active segment of the PP fault has an inferred length of 82 km and a mostly strike-slip sense of movement. The PP fault slip makes up approximately 10% of the total 37 mm/yr margin-parallel plate motion and also comprises a significant proportion of the total strain budget in regional tectonics. Given that the closest segment of the fault is less than 45 km from Christchurch city, the PP fault is crucial for accurate earthquake hazard assessment for this major population centre. We have employed the hybrid simulation methodology of Graves and Pitarka (2010, 2015), which combines low (f<1 Hz) and high (f>1 Hz) frequencies into a broadband spectrum. We have used validations from three moderate magnitude events (𝑀𝑤4.6 Sept 04, 2010; 𝑀𝑤4.6 Nov 06, 2010; 𝑀𝑤4.9 Apr 29, 2011) to build confidence for the 𝑀𝑤 > 7 PP simulations. Thus far, our simulations include multiple rupture scenarios which test the impacts of hypocentre location and the finite-fault stochastic rupture representation of the source itself. In particular, we have identified the need to use location-specific 1D 𝑉𝑠/𝑉𝑝 models for the high frequency part of the simulations to better match observations.
