The 2010–2011 Canterbury earthquake sequence (CES) caused up to 10 episodes of liquefaction at highly susceptible sites in eastern Canterbury, resulting in severe damage to land and infrastructure. Subsurface investigations at five sites over two study areas revealed CES dikes and sills that align with and crosscut pre‐CES liquefaction features, including dikes, a lateral sill, a sandblister, and a buried compound sandblow. Crosscutting relationships combined with carbon‐14 (14C) dating constrain the timing of the pre‐CES liquefaction features to likely post‐A.D. 1321 and pre‐1960 in one study area. Pre‐CES features in the second study area likely formed in three distinct episodes: post‐A.D. 1458 and possibly during the 1901 Cheviot earthquake, between A.D. 1297 and 1901, and pre‐A.D. 1458. The liquefaction potential of known active faults within the wider Canterbury region are evaluated from back‐calculated magnitude‐bound curves and peak ground acceleration (PGA) approximated using a New Zealand‐specific ground‐motion prediction equation and compared with global liquefaction triggering thresholds. Analysis indicates that many active faults within North Canterbury and offshore that are within 50 km of the study sites and capable of triggering Mw>6.5 earthquakes have the potential to cause widespread liquefaction. Ruptures of these faults may have formed the pre‐CES liquefaction features. Combining the backcalculation approach with the modeled PGA proves effective in determining the active faults capable of triggering liquefaction at the study sites and are therefore capable of triggering liquefaction in the future.
Online Material: Overview and further discussion of the probabilistic magnitude‐bound methodology framework and derivative curves, description of sediment units, and table of peak ground acceleration (PGA).