PALEOLIQUEFACTION LESSONS LEARNED FROM THE 2010-2011 CANTERBURY, NEW ZEALAND, EARTHQUAKES

The September 4, 2010, moment magnitude, M_w, 7.1 Darfield and February 22, 2011, M_w 6.2 Christchurch earthquakes, which caused severe and extensive liquefaction particularly in Christchurch and nearby areas with a high water table and underlain by fluvial deposits of the Waimakariri, Heathcoat, and Avon Rivers, were produced by previously unknown faults. The Greendale fault, source of the Darfield earthquake, ruptured the surface for 29 km across the alluvial plain west of Christchurch, laterally displaced cultural and geomorphic features by 2 m on average, and produced scarps up to 1 m high. The Port Hills fault, source of the Christchurch earthquake, did not rupture the surface but caused centimeters of subsidence and uplift across its surface projection. The Darfield and Christchurch earthquakes, as well as two smaller (M 5-6) shocks, all within a 9-month period, induced recurrent liquefaction at numerous sites resulting in extrusion and deposition of water-entrained sediment and formation of compound sand blows. Related feeder dikes exhibit multiple phases of injection. In trenches of modern sand blows, paleoliquefaction features have been found. They are larger and more weathered than and crosscut by the modern sand dikes. In some locations, modern sand dikes intrude the margins of older sand dikes. Preliminary findings indicate that a paleoearthquake of at least M > 6.2 occurred between A.D. 1000 and A.D. 1400 in the Canterbury region. These finding suggests a shorter recurrence interval for large earthquakes than previously thought (M_w 6.2 in 2-3k yr). Lessons drawn from the Canterbury earthquakes include:

(1) Earthquake-induced liquefaction features are critical indicators of strong ground shaking in regions where fault ruptures are difficult to recognized or do not propagate to the surface;

(2) Sand blows composed of several depositional units form as the result of multiple earthquakes in a sequence lasting months;

(3) Sites of modern earthquake-induced liquefaction are prime targets for paleoliquefaction features due to suitable sedimentological and ground water conditions;

(4) In fluvial settings, paleoliquefaction studies may provide a more complete record of paleoarthquakes, and therefore a more accurate assessment of earthquake potential, than would be ascertained from fault studies alone.