In what settings do turbidites provide a record of major earthquakes?

Abstract:

Turbidite paleoseismology has the potential to reveal extended earthquake records. The concept of turbidite paleoseismology is underpinned by two assumptions: i) every large earthquake produces a turbidite (i.e. the integrated record for an area is complete); and ii) earthquake-triggered turbidites have special characteristics that enable them to be distinguished from turbidites triggered by other mechanisms (i.e. the record is extractable). The most reliable characteristic for inferring an earthquake trigger is probably evidence of widespread triggering (due to widespread ground shaking), which may be inferred from synchronous triggering or confluence tests given sufficiently precise age control.

Testing whether every large earthquake produces a widespread turbidite requires independent evidence of the timing of earthquakes. Coring in the region of the 2004 (Mw 9.1) and 2005 (Mw 8.7) Sumatran earthquakes suggests that they did not produce widespread turbidites within intraslope basins. Coherent landslides are rare on this margin and as yet no landslide has been found that correlates in age with a known earthquake. Whilst smaller-scale landslides are more common, analogy with larger scale failures suggests that any earthquake record encapsulated within them is likely to be complicated (including potentially by other triggers) and incomplete.

Some settings may be more or less appropriate to apply turbidite paleoseismology because of earthquake characteristics such as the nature of slip, or margin characteristics such as sedimentation rates, material properties, mechanism of slope failure and the frequency of seismic events. Further work is needed in locations with well characterized records of recent major earthquakes, and comparisons between the sedimentary records on these margins, in order to understand the types of earthquakes and settings that provide a complete and extractable record of major earthquakes.