Slip to Trench: Coseismic, Postseismic, or Interseismic?

Abstract:

It is poorly known whether the shallow parts of subduction megathrust faults commonly weaken in great earthquakes to cause coseismic slip to trench, or strengthen to resist coseismic slip but slip to trench afterwards. The 2011 M 9.0 Tohoku-oki earthquake offers an example of coseismic slip to trench. Comparison of high-resolution bathymetry data collected over the Japan Trench before and after the earthquake shows an overall seafloor uplift of ~ 10-20 m landward of the trench. By modeling the bathymetry change, we determine a coseismic fault slip up to ~ 60 m at the trench. In comparison, GPS displacements recorded on the near-trench (~ 60 km) islands after the 2005 M 8.7 Nias, Sumatra, earthquake suggest large shallow afterslip, although coseismic slip at the trench cannot be resolved. The 2012 M 7.6 Nicoya, Costa Rica, earthquake provides an unambiguous example for slip to trench only after the earthquake. The difference in seafloor pressures observed at sites < 1 km apart on the seaward and landward sides of the thrust outcrop did not show any change during the earthquake, indicating no coseismic slip to trench; only afterwards did pressure differences indicate gradual uplift of the prism toe relative to the incoming plate, suggesting afterslip reaching the trench. Finally, at Nankai, occurrences of very-low-frequency earthquakes at shallow depths (< 10 km below seafloor) and concurrent borehole pressure transients near the trench suggest that episodic slip to trench occurs well into the interseismic interval there. These observations showing distinctly different shallow fault behaviour raise several questions: What factors control the behaviour of the shallow fault? Is the shallow fault behaviour margin-dependent? Can the same shallow fault behave differently in different earthquakes? We speculate that earthquake size and the amount of slip play dominant roles. The shallow megathrust tends to exhibit rate-strengthening at low slip rates and resists seismic rupture such as in the Costa Rica earthquake, but a sufficiently large slip of the deeper rupture zone may force the shallow part to slip and may even cause dynamic weakening at high slip rates such as in the Tohoku-oki earthquake.