Kinematic ruptures on faults with multiple segments

1
Kinematic ruptures on faults with multiple
segments

• Chen Ji, UCSB

2
Most kinematic models dont include multiple
segments because

• Fault geometry is not very clear.
• Observations cant constrain

3
1992 Landers Earthquake
Wald Heaton, 1994, BSSA
4
1992 Landers Earthquake Wald Heaton, 1994
Rupture velocity drops when rupture meets the
compression stepover
5
1999 Chi-Chi earthquake Ji et al., 2003

• Seismic moment 2.7 1020 Nm.
• Average rupture velocity is about 2.0 km/sec.
• Average rise time 7.2 sec

6
Irregular fault surface

• Slip distribution shows eastward low slip zones
  (pointed by arrows) when the fault trace changes
  directions
• The downward extensions of the low slip zones
  suggest that the surface curvature is a
  deep-seated feature.

Slip distribution
7
Snapshot
North
South
Dashed contours shows the rupture fronts in every
3 sec if the rupture velocity is 2.1 km/sec.
8
Rupture propagation in an irregular fault plane
North
South
Rupture velocity drops when rupture meets the
compression stepover
9
Rupture turns

• The intersection is not normal to the fault
  strike direction.
• Near the hinge line, the uplift of the hanging
  wall will create additional space for extension.

10
1999 Hector Mine Earthquake Ji et al, 2002, BSSA
11
1999 Hector Mine Earthquake Ji et al, 2002,
BSSA
12
What is next in finite fault inversion?

• Good spatial and frequency data coverage
• - High rate GPS
• Good Greens functions, 3D structure
• Accurate fault geometry, 3D planes

• good inverse programs.

13
Summary

• Kinematic slip distributions of several large
  earthquakes support the argument that compression
  stepovers slow down the rupture propagation.
• No clear evidence so far in my study that
  dilatation stepovers slow down the kinematic
  rupture.

14
Observations

• Rupture of one fault segment can be a single
  earthquake or part of a large event Bad !!!
• Regions with high inter-seismic seismicity rate
  may be the earthquake nucleation region but
  unlike to have large slip during the main-shock
  rupture Good !!!

Note Earthquakes I have studied so far are
larger than 6 (rupture planes are larger than 100
km2).
15
Afterslip
2004 Sumatra-Andaman Earthquake