Implications for friction and fault
zone rheology from early and repeating aftershocks

Zhigang Peng*
Postdoctoral Researcher
Department of Earth and Space Sciences
3686B Geology Building
595 Charles E Young Drive East
University of California, Los Angeles, 90095-1567
Phone: (310)8251420
Email: zpeng@ess.ucla.edu
Web: http://moho.ess.ucla.edu/~zpeng


* Work done in collaboration with John Vidale, Miaki Ishii, Agnes
Helmstetter, Chris Marone, and Allan Rubin

ABSTRACT

Part I: Mainshock rupture is typically followed by aftershocks that
diminish in rate approximately as the reciprocal of the elapsed time.
However, many aftershocks are missing in existing seismicity catalogs in
the first few minutes after the mainshock.  Yet this period holds valuable
information about the transition from mainshock rupture to sporadic
aftershocks, and about the friction laws that control the occurrence of
earthquakes.  We have analyzed waveforms of ~80 M3-5 shallow earthquake
sequences recorded by the Hi-Net borehole array in Japan.  By scrutinizing
the high-frequency signal, we have detected up to 8 times more aftershocks
in the first 900 s than in the JMA catalog.  The aftershock activity
appears to evolve from fairly steady rate in the first 50-100 seconds to a
power law decay rate later (with p value in the range 0.8-0.9), as
predicted by the Dieterich's rate-and-state friction model.  Other
possible mechanisms consistent with data include dynamic shaking of the
mainshock, fault healing, or fluid diffusion.

Part II: The recurrence intervals for 194 repeating clusters on the
Calaveras fault and their elapsed times after the 1984 M6.2 Morgan Hill,
California, earthquake follow a power-law decay relation.  The decay rates
of deep repeating clusters (4-7 km) systematically exceed those of the
shallow clusters (1-4 km).  The trends between relative moment and
recurrence interval, which is a measure of the fault-healing rate, vary
systematically with hypocentral depth of the repeating clusters.  A
strain-rate dependent transient loading in the creeping zone surrounding a
mainshock slip patch can explain both variations in decay rates and moment
trends.  Alternatively, fault-healing rate can be affected by the average
shear stress level or temperature.  Our observations suggest that
systematic behavior of repeating aftershocks reflect variations in the
fault zone rheology.