ISSN:
1420-9136
Keywords:
Key Words: Time-to-failure analysis; seismicity; temporal variation.
Source:
Springer Online Journal Archives 1860-2000
Topics:
Geosciences
,
Physics
Notes:
Abstract —The M 5.1 event (May 23, 1993) which occurred in one of the most active swarm areas of Japan was preceded by foreshock activity. We obtained precise hypocenters of the foreshock-mainshock-aftershock sequences with a temporary seismic network installed just above the source region twenty days before the mainshock. The foreshocks are very unique in their accelerating activity; the acceleration in the number of foreshocks enabled us to estimate the time of the mainshock with time-to-failure analysis proposed by . Although substantial snow remained in the swarm area, we quickly installed the network because the time-to-failure analysis disclosed that the mainshock was impending. The temporary network provided detailed information on both the temporal and spatial distribution of the foreshock-mainshock-aftershock sequences. Foreshocks started fifty days before the mainshock and were distributed linearly at the base of the seismogenic layer with a length of 5 km and horizontal and vertical widths of about 1 km. The temporal change of the number of foreshocks is approximated by a power law, and the time of the mainshock can be estimated by extrapolating plots of the inverse of the daily number of events. An area of seismic quiescence appeared 40 hours before the mainshock and propagated with a rate of 20 m/hour. The mainshock occurred 2 km westward from the primary foreshock area. It was located at the base of the aftershock region. This process can be interpreted as source nucleation; preslip on the fault prior to the mainshock.
Type of Medium:
Electronic Resource
URL:
http://dx.doi.org/10.1007/s000240050268
Permalink