HR: 0800h
AN: S51B-1005    [Abstracts]
TI: Relocation and Assessment of the December 2004 Sumatra-Andaman great earthquake sequence and background seismicity
AU: * Abend, H
EM: habend@ldeo.columbia.edu
AF: Lamont Doherty Earth Observatory, 61 Route 9W - PO Box 1000, Palisades, NY 10964 United States
AU: Waldhauser, F
EM: felixw@ldeo.columbia.edu
AF: Lamont Doherty Earth Observatory, 61 Route 9W - PO Box 1000, Palisades, NY 10964 United States
AU: Engdahl, R
EM: engdahl@colorado.edu
AF: University of Colorado, Box 390 UCB, Boulder, CO 80309 United States
AB: We present earthquake relocation results from the Sumatra-Andaman region, covering the period from January 1964 to May 2005 and including the aftershock sequences of the December 2004 M 9.3 and March 28 M 8.7 earthquakes. Single event hypocenters as provided by the ISC or USGS are often in error by tens of kilometers, lacking the resolution needed to gain insight into the seismic and tectonic working of the region, and the mechanics of large earthquakes. The ISC/USGS bulletin data has been processed using a rigorous quality selection criterion to improve phase associations and pick accuracy. Reported depth phases are used to constrain absolute depths, as most events are recorded at far regional and teleseismic distances only. This data set of ~3200 events is then relocated using the double-difference algorithm, to improve relative hypocenters by inverting differential arrival times for event separation. Residual RMS values are typically 0.6 sec, and mean errors in relative locations in the range of about 6 km and a mean error in depth of about 7km. Much of the pre-mainshock seismicity occurred downdip (h > 35 km) along the thrust plane while most post- mainshock seismicity occurred trenchward (h < 35 km) in an area that lacks previous seismicity. Along the trench, the December 2004 mainshock ruptured a 1600-km-long segment of the plate boundary and the aftershock zone developed northward from the mainshock. The March 2005 mainshock ruptured a 300-km-long segment of the plate boundary and the aftershock zone developed southwestward from the mainshock. Although the immediate area of the December 2004 mainshock had been previously active, only a few aftershocks occurred there. Double-difference relocation of earthquake within the interplate zone indicate a seismogenic thickness of ~20-25 km in the area of the March 2005 event, and ~25-35 km in the northern tip of the Andaman microplate. Crustal events along the Sumatra fault reveal sharp boundaries along the northern part of the Andaman microplate and the Burma subplate. A previously defined straight line plate boundary appears as a step- like feature in the relocated seismicity. No specific fault structures are revealed in one of the major crustal swarms of about 700 earthquakes that occurred between 7.5-8.5 degrees N and 94-95 degrees E, indicating that this may be a complex step over region in the plate boundary, or that we lack the necessary resolution to image such complicated structures with the available data. Higher resolution is likely to be achieved by using waveform cross correlation measurements to improve on the phase pick data.
DE: 7230 Seismicity and tectonics (1207, 1217, 1240, 1242)
DE: 7240 Subduction zones (1207, 1219, 1240)
SC: Seismology [S]
MN: Fall Meeting 2005