HR: 10:35h
AN: S12A-02    [Abstracts]
TI: Radiated Seismic Energy and Mean Square S-Wave Spectral Comparisons for Large Earthquakes
AU: * Boatwright, J
EM: boat@usgs.gov
AF: US Geological Survey, 345 Middlefield Road, MS 977, Menlo Park, CA 94025 United States
AU: Venkataraman, A
EM: anupamav@pangea.stanford.edu
AF: Stanford University, Dept Geophysics/Mitchell Bldg, Stanford, CA 94305-2215 United States
AU: Mayeda, K
EM: kmayeda@llnl.gov
AF: Lawrence Livermore National Laboratory, L-205, P.O. Box 808, Livermore, CA 94551 United States
AU: Choy, G L
EM: choy@usgs.gov
AF: US Geological Survey, Denver Federal Ctr P.O. Box 25046 MS 967, Denver, CO 80225 United States
AU: Cocco, M
EM: cocco@ingv.it
AF: Istituto Nazionale Geofisica e Vulcanologi, Via di Vigna Murata 605, Rome, 00143 Italy
AB: Refinements in the analyses of regional waves, coda waves, and teleseismic body waves have produced a convergence of estimates of the seismic energy radiated by large strike-slip earthquakes. In particular, estimates of the seismic energy radiated by the 1992 Landers earthquake (~3-7x1015 joules) and the 1999 Hector Mine earthquake (~3-4x1015 joules) from regional S+ Lg+surface waves, (either analyzed directly or by using aftershocks as EGFs), the regional coda-waves, and the teleseismic P+pP+sP wave-groups are remarkably consistent. We test this convergence by comparing the mean square S-wave velocity spectrum, <u̇s*(ω)>2 = r2\over{4π} ∫∫r>>ζ | u̇s*(x,ω)|2 sinθ dθ dφ estimated from these different analyses. Here u̇s*(x,ω) is the radiated S-wave velocity spectrum at x(r,θ,φ), corrected for attenuation, and ζ is the source extent. We assume that the coda-wave analysis estimates this quantity directly because coda waves are multiply scattered. In contrast, the direct and EGF analyses of regional waves are weighted by station to yield appropriate estimates. The limited takeoff angles of the teleseismic P+pP+sP phases make the teleseismic estimates the weakest approximations of the mean square spectrum. We also consider the 1979 Imperial Valley, 1994 Northridge, and 2003 San Simeon earthquakes. While the mean square S-wave velocity spectra calculated from these different parts of the wave-field compare well for the strike-slip earthquakes, the regional spectra for the two thrust earthquakes exceed the teleseismic and coda spectra at frequencies above 0.5 Hz. Additionally, we use published kinematic slip models for these earthquakes to estimate the mean square S-wave spectra and seismic energy radiated by these earthquakes. Surprisingly, the mean square S-wave spectrum estimated from the Ji et al. (2002) model for the 1999 Hector Mine earthquake underestimates the observational estimates from 0.4 to 1.0 Hz. We intend to test this discrepancy more thoroughly by analyzing estimates of the mean square S-wave spectra from slip models for the four other large events.
DE: 7200 SEISMOLOGY
DE: 7203 Body waves
DE: 7209 Earthquake dynamics (1242)
DE: 7215 Earthquake source observations (1240)
SC: Seismology [S]
MN: Fall Meeting 2005