Imaging the Rupture Process of the 2004 Sumatra-Andaman Earthquake Using Deconvolved Surface Wave Source Time Functions Inversion of surface wave source time functions for a two-dimensional slip distribution on the fault for the 2004 Sumatra-Andaman earthquake. Observed and modeled (red) surface wave source time functions for 12 GSN stations are shown on the upper left. The overall source time function for the rupture obtained from the finite source model is shown on the lower left. The distribution of slip along the interplate fault is shown in map view on the right, with larger symbols indicating larger dip-slip sliding on the fault.
Rapid determination of the slip distribution for great earthquakes is important for the assessment of their tsunamigenic potential. While analysis of high frequency radiation offers good potential for rapidly constraining total rupture duration, it is important to characterize the magnitude of slip on the fault using longer period signals. This was demonstrated by the tsunami excitation for the 2004 Sumatra-Andaman earthquake, for which reduced slip in the northern portion of the fault plane results in weak tsunami excitation near the Andaman Islands. We have developed a rapid analysis procedure that requires relatively little a priori information about the faulting in order to extract a fairly robust estimate of the slip distribution along an interplate fault. This involves deconvolution of global GSN recordings of long-period Rayleigh waves by theoretical Greenʼs functions for an interplate thrust mechanism, followed by either inverse radon transform for a one-dimensional slip distribution or inversion of the surface wave source time functions for a two-dimensional slip distribution. This procedure can be applied to the full surface wave recordings at relatively short distances from the earthquake to extract a first-order characterization of the earthquake rupture within 20-30 minutes after the event.
Ammon, C., et al., Rupture process of the 2004 Sumatra-Andaman earthquake, Science, 308, 1133-1139, 2005.
Supported by NSF grant EAR-0125595.