Earthquake source characterization by the isochrone back projection method using near-source ground motions

Abstract

Rapid and accurate assessment of source characteristics of moderate to large earthquakes is extremely useful for hazard assessment and to guide response of emergency services. Using the back projection method (BPM) it is possible to obtain an image of the source rupture process rapidly. The potential of the method in identifying the rupture propagation and its slip distribution has been shown in previous studies. However, most of the earlier back projection implementations obtain only slip intensities not slip amplitude. Here, we propose a method that is capable of providing quick estimates of the slip amplitude in addition to its distribution across the fault plane, using high frequency near-source records. First, we explore the advantages and limitations of the proposed BPM through a series of synthetic examples. We demonstrate the utility of the method to identify slip asperities and their associated intensities, with a limited number of stations (<5) distributed azimuthally around the source in the ideal case. As expected, the accuracy of asperity locations and amplitudes are improved as the number of stations increases and an appropriate station-weighting scheme is introduced. To test the BPM, we apply the method to the 2004 Mw 6.0 Parkfield earthquake using available near-source seismic data. The method identifies similar locations and amplitudes of slip using either P- or S-wave displacement records. And, for real earthquakes, we find that a significant number of observations are needed around the source to reduce the influence of local propagation and site effects.