American Geophysical Union, Fall Meeting, San Francisco, California, 5-9 December 2005
Festa G, Mai P M, Francois-Holden C, Burjanek J, Di Carli S, Emolo A, Gallovic F, Zollo A, Delouis B, Madariaga R.
Kinematic inversions of near-fault data allow to retrieve certain properties of earthquake rupture, such as the slip distribution, the rupture velocity and the slip rate which depends on the choice of the source time function. Accurate images of the spatio-temporal rupture process are also a necessary ingredient to move from kinematic source characterizations to reliable models of the dynamic fracture process. Several techniques have been developed in the past to carry out such source inversions, with an apparent increase in estimated rupture complexity which also reflects the increasing amount and quality of near-fault strong-motion data. However, the resolution, reliability and quality of these finite-source rupture models still remains an open question. Particular choices in the inversion method (fault parametrization, selection and processing of data, minimum search alogrithm and fitness function) influence the inversion results, leading to different and sometimes incompatible slip maps for the same earthquake. We therefore construct a blind test inversion exercise in which several research groups attempt to derive a kinematic rupture model from synthetic seismograms calculated for an input model unknown to the people performing the inversion. During several stages of this blind test, we increase the complexity of the input model while also the "data quality" will be decreased by adding random noise to the synthetic seismograms. The major aim of this study is to compare different source-imaging techniques, to understand their strengths and weaknesses and to check the processing of data. Besides inferring the general properties of the rupture, we will also be able to asses the resolution and possibly uncertainties in the model parameters, which will furnish guide lines and directions for future work. The test is carried out in the framework of the EU-SPICE project (Seismic wave Propagation and Imaging in Complex media - a European project), but is also open to other interested researchers.
UR: http://www.spice-rtn.org/members/mai/BlindTest
DE: 0500 COMPUTATIONAL GEOPHYSICS (3200, 3252, 7833) DE: 4450 Nonlinear maps DE: 7215 Earthquake source observations (1240) DE: 7290 Computational seismology SC: Nonlinear Geophysics [NG] MN: Fall Meeting 2005.