A Multi-fault Model Estimation from Tsunami Data: An Application to the 2018 M7.9 Kodiak Earthquake

In this study, we developed a new search algorithm to find a multi-fault model of a complex earthquake using tsunami data, and applied it to the January 23, 2018 M7.9 Kodiak earthquake. Our method includes a Green’s function based time reverse imaging (GFTRI) approach to invert for sea surface displacement using tsunami waveforms, followed by inversion of the sea surface displacement for the earthquake slip distribution. The global CMT focal mechanism for this event indicates that faulting occurred on a steeply dipping fault striking either N–S (right lateral) or E–W (left lateral), while subsequent work reveals a more complex pattern of strike-slip faulting. We carried out a number of source inversions using different combinations of faults to find the model based on an extremum for residual errors. Our results suggest that the rupture occurred on at least three faults oriented in approximately N–S and E–W directions. We further explored the fault-geometry parameters by perturbing them within a range suggested by previous work. We found that the sea surface displacement model is best fit by our preferred three fault-model with the set of parameters (strike, dip, rake): ($$165^{\circ }$$, $$60^{\circ }$$, $$154^{\circ }$$) and ($$265^{\circ }$$, $$60^{\circ }$$, $$10^{\circ }$$) for N–S and E–W directions, respectively.