A two step inversion of the 1992 Landers earthquake rupture process using radar interferometry and strong motion data.
Abstract. A radar interferogram is used in combination with classical geodetic and seismological data to study the rupture process.
First, we use interferometric data alone to constrain slip distribution on a planar faults model along both strike and dip direction. The results are then compared with other inversions to know if such rich data set provide additional information compared to classical data. Two inversions techniques (least-squares and genetic algorithm) are applied to interferometric and GPS data, alone and joint. The uniqueness of the solution and the parameter resolution with depth are discussed. These techniques are applied to synthetic data to analyze the quality of the inversion. Finally, strong motion data are inverted with the a priori final slip amplitude and its resolution worked out thanks to geodetic data to constrain the temporal development of the rupture process.
We conclude from the results of these inversions that in the case of the Landers earthquake, interferometric data constrain the slip distribution due to the rich spatial distribution of radar interferometry. The parameters resolution decreases with depth. Below 10 km several models with various values of slip at depth can equally well explain the data. The results are in agreement with geological observations and confirm that the slip distribution of the Landers earthquake is heterogeneous. The inversion of the strong ground motion with the a priori final slip distribution determined by geodetic data, emphasize a strong variation of the temporal development of the earthquake. Fast rupture front velocities appear within asperities and the rupture slows when it encounters a resistance along the fault. On average the rupture front propagates with velocities close to the S-wave speed and terminates about 20 s after initiation.