Kinematic and Dynamic Source Inversions using Elliptical Patches constrained by Teleseismic Source Time Functions and Source Spectra Estimation

Abstract:

Our understanding of the dynamic nature of the earthquake rupture process or natural earthquakes is still limited. On one side, seismologist try to estimate source parameters using simple point-source models to obtain source dimensions, average slip, stress drop, and seismic efficiencies, and in some cases rupture velocities as well. On the other side, kinematic source inversions are routinely performed on large earthquakes, and from the results estimates of these source parameters can also be obtained, in addition of fault geometry and slip distribution. More recently, dynamic source inversions are also performed, which results in a more physics-based inversion of the dynamic processes that act on the fault surfaces, including stress and friction parameters. All these methods have advantages and limitations. In particular, even though the physics-based dynamic source inversions are probably the more appropriate approach, the large number of parameters that need to be inverted for and the significant trade-offs between these parameters result in a infinite number of solutions that need to be explored using Monte Carlo methods. A particular approach has been put forward in the last 10 years to perform these kinematic and dynamic source inversions using elliptical patches, which significantly reduces the number of parameters to invert for, while still fitting the data. We have found that even in these cases, dynamic source inversions show significant trade-offs between the dynamic source parameters. We combine these three approaches to try an better constrain the dynamic rupture processes of moderate-to-large intermediate-depth earthquakes (M~6.0) using one and at most two elliptical patches. The teleseismic point-source estimates of stress drop, seismic efficiency and more importantly rupture velocity provide key parameters that can help ous constrain both kinematic and dynamic source inversions. We believe this may be a step towards improving our understanding of the earthquake’s energy budgets using a combination of source inversions and source parameter estimates.