Which Dynamic Rupture Parameters Can Be Estimated from Strong Ground Motion and Geodetic Data?

We have tested to which extent commonly used dynamic rupture parameters can be resolved for a realistic earthquake scenario from all available observations. For this purpose we have generated three dynamic models of the Landers earthquake using a single, vertical, planar fault with heterogeneity in either the initial stress, the yield stress, or the slip-weakening distance. Although the dynamic parameters for these models are inherently different, all the simulations are in agreement with strong motion, GPS, InSAR, and field data for the event. The rupture propagation and slip distributions obtained for each model are similar, showing that the solution of the dynamic problem is non-unique. In other words, it is not always possible to separate strength drop and the slip-weakening distance using rupture modeling, in agreement with the conclusions by GUATTERI and SPUDICH (2000).