Fault Weakening and Shear Localization during Crater Collapse

Introduction: The observed morphologies of impact basins can only be explained by invoking dramatic weakening of the crater walls and floor during crater collapse, but the mechanism underlying the drop in strength is unknown. Here, we explore the possibility that strain-weakening along faults is the primary mechanism. We present a model of temperatureand pressure-dependant strain-weakening along faults during crater collapse. Background: The depths of complex craters are far shallower than predicted using the quasi-static frictional strength of fractured rock [1, 2]. Therefore, some physical process must be causing transient weakening of the rock. The mechanism responsible for this dramatic drop in strength remains unknown. Proposed explanations for weakening include acoustic fluidization [3], or dynamic weakening along faults, possibly caused by melt lubrication [4]. Acoustic fluidization has been modeled extensively [e.g. 1, 5, 6, 7], however, little work has been done on fault weakening [8]. Simulations of crater collapse including a simple model of dynamic strain weakening were performed by Senft and Stewart [8]. Their work was successful at producing the characteristic features of large craters (fig.1), indicating that fault weakening could explain the strength drop during crater collapse. However, the simplicity of their model presents some limitations. One of the limitations is that the criterion for weakening is not pressure-dependent. This asumption is not a problem in large craters because the pressures are high everywhere, but in smaller complex craters, strain weakening must account for pressure effects because the weakening criterion is not met everywhere. Another limitation is that detailed work on finding the average spacing between faults cannot be done reliably with such a simple model. Also, the thickness and other characteristics of the faults are not constrained. Therefore, a more sophisticated theory is necessary to apply a fault-weakening model to any size complex crater and to predict specific propeties of the faults, such as their spacing and thickness. With the benefits of an improved model as motivation, the present work aims to improve upon the model by Senft and Stewart [8] by including temperature, pressure, strain rate, and localization effects in the simulations of fault-driven crater collapse. Figure 1: A simulation of an 8 km-diameter body impacting a rocky planet 140 seconds after impact by Senft and Stewart [8]. The color denotes total accumulated shear strain. Quasilinear zones of localized shear strain are interpreted as fault zones. The numbers indicate the relative order of fault activation, with 1 being the earliest. Faults naturally arise in this model as rocks are damaged, and the final crater shape predicted using fault weakening is consistent with observations [8].