Fault Interactions in Extensional Regimes

Fault Interactions in Extensional Regimes M. Streepey and C. Lithgow-Bertelloni Department of Geological Sciences, University of Michigan, Ann Arbor, MI 48109 Studies have shown that faults generally tend to reactivate over long histories of deformation, often in spite of less favorable orientations or changing stress regimes in the region. Reactivation of faults suggests that rheology is a key determining factor in the localization of intense deformation in orogenic belts. It is evident in these studies that stresses are preferentially partitioned into pre-existing weak zones of the crust. This is shown commonly in orogenic belts, where thrust faults reactivate as normal faults during syn- to post-orogenic extension. Therefore, the interaction of faults might be an important element in the deformation of the lithosphere during pre- and post-orogenic tectonics. On shorter timescales, it has been suggested that fault interactions are commonplace in areas of active seismicity, and that those interactions can be related to earthquake triggering and therefore may be critically important in assessing the behavior of the lithosphere during deformation. We investigate this problem concentrating on the time evolution of faults in extensional regimes. Geologic evidence in ancient orogenic belts shows periods of protracted normal fault motion over timescales of hundreds of millions of years after orogenesis. This motion is likely episodic rather than continuous; however, this is not constrained by field and geochronological studies. Fault evolution on these timescales is modeled using the finite element code ABAQUS. Our elastic results show, as expected from dislocation theory, that stress shadows produced by motion along faults can be linearly superposed and that faults do not have a high degree of interaction. We have constructed new models of two-dimensional finite elements that represent a block of crust under extensional stresses. Sited in these blocks are weak zones that pre-exist simply as weaker elements in our block and therefore are sites of highest plastic strain. These new models also incorporate a time-dependent, elasto-viscous-plastic rheology, and stresses resulting from motion along faults in a block of crust, not surprisingly, do not appear to have simple, linear solutions. Our preliminary results show that there is some connection between geometry of faults and order of activation (faults do not all activate at the same time) and that plastic strain is generally completely confined to weaker areas, regardless of the rheology of the surrounding crustal material. These areas of plastic deformation do not propagate through stronger regions of the crust but remain isolated zones of weakness that reactivate over the history of deformation, even under different stress regimes. This has implications for the degree of interaction between faults and their role in stress transfer, which may be more complex than has been suggested.