Abstract -
Faults typically form components of complex systems in which dynamic and kinematic interactions are an essential element of fault growth on earthquake and geological timescales (Walsh and Watterson, 1991). These interactions, which on geological timescales are typically manifest as displacement transfer associated with fault intersection (hard-linkage) or distributed deformation (soft-linkage), influence fault dimensions, displacement patterns and growth. Fault interaction and growth over timescales of up to millions of years have been constrained by analysis of faulting with displacements of millimetres to kilometres using outcrop and seismic-reflection data. Many of the faults analysed experienced rapid fault propagation followed by prolonged displacement accumulation. The resulting near-constant lengths established from an early stage in the fault history are attributed to retardation of lateral propagation by interaction between fault tips. These interactions are interpreted to result in near-linear along strike displacement profiles and increasing displacement gradients with fault size (i.e. length and maximum displacement). The increase in displacement gradients is associated with a steepening of slope on fault displacement vs length (D-L) plots and with rising strain across fault systems. For immature low-strain fault systems in which individual faults have experienced few slip events the slope of D-L plots is typically <1, while in mature higher-strain systems this slope is generally >1. The increasing D-L slopes is interpreted to primarily reflect a combination of tip-line retardation arising from interactions and preferential death (or longer recurrence intervals between individual slip increments) of smaller faults. As a consequence, large faults accrue their cumulative displacement in many more slip events than small faults. The available data are consistent with fault growth models in which the hierarchy of fault size is established rapidly, perhaps in many cases due to fault reactivation, with the longest faults moving fastest, for the longest duration and being most likely to benefit from strain localisation. Our fault growth models draw heavily on the ideas and observations of Juan Watterson, many of which are just as relevant today as when they were first developed over 20 years ago.
Abstract of talk given to:
Geometry and Growth of Normal Faults, Geological Society of London, June 2014.