EVOLVING EFFICIENCY OF SEGMENTED STRIKE-SLIP FAULT SYSTEMS

Strike-slip faults link over multiple earthquake cycles to form through-going structures, as demonstrated by the more continuous nature of the mature San Andreas Fault in California, USA, relative to the younger, segmented San Jacinto Fault nearby. However, despite its relative immaturity, the San Jacinto Fault supports between one third and one half of the relative motion of the Pacific and North American plates. This suggests that its segmented structure efficiently accommodates applied plate motion. We ask: what spacing and configuration of segmented faults allows them to accommodate applied plate displacement as efficiently as a single, non-planar fault resulting from hard-linkage? Using boundary element method models, we calculate the efficiency of hard-linked and segmented systems of two 20 km, right-lateral, strike-slip faults. Faults range from 1 km to 8 km from one another and vary between right-stepping and left-stepping, underlapping and overlapping. Assuming that higher efficiency correlates with a higher potential for a single, large earthquake to rupture both fault segments, this approach has utility for assessing the seismic hazard associated with segmented fault systems. On the meter scale, en echelon fractures propagate by both tensile and shear failure. Understanding the propensity for opening and hard-linkage between fractures in these systems is important for characterizing the potential for fluid flow through the system. The algorithm GROW (GRowth by Optimization of Work) honors tensile and frictional failure criteria along pre-existing surfaces and at fracture tips, while allowing the orientation of fracture growth within the system to be governed by the global energy budget. In GROW, fractures evolve in order to minimize the energy expended during deformation, thereby maximizing the mechanical efficiency of the entire system. Using GROW, we propagate systems of three, en echelon, right-lateral shear fractures that are initially right- or left-stepping and over- or underlapping. Tracking the efficiency of these fault systems as they evolve provides further quantitative insight into the efficiency gains provided by hard- versus soft-linkage.