Fault growth Models: observations from historical earthquakes and active faults in New Zealand

Faults in the brittle upper crust are thought to primarily grow due to repeated earthquakes. To understand better fault growth during incremental slip events we analyse geometric and displacement data for timescales of individual earthquakes (since 1840 AD) to millions of years on New Zealand active faults. The active faults studied are from connected networks with a range of orientations, lengths (1–200 km), displacement rates (0.1–27 mm/yr) and slip types. Our data indicate that individual earthquakes produce slip on multiple faults with variable sizes, orientations and slip types; in some cases these earthquakes cross tectonic domain boundaries. Earthquakes generally produce partial rupture of reactivated bedrock faults and show little evidence of tip propagation, characteristics most closely resembling the constant-length fault growth model, with growth primarily achieved by increases in cumulative slip. Earthquake slip profiles display a range of shapes with one or more maxima. High gradients along faults and approaching fault tips reflect coseismic slip transfer to nearby faults. These high slip gradients are consistent with stress interactions and kinematic coherence between faults during individual earthquakes (i.e., timescales of seconds to minutes). Coseismic increments of slip increase with rupture length and are described by a power function of ~0.5, while the power function for cumulative displacement and final length is ≥1. An important consequence of these divergent power functions is that larger faults broadly accrue their finite displacements in more earthquakes than smaller faults. The increase in earthquake number with fault size is achieved by a combination of shorter recurrence intervals and longer faulting histories for larger faults. We believe that our New Zealand observations have global application.