The seismic cycle, also known as sequences of earthquakes and aseismic slip (SEAS), is characterized by periodic accumulation and release of tectonic stress. This multi-timescale process encompasses both rapid coseismic rupture, which unfolds over seconds, and prolonged interseismic deformation lasting up to thousands of years. Physics-based numerical models of the seismic cycle aim to capture SEAS within a unified framework and to enhance our understanding of earthquake generation. In this dissertation, I employ fully dynamic seismic cycle models, including dynamic inertial effects, to investigate earthquake nucleation and the coupled evolution of fault slip and inelastic fault zone deformation governed by damage rheology.

Chapter 2 focuses on exploring the influence of the characteristic weakening distance (DRS) in rate-and-state friction (RSF) on earthquake nucleation. The findings indicate that a larger value of a/b (>0.75), rather than the traditionally assumed 0.5, is needed to produce expanding crack nucleation for a relatively small DRS. This suggests that fixed-length nucleation may be more common on both natural and laboratory faults, and therefore earthquake nucleation style is strongly governed by both a/b and DRS.

Chapter 3 aims to develop a novel seismic cycle model integrating RSF and damage rheology to capture the coevolution of fault slip and fault zone deformation. Simulations reveal coseismic velocity drops consistent with seismological observations and a persistent shallow slip deficit (SSD). Off-fault damage is predominantly generated during earthquakes, concentrating at shallow depths in a flower-like structure, characterized by a distributed damage area surrounding a localized, highly damaged inner core. Utilizing an experimentally based logarithmic healing law, the model shows that coseismic reductions in off-fault rigidity only partially heal, leading to a cumulative, permanent rigidity loss over multiple seismic cycles. Consequently, the fault zone width and rigidity eventually stabilize, reaching a mature state with a large cumulative fault slip.

In Chapter 4, I apply this earthquake coevolution model to examine the role of weak fault zone deformation on the generation of multiscale seismicity. The results demonstrate that relatively weak fault zones (i.e., when surrounding rocks have a low internal friction coefficient) facilitate the production of both large and small earthquakes, reproducing key earthquake scaling relations observed in nature, such as power-law magnitude-frequency distribution, magnitude-invariant static stress drop, and non-linear fracture energy scaling in a unified framework. These findings highlight the fundamental role of the coevolution of earthquakes and fault zone inelastic deformation in controlling earthquake behaviors.

In conclusion, these findings highlight the significant role that fault friction and fault zone deformation play in governing earthquake nucleation, slip behavior, and earthquake scaling relations. Particularly, by capturing the coevolution of fault slip and fault zone deformation over multiple seismic cycles, Chapters 3 and 4 underscore the fundamental importance of fault zone inelastic deformation in shaping earthquake dynamics beyond fault friction and rock elasticity. The spontaneously generated inelastic deformation dissipates elastic strain energy and acts as natural rupture barriers to modulate earthquake size. The coevolution of fault slip and fault zone inelastic deformation provides an efficient way to generate fault stress heterogeneity and a wide spectrum of earthquake size over multiple seismic cycles. These insights offer new directions for interpreting natural fault systems and assessing seismic hazards.